WO2021068746A1 - 隧道巡检图像采集装置、隧道巡检系统及隧道巡检方法 - Google Patents

隧道巡检图像采集装置、隧道巡检系统及隧道巡检方法 Download PDF

Info

Publication number
WO2021068746A1
WO2021068746A1 PCT/CN2020/117088 CN2020117088W WO2021068746A1 WO 2021068746 A1 WO2021068746 A1 WO 2021068746A1 CN 2020117088 W CN2020117088 W CN 2020117088W WO 2021068746 A1 WO2021068746 A1 WO 2021068746A1
Authority
WO
WIPO (PCT)
Prior art keywords
tunnel
image
ccd cameras
auxiliary light
light source
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2020/117088
Other languages
English (en)
French (fr)
Inventor
余天乐
匡俊
姚继东
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai Oriental Maritime Engineering Technology Co Ltd
Original Assignee
Shanghai Oriental Maritime Engineering Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from CN201921669806.4U external-priority patent/CN210720194U/zh
Priority claimed from CN201910949526.7A external-priority patent/CN110593957B/zh
Priority claimed from CN201910949507.4A external-priority patent/CN110487816A/zh
Application filed by Shanghai Oriental Maritime Engineering Technology Co Ltd filed Critical Shanghai Oriental Maritime Engineering Technology Co Ltd
Priority to US17/767,437 priority Critical patent/US12045966B2/en
Priority to EP20873699.1A priority patent/EP4043693B1/en
Publication of WO2021068746A1 publication Critical patent/WO2021068746A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • H04N7/188—Capturing isolated or intermittent images triggered by the occurrence of a predetermined event, e.g. an object reaching a predetermined position
    • E—FIXED CONSTRUCTIONS
    • E21—EARTH OR ROCK DRILLING; MINING
    • E21F—SAFETY DEVICES, TRANSPORT, FILLING-UP, RESCUE, VENTILATION, OR DRAINING IN OR OF MINES OR TUNNELS
    • E21F17/00—Methods or devices for use in mines or tunnels, not covered elsewhere
    • E21F17/18—Special adaptations of signalling or alarm devices
    • G—PHYSICS
    • G03—PHOTOGRAPHY; CINEMATOGRAPHY; ANALOGOUS TECHNIQUES USING WAVES OTHER THAN OPTICAL WAVES; ELECTROGRAPHY; HOLOGRAPHY
    • G03B—APPARATUS OR ARRANGEMENTS FOR TAKING PHOTOGRAPHS OR FOR PROJECTING OR VIEWING THEM; APPARATUS OR ARRANGEMENTS EMPLOYING ANALOGOUS TECHNIQUES USING WAVES OTHER THAN OPTICAL WAVES; ACCESSORIES THEREFOR
    • G03B15/00—Special procedures for taking photographs; Apparatus therefor
    • G03B15/02—Illuminating scene
    • G03B15/03—Combinations of cameras with lighting apparatus; Flash units
    • G—PHYSICS
    • G03—PHOTOGRAPHY; CINEMATOGRAPHY; ANALOGOUS TECHNIQUES USING WAVES OTHER THAN OPTICAL WAVES; ELECTROGRAPHY; HOLOGRAPHY
    • G03B—APPARATUS OR ARRANGEMENTS FOR TAKING PHOTOGRAPHS OR FOR PROJECTING OR VIEWING THEM; APPARATUS OR ARRANGEMENTS EMPLOYING ANALOGOUS TECHNIQUES USING WAVES OTHER THAN OPTICAL WAVES; ACCESSORIES THEREFOR
    • G03B37/00—Panoramic or wide-screen photography; Photographing extended surfaces, e.g. for surveying; Photographing internal surfaces, e.g. of pipe
    • G03B37/005—Photographing internal surfaces, e.g. of pipe
    • G—PHYSICS
    • G03—PHOTOGRAPHY; CINEMATOGRAPHY; ANALOGOUS TECHNIQUES USING WAVES OTHER THAN OPTICAL WAVES; ELECTROGRAPHY; HOLOGRAPHY
    • G03B—APPARATUS OR ARRANGEMENTS FOR TAKING PHOTOGRAPHS OR FOR PROJECTING OR VIEWING THEM; APPARATUS OR ARRANGEMENTS EMPLOYING ANALOGOUS TECHNIQUES USING WAVES OTHER THAN OPTICAL WAVES; ACCESSORIES THEREFOR
    • G03B37/00—Panoramic or wide-screen photography; Photographing extended surfaces, e.g. for surveying; Photographing internal surfaces, e.g. of pipe
    • G03B37/04—Panoramic or wide-screen photography; Photographing extended surfaces, e.g. for surveying; Photographing internal surfaces, e.g. of pipe with cameras or projectors providing touching or overlapping fields of view
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/20—Image enhancement or restoration using local operators
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes
    • G06V20/176—Urban or other man-made structures
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/70—Labelling scene content, e.g. deriving syntactic or semantic representations
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/56—Cameras or camera modules comprising electronic image sensors; Control thereof provided with illuminating means
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/70—Circuitry for compensating brightness variation in the scene
    • H04N23/74—Circuitry for compensating brightness variation in the scene by influencing the scene brightness using illuminating means
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/90—Arrangement of cameras or camera modules, e.g. multiple cameras in TV studios or sports stadiums
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N5/00—Details of television systems
    • H04N5/04—Synchronising
    • H04N5/06—Generation of synchronising signals
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20112—Image segmentation details
    • G06T2207/20132—Image cropping
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30181—Earth observation
    • G06T2207/30184—Infrastructure

Definitions

  • the invention belongs to the technical field of tunnel flaw detection, and particularly relates to a tunnel inspection image acquisition device, a tunnel inspection system and a tunnel inspection method.
  • the present invention provides a tunnel inspection image acquisition device, a tunnel inspection system and a tunnel inspection method, which have simple structure, large detection range, high timeliness, fast detection speed, high recognition accuracy, and inspection Technical features of accurate positioning and clear images.
  • the invention provides a tunnel inspection image acquisition device, which is applied to the image acquisition of the inner surface of the tunnel, and includes: a plurality of CCD cameras, a plurality of auxiliary light sources used for shooting and exposure of the CCD camera, a fixed bracket, a CCD camera and an auxiliary light source Fixed on a fixed bracket; among them, the fixed positions of multiple CCD cameras are located on the same circular ring. In working state, the center of the circular ring is located at the center of the tunnel to ensure that the distances between the multiple CCD cameras and the inner surface of the tunnel are consistent. And the overall shooting angle of multiple CCD cameras covers the range of the tunnel inner surface.
  • the CCD camera and the auxiliary light source are sequentially spaced and fixed on the fixing bracket.
  • the shooting angles between adjacent CCD cameras partially overlap.
  • the fixed bracket includes a base frame, a fixed block, and a plurality of support rods.
  • the base frame is fixedly connected to the fixed block, one end of the support rod is fixedly connected to the fixed block, and the other end of the support rod is connected to the CCD camera or auxiliary light source. Fixed connection.
  • the auxiliary light source is an LED area array light source.
  • the present invention also provides a tunnel inspection system, which includes a mobile testing platform for rail travel, and also includes a computer, a display, a synchronous encoder, a data collector, and the tunnel inspection platform provided on the mobile platform. Inspection image acquisition device; among them, the fixed bracket is set on the mobile platform for flaw detection.
  • the center of the ring where the multiple CCD cameras are located is at the center of the tunnel to ensure that the distance between the multiple CCD cameras and the tunnel lining is consistent;
  • the synchronous encoder is electrically connected with the CCD camera and the auxiliary light source, and is used to send synchronous pulse signals at fixed distances according to the displacement of the mobile platform to trigger multiple CCD cameras and multiple auxiliary light sources to work synchronously;
  • the data collectors are connected to the multiple A CCD camera and a synchronous encoder are electrically connected to collect the tunnel lining image taken by the CCD camera, and the tunnel lining image is numbered in combination with the synchronous encoder;
  • the computer is electrically connected to the data collector and the display to monitor the tunnel
  • the lining image performs image recognition and disease point labeling. At the same time, the tunnel lining image after the disease point labeling is stitched and displayed in real time.
  • it further includes a positioning device.
  • the positioning device is electrically connected to the computer, and the computer is electrically connected to the auxiliary light source.
  • the positioning device is used to read and write the RFID tag in the tunnel, wherein the location in the RFID tag is read.
  • the information is fed back to the computer for mileage correction of the flaw detection mobile platform; the tunnel structure information in the RFID tag is read, and the information is fed back to the computer to adjust the brightness and exposure parameters of the auxiliary light source to change the image depth effect to obtain a clear image;
  • the feedback information of the disease point is written into the RFID tag to mark the position of the disease point.
  • the computer, the display, the synchronous encoder, the data collector, the tunnel inspection image acquisition device, and the positioning device are respectively detachably arranged on the mobile platform for flaw detection.
  • the present invention also provides a tunnel inspection method, which includes the following steps:
  • S4 Perform image recognition, mark disease points and online alarm on the numbered tunnel lining image. At the same time, stitch the tunnel lining image after the disease point mark and display it in real time.
  • steps S1 to S4 during the execution of steps S1 to S4, the following steps are further included: reading the RFID tag in the tunnel, and adjusting the brightness and exposure parameters of the auxiliary light source according to the tunnel structure information of the RFID tag to change the image depth The effect is to obtain a clear image.
  • step S4 further includes the following step: if the diseased point is detected by the image recognition, the diseased point information is written into the RFID tag in the tunnel.
  • step S2 specifically includes the following steps:
  • S22 Integrate and process the pulse signal to obtain the synchronous pulse period signal, and determine the current traveling direction of the flaw detection mobile platform and the number of pulses to obtain the mileage of the flaw detection mobile platform.
  • the synchronous pulse period signal is multi-frequency output to trigger the CCD
  • the camera and auxiliary light source work synchronously to capture images of the tunnel lining.
  • steps S1 to S4 during the execution of steps S1 to S4, the following steps are further included: reading the RFID tag in the tunnel, and performing mileage correction on the flaw detection mobile platform according to the positioning information in the RFID tag.
  • step S4 image recognition specifically includes the following steps:
  • A1 Receive and store the tunnel lining image
  • A2 Recognize and crop the repeated area of the tunnel lining image
  • A3 Perform image filtering on the trimmed tunnel lining image, and smooth the impurity factors in the tunnel lining image obtained by the image filtering;
  • A4 The overall suppression of the pixel gray value of the impurity factor, the preliminary identification of the tunnel lining image after the overall suppression, and the overall enhancement of the pixel gray value of the suspected disease point;
  • A5 Compare the similarity between the suspected disease point and the theoretical model, judge whether the suspected disease point is a disease point, and feed back the disease point information.
  • the present invention has the following advantages and positive effects:
  • the tunnel inspection image acquisition device of the present invention multiple CCD cameras are provided, and the fixed positions of the multiple CCD cameras are located on the same circular ring. In the working state, the center of the circular ring is located at the center of the tunnel. This arrangement can be Ensure that the distance between the CCD industrial camera lens and the inner surface of the tunnel is the same. There is no need to adjust the focal length and other parameters during the shooting process. The captured image is not easy to be deformed. The actual size of the image captured by each lens is the same, which greatly reduces the difficulty of image processing. It is convenient for real-time image processing, and improves the timeliness of image processing.
  • the detection range can be greatly increased, and the maximum can reach 360°, which is convenient to cover the range of the tunnel lining surface, and can complete the full-line tunnel section ( (Such as round, rectangular, horseshoe, etc.) Clear photographs at different viewing distances have achieved the technical effects of easy image processing and large detection range;
  • the CCD camera and the auxiliary light source are sequentially spaced apart on the fixed bracket, and each auxiliary light source can be used for exposure of two CCD cameras, which improves the utilization rate of the auxiliary light source and reduces the cost , At the same time, under the same shooting requirements, the exposure difference between CCD cameras is reduced, the consistency between images is further improved, and the timeliness of image processing is improved;
  • the shooting angles between adjacent CCD cameras are partially overlapped, which not only reduces the possibility of missed detection areas, but also uses the overlapped parts of the images to reduce the risk of image stitching. Difficulty, which facilitates real-time image processing and improves the timeliness of image processing;
  • the fixed support structure is simple and lightweight, easy to transport and store, and reduces the cost.
  • the LED area array light source can be used to form a light band with uniform brightness on the tunnel wall. Under the shooting requirements, the exposure difference between CCD camera shooting is reduced, the consistency between images is further improved, and the timeliness of image processing is improved;
  • the tunnel inspection image acquisition device of the present invention is set on a mobile platform for flaw detection, and cooperates with a computer, a monitor, a synchronous encoder, and a data acquisition device to perform synchronous shooting, disease identification, real-time splicing and display of the tunnel, among which, based on
  • the easy processing of the images collected by the tunnel inspection image acquisition device improves the recognition speed and splicing speed of the tunnel inspection system, so as to realize the real-time display on the spot.
  • the user can perform the on-site processing of the inspection through the real-time displayed content or notify in time Relevant departments deal with it, which greatly improves the timeliness of tunnel inspection, and achieves the technical effects of high efficiency, short inspection time, high accuracy, and less likely to miss inspection;
  • a detachable modular design is adopted, which is convenient for independent storage and transportation, and improves the convenience of the system;
  • the center of the ring where the multiple CCD cameras on the mobile platform for flaw detection are located is adjusted to the center of the tunnel to ensure the distance between the multiple CCD cameras and the inner surface of the tunnel In this way, there is no need to adjust the focal length and other parameters in the shooting process, and the captured image is not easy to be deformed.
  • the actual size of the image captured by each lens is the same, which greatly reduces the difficulty of image processing, facilitates the real-time processing of the image, and improves the tunnel inspection
  • the recognition speed and splicing speed of the system can realize real-time display on site. Users can perform on-site inspection of inspections through real-time display content or notify relevant departments for processing in time, which greatly improves the timeliness of tunnel inspections and achieves high efficiency. , The technical effect of short detection time, high accuracy, and not easy to miss detection;
  • the operations of reading and writing RFID tags can mark disease points, adjust the auxiliary light source, and calibrate the mileage of the flaw detection mobile platform, which facilitates later re-inspection and automatically changes the light source
  • Brightness and exposure parameters can change the image depth of field to produce clear images.
  • mileage correction also improves the accuracy of disease point positioning, achieving the technical effects of disease point traceability, accurate inspection and positioning, and clear images;
  • image cropping is used to reduce the recognition area, image filtering to reduce impurity factors that interfere with recognition, image enhancement to suppress impurity factors, to identify and highlight suspected disease points, and to confirm whether they are disease points through similarity comparison, The technical effect of fast recognition speed and high recognition accuracy is achieved.
  • Figure 1 is a three-dimensional structure diagram of a tunnel inspection image acquisition device of the present invention
  • FIG. 2 is a plan structure diagram of a tunnel inspection image acquisition device of the present invention.
  • Figure 3 is a schematic diagram of the overall structure of a tunnel inspection system of the present invention.
  • FIG. 4 is a schematic diagram of the installation position of a tunnel inspection system of the present invention.
  • Figure 5 is a block diagram of the architecture of a tunnel inspection system of the present invention.
  • Figure 6 is a flow chart of synchronous triggering of a tunnel inspection system of the present invention.
  • Fig. 7 is a disease identification flowchart of a tunnel inspection system of the present invention.
  • Fig. 8 is an overall flow chart of a tunnel detection method of the present invention.
  • 1-Tunnel inspection image acquisition device 11-CCD camera; 12-auxiliary light source; 13-fixed bracket; 131-support rod; 132-fixed block; 133-underframe; 2-flaw detection mobile platform; 31-computer; 32 -Display; 33-Synchronous encoder; 34-Data collector; 35-Positioning device.
  • the present application provides a tunnel inspection image acquisition device, which is applied to the image acquisition of the inner surface of the tunnel, and includes: a plurality of CCD cameras 11, a plurality of auxiliary light sources 12 for shooting and exposure of the CCD camera 11, and a fixed
  • the bracket 13, the CCD camera 11 and the auxiliary light source 12 are fixed on the fixed bracket 13; wherein the fixed positions of the multiple CCD cameras 11 are located on the same ring.
  • the center of the ring is located at the center of the tunnel to ensure more The distance between the two CCD cameras 11 and the inner surface of the tunnel is the same, and the overall shooting angle of the plurality of CCD cameras 11 covers the range of the inner surface of the tunnel.
  • the tunnel inspection image acquisition device of this embodiment can be installed on any platform, and connected and installed through the fixed bracket 13 to perform image acquisition of the tunnel inspection.
  • this embodiment uses 6 CCD cameras 11 and 7 auxiliary light sources 12, the auxiliary light source 12 is a stroboscopic area array light source, distributed on the same ring through the fixed bracket 13, the advantage of this layout is The auxiliary light source 12 and the CCD camera 11 are closely arranged on the same support, and the overall structure is simple. Among them, each auxiliary light source 12 of this embodiment is separated from the CCD camera 11 by 22.5°, and there are 6 CCD camera 11 channels. Each CCD camera 11 has a circular shooting angle of 48°, so as to fully cover the 270° tunnel (with partial overlap) Area), the auxiliary light source 12 adopts an LED area array light source, arranged at an interval of 45°, and can form a light band with uniform brightness on the wall of the tunnel.
  • the above-mentioned angle setting, quantity setting, and equipment selection are only a specific technical solution of this application.
  • the above design can be adjusted according to the structure of the tunnel, the camera's circular shooting angle, etc., and the maximum 360° shooting coverage can be achieved.
  • the overall shooting angle of 270° is best selected in this application to achieve clear photos of different viewing distances of tunnel sections (such as circles, rectangles, horseshoes, etc.) across the entire line.
  • multiple CCD cameras 11 are set, and the fixed positions of the multiple CCD cameras 11 are located on the same circular ring.
  • the center of the circular ring is located at the center of the tunnel.
  • the actual size of the image taken by each lens is the same, which greatly reduces the difficulty of image processing, facilitates the real-time processing of the image, and improves the image Processing timeliness, and at the same time, due to the use of the same ring setting method, the detection range can be greatly increased, and the maximum can reach 360°, which is convenient to cover the range of the tunnel lining surface, and can complete clear photographs of different viewing distances of the full-line tunnel section.
  • the fixing bracket 13 of this embodiment includes a base frame 133, a fixing block 132, and a plurality of supporting rods 131.
  • the base frame 133 is fixedly connected to the fixing block 132, and one end of the supporting rod 131 is fixedly connected to the fixing block 132.
  • the other end of the support rod 131 is fixedly connected with the CCD camera 11 or the auxiliary light source 12.
  • the fixed bracket 13 may also be a ring-shaped supporting plate, and the CCD camera 11 and the auxiliary light source 12 are mounted on the supporting plate, and the auxiliary bracket is used for integral fixed support.
  • the fixed bracket 13 of this embodiment has a simple and lightweight structure, is easy to transport and store, and reduces costs. At the same time, using an LED array light source, it can form a light band with uniform brightness on the tunnel wall, reducing the CCD camera under the same shooting requirements. 11 The exposure difference between shots further improves the consistency between images and improves the timeliness of image processing.
  • each auxiliary light source 12 can be used for the exposure of two CCD cameras 11, which improves the utilization rate of the auxiliary light source 12 and reduces the cost.
  • the shooting time of the CCD camera 11 is reduced.
  • the exposure difference improves the consistency between images, thereby improving the timeliness of image processing.
  • the use of LED area array light source can form a uniform brightness light band on the tunnel wall, which is also to reduce the CCD camera 11 shooting.
  • the exposure difference further improves the consistency between images, thereby improving the timeliness of image processing.
  • the shooting angles between adjacent CCD cameras 11 in this embodiment partially overlap.
  • the shooting angles of the adjacent CCD cameras 11 of the present invention partially overlap, which not only reduces the possibility of missed detection areas, but also uses the overlapped parts of the images to reduce the difficulty of image stitching, facilitate the real-time processing of the images, and improve the image Timeliness of treatment.
  • the present application provides a tunnel inspection system based on Embodiment 1, which includes a mobile platform for flaw detection 2 used for rail travel, and also includes a computer 31, a display 32, and a synchronization code set on the mobile platform 2 for flaw detection.
  • the fixed bracket 13 is set on the mobile platform 2 for flaw detection.
  • the encoder 33 is electrically connected to the CCD camera 11 and the auxiliary light source 12 respectively, and is used to send synchronization pulse signals at regular intervals according to the displacement of the flaw detection mobile platform 2 to trigger multiple CCD cameras 11 and multiple auxiliary light sources 12 to work synchronously;
  • the collector 34 is respectively electrically connected with a plurality of CCD cameras 11 and a synchronous encoder 33 to collect the tunnel lining image taken by the CCD camera 11, and combined with the synchronous encoder 33 to number the tunnel lining image;
  • the collector 34 and the display 32 are electrically connected for image recognition and disease point marking of the tunnel lining image. At the same time, the tunnel lining image after the disease point marking is spliced and displayed in real time.
  • the flaw detection mobile platform 2 of this embodiment is a double-track flaw detection vehicle, which can move on steel rails.
  • the tunnel inspection image acquisition device 1 is fixedly installed at the rear of the vehicle body and protrudes from the vehicle body.
  • the front of the dual-track flaw detection vehicle is equipped with a computer 31 and a display 32 to control the entire dual-track flaw detection vehicle and the tunnel inspection system.
  • the user can adjust the relevant parameters of the tunnel inspection system in real time on site, and can view the tunnel inspection in real time on site
  • the photos taken by the system can be used to confirm whether there are any disease points in the tunnel on the spot, and if there are any defects, the relevant departments can be notified in time for maintenance and other operations.
  • the synchronous encoder 33 of this embodiment can be installed on the moving bearing of the flaw detection mobile platform 2, and generates a pulse signal with direction and displacement according to the rotation of the bearing, so as to trigger the CCD camera 11 and the light source synchronously.
  • the distance synchronization triggers a shooting mode.
  • A+, B+, A-, B- signals are generated according to the movement of the vehicle body, and the generated signals are transmitted to the signal processor, and the signal processor pairs the synchronous encoder
  • the pulse signal generated by 33 is integrated and processed to determine the walking direction and the number of pulses of the current dual-track flaw detection vehicle equipped with the intelligent tunnel inspection system, and the corresponding code value is judged and uploaded to the computer 31 for image storage, and passed the appropriate multi-point Frequency transmission, trigger the CCD area array camera and the light source controller to work synchronously, the CCD area array camera and the light source controller receive the corresponding pulse signal, the CCD area array camera triggers the work to collect the tunnel lining image data, the light source controller receives the pulse signal After that, the switching frequency of the light source is controlled to synchronize the shooting frequency of the CCD area camera to ensure that the tunnel is bright enough when shooting by the CCD area camera.
  • the data collector 34 in this embodiment may be a switch, and the switch may be a multi-channel switch to one switch, which can realize data collection and simplify the communication connection line between the camera and the computer 31.
  • the computer 31 After the computer 31 receives the image of the tunnel lining transmitted by the data collector 34, the computer 31 performs the image identification of the diseased point. Among them, a large number of on-site internal images of the tunnel are collected in the early stage, and a large number of statistics are analyzed for pedestrian platforms, electrical cables, pipes, and tunnel segments The regional gray-scale difference formed by the normal image and the problem image content such as seams, fire channel indicator lights, electrical boxes, fire emergency telephones, train platforms, fire pipes, stairs, etc., prepares the basic technology for various intelligent identification of tunnel inspections.
  • First step The system continuously collects the 270° tunnel lining image of the entire tunnel through the cooperation of the stroboscopic area array LED light source and the area array CCD industrial camera, and saves it.
  • Step 2 Filter some randomly distributed impurities such as tunnel segments and platforms through image filtering to make the surface of tunnel segments and platforms smoother, which helps to improve the efficiency of tunnel intelligent algorithms;
  • the third step Image enhancement is to suppress the overall pixel gray value of the surface image of ordinary tunnel segments, platforms, etc., and to enhance the overall enhancement of special points such as cracks and water leakage, which is conducive to the rapid screening of the tunnel disease recognition algorithm Suspected disease point;
  • Step 4 Compare the similarity between the suspected damage point and the theoretical model. If the similarity conforms to the theoretical model, an abnormal alarm will be issued and a problem point report will be generated.
  • the positioning device 35 is electrically connected to the computer 31, and the computer 31 is electrically connected to the auxiliary light source 12, and the positioning device 35 is used for reading and writing RFID tags in the tunnel, wherein, Read the positioning information in the RFID tag and feed it back to the computer 31 for mileage correction of the flaw detection mobile platform 2; read the tunnel structure information in the RFID tag and feed it back to the computer 31 to adjust the brightness and exposure parameters of the auxiliary light source 12 to Change the image depth effect to obtain a clear image; write the disease point information fed back by the computer 31 into the RFID tag to mark the position of the disease point.
  • the positioning device 35 is equipped with a vehicle-mounted reader/writer, and the tag information is read in time through the positioning system, and the tag information is transmitted to the computer 31 by communication methods such as Ethernet, Bluetooth, Zigbee, WLAN, or RS232, RS485, and the station is generated in the background.
  • the tag is a full kilometer mileage tag
  • the system will automatically trigger mileage correction to reduce the frequency of mileage correction.
  • the computer 31 will automatically mark it, that is, control the on-board reading and writing The device writes disease information in the RFID tag for later review.
  • the vehicle-mounted reader/writer is based on RFID technology.
  • the tunnel inspection system of this embodiment reads or writes RFID tags through the positioning device 35, and can mark the diseased points and calibrate the mileage of the flaw detection mobile platform 2, which facilitates the later re-inspection, and the mileage correction also improves the diseased points.
  • the accuracy of positioning has achieved the technical effect of accurate positioning for inspection.
  • the computer 31, the display 32, the synchronous encoder 33, the data collector 34, the tunnel inspection image acquisition device 1, and the positioning device 35 are respectively detachably arranged on the mobile platform 2 for flaw detection.
  • the tunnel inspection system of this embodiment adopts a detachable modular design, which is convenient for independent storage and transportation, and improves the convenience of the system.
  • the overall workflow of the tunnel inspection system is as follows:
  • S1 The vehicle enters the tunnel area and starts to detect
  • the synchronous encoder 33 automatically triggers the synchronous pulse period signal as the dual-track flaw detection vehicle advances, and sends the signal to the CCD industrial camera and the light source controller at the same time to make it work synchronously, and complete an image acquisition after traveling a fixed distance;
  • the data collector 34 collects the images in real time, and combines with the synchronous encoder 33 to number the corresponding images and upload them to the computer through various communication methods: Ethernet, Bluetooth, Zigbee, WLAN or RS232, RS485 and other communication methods 31;
  • S5 The intelligent recognition system in the computer 31 realizes leakage recognition through partial image gray-scale difference features, and realizes automatic recognition functions such as pipeline shedding through linear gray-scale difference features. For the identified disease points, the corresponding information is written in the RFID tag through the RFID positioning device 35, which is convenient for later review;
  • the computer 31 calls the tunnel images that have been identified and marked with disease points, and uses algorithms to stitch the images acquired by the six CCD cameras 11 at the same time to more intuitively display the complete tunnel lining image.
  • the tunnel inspection image acquisition device 1 of this embodiment is set on the mobile platform 2 for flaw detection, and cooperates with the computer 31, the display 32, the synchronous encoder 33, and the data collector 34 to perform tunnel synchronous shooting, disease identification, real-time splicing, and Display, among which, the easy processing of the images collected by the tunnel inspection image acquisition device 1 based on the embodiment 1 improves the recognition speed and splicing speed of the tunnel inspection system, so as to realize the real-time display on the spot, and the user can use the real-time display
  • the content is processed on-site for inspection or timely notification to relevant departments for processing, which greatly improves the timeliness of tunnel inspection and achieves the technical effects of high efficiency, short inspection time, high accuracy, and less likely to miss inspection.
  • the present application provides a tunnel inspection method, which includes the following steps:
  • S4 Perform image recognition, mark disease points and online alarm on the numbered tunnel lining image. At the same time, stitch the tunnel lining image after the disease point mark and display it in real time.
  • the flaw detection mobile platform 2 of this embodiment can be a dual-track flaw detection vehicle that can move on steel rails.
  • a tunnel inspection system is installed on the dual-track flaw detection vehicle to perform the tunnel inspection of this embodiment. Inspection method, the tunnel inspection system includes: computer 31, display 32, synchronous encoder 33, data collector 34, positioning device 35, tunnel inspection image acquisition device 1, tunnel inspection image acquisition device 1 including CCD camera 11, auxiliary The light source 12 and the fixing bracket 13, and the fixing bracket includes a supporting rod 131, a fixing block 132 and a bottom frame 133.
  • the executable body of the tunnel inspection method of this embodiment includes, but is not limited to, the above-mentioned dual-track flaw detection vehicle.
  • step S1 of this embodiment is: the mobile testing platform enters the tunnel area, and the center of the ring where the multiple CCD cameras on the mobile testing platform are located is adjusted to the center of the tunnel to ensure that the multiple CCD cameras are in the tunnel.
  • the distance of the lining is the same, and then the tunnel inspection is started.
  • the center of the ring where the multiple CCD cameras on the mobile testing platform are located is adjusted to the center of the tunnel to ensure that the distances between the multiple CCD cameras and the inner surface of the tunnel are consistent.
  • the actual size of the images taken by each lens is the same, which greatly reduces the difficulty of image processing, facilitates the real-time processing of images, and improves the recognition speed and splicing speed of the tunnel inspection system. In this way, real-time on-site display can be realized.
  • the user can perform on-site inspection of the inspection through the real-time displayed content or notify the relevant departments for processing in time, which greatly improves the timeliness of the tunnel inspection, and achieves high efficiency, short inspection time and high accuracy. , The technical effect of not prone to missed inspection.
  • the step S2 of this embodiment is: the synchronous encoder automatically triggers the synchronous pulse period signal as the dual-track flaw detection vehicle advances, and sends the signal to the CCD camera and auxiliary light source at the same time, so that it works synchronously, and is fixed every time the vehicle travels. Complete an image acquisition after the distance;
  • Step S2 specifically includes the following steps: S21: When the dual-gauge flaw detection vehicle is moving, the encoder generates A+, B+, A-, B- signals with directions and displacements according to the displacement of the vehicle body, and transmits the generated signals to the signal Processor; S22: The signal processor integrates and processes the pulse signal generated by the synchronous encoder, judges the traveling direction and the number of pulses of the dual-track flaw detection vehicle currently equipped with the tunnel inspection system, judges the corresponding code value and uploads it to the computer, and passes Suitable multi-frequency transmission, trigger the CCD area array camera and the light source controller to work; S23: The CCD camera and auxiliary light source controller receives the corresponding pulse signal, the CCD camera triggers the work to collect the tunnel lining image data, the light source controller After receiving the pulse signal, the switching frequency of the light source is controlled to synchronize the shooting frequency of the CCD area camera to ensure that the tunnel is bright enough when shooting by the CCD area camera.
  • Step S3 of this embodiment is: the data collector collects the images in real time, and combines the synchronous encoder to perform numbering processing on the corresponding images through various communication methods: Ethernet, Bluetooth, Zigbee, WLAN or RS232, RS485, etc. Way to upload to the computer;
  • Step S4 of this embodiment is: the computer performs image recognition, marking and online alarming of the numbered tunnel lining image, and at the same time, splicing the tunnel lining image after marking the disease point and displaying it in real time.
  • the computer realizes the recognition of water leakage through local image gray-scale difference characteristics, and realizes automatic recognition functions such as pipeline fall off through linear gray-scale difference characteristics. According to the recognition results, it performs disease point marking and online alarm. Online alarm can be through the display Prompt alarm, also can alarm by indicator light and/or sound, etc.
  • the computer calls the tunnel lining image that has been identified and marked with disease points, and uses the algorithm to stitch the images acquired by the CCD camera at the same time to display the complete tunnel lining image more intuitively.
  • image recognition specifically includes the following steps: A1: through the cooperation of the auxiliary light source of the strobe area LED light source and the CCD area camera, continuously collect and save the entire tunnel lining image and save it; A2: the tunnel The repeating area of the lining image is identified and the image is cut; A3: filter some randomly distributed impurities such as the tunnel segment and platform through image filtering, so that the surface of the tunnel segment, platform, etc.
  • A4 Image enhancement: The overall suppression of the pixel gray value of the surface image of ordinary tunnel segments and platforms, etc., and the overall enhancement of special points such as cracks and water leakage are conducive to the tunnel disease recognition algorithm Quickly screen out the suspected disease points;
  • A5 compare the similarity between the suspected disease points and the theoretical model, and the similarity conforms to the theoretical model to judge the suspected disease points as disease points, and feed back the disease point information.
  • image cropping is used to reduce the recognition area, image filtering to reduce impurity factors that interfere with recognition, image enhancement to suppress impurity factors, and to identify and highlight suspected disease points, and to confirm whether they are disease points through similarity comparison, achieving fast recognition speed and accurate recognition High degree of technical effect.
  • step S4 further includes the following steps: for the identified disease point, write corresponding information on the RFID tag through the positioning device to facilitate later review, and the positioning device may be based on RFID technology.
  • the positioning device reads the tag information through a vehicle-mounted reader, uses Ethernet, Bluetooth, Zigbee, WLAN, or RS232, RS485 and other communication methods to transfer the tag information to the computer and generates a ledger in the background.
  • the system will automatically trigger the mileage correction, if there is a disease in the place after intelligent identification, the inspection system will automatically mark. If you drive through a section where the size of the tunnel structure changes, you can notify the inspection system in advance through an RFID tag, and the inspection system will automatically change the brightness and exposure parameters of the light source when it reaches the location, thereby changing the image depth effect and shooting clear images.
  • the operation of reading and writing RFID tags in this embodiment can mark disease points, adjust the auxiliary light source, and calibrate the mileage of the flaw detection mobile platform, which facilitates later re-inspection, and automatically changes the brightness and exposure parameters of the light source, thereby changing
  • the image depth of field effect captures clear images, and the mileage correction also improves the accuracy of disease point positioning, achieving the technical effects of disease point traceability, accurate inspection and positioning, and clear images.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Signal Processing (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Mining & Mineral Resources (AREA)
  • Quality & Reliability (AREA)
  • Computational Linguistics (AREA)
  • Geology (AREA)
  • Geochemistry & Mineralogy (AREA)
  • General Life Sciences & Earth Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)

Abstract

一种隧道巡检图像采集装置（1），通过多个CCD相机（11）、多个辅助光源（12）、固定支架（13），CCD相机（11）与辅助光源（12）固定于固定支架（13）上，并且多个CCD相机（11）的固定位置位于同一圆环上，使得工作状态下，圆环的中心位于隧道的圆心位置，多个CCD相机（11）与隧道内衬面的距离一致，并且多个CCD相机（11）的整体拍摄角度覆盖隧道内衬面的范围，同时，配合探伤移动平台（2）上，通过计算机（31）、显示器（32）、同步编码器（33）、数据采集器（34），可以实现隧道的同步拍摄、病害识别、实时拼接与显示等功能，该采集装置具有结构简单、检测范围大、时效性高、检测速度快、识别准确度高、巡检定位准、拍摄图像清晰等技术特点。还涉及一种隧道巡检系统及隧道巡检方法。

Description

隧道巡检图像采集装置、隧道巡检系统及隧道巡检方法 技术领域
本发明属于隧道探伤技术领域，尤其涉及隧道巡检图像采集装置、隧道巡检系统及隧道巡检方法。
背景技术
随着中国铁路及城市轨道交通的高速发展，地铁隧道的安全越来越受到重视，隧道中的开裂、渗漏、变形等问题无可避免。而这些问题将会严重影响铁路及轨道交通的行驶安全，因此隧道巡检工作就显得尤其重要。
目前，国内隧道巡检工作主要依靠人工作业，这导致目前隧道巡检作业整体效率低下，准确率低，容易漏检，耗时长等问题。而目前自动智能巡检装置大多用于大型探伤车结构复杂、成本较高，检测范围较小，特别地，目前的隧道巡检探伤车时效性不高，首先需要进行隧道图像采集，待收集完隧道的所有图像，再在后台平台进行隧道图像处理及病害点检测，在发现病害之后再去通知检修等部门去处理，如此导致隧道的巡检及后期处理的周期十分长。
发明内容
本发明为解决上述问题，提供了隧道巡检图像采集装置、隧道巡检系统及隧道巡检方法，其具有结构简单、检测范围大、时效性高、检测速度快、识别准确度高、巡检定位准、拍摄图像清晰的技术特点。
本发明的技术方案为：
本发明提供了一种隧道巡检图像采集装置，应用于隧道内衬面的图像采集，包括：多个CCD相机、多个用于CCD相机拍摄曝光的辅助光源、固定支架，CCD相机与辅助光源固定于固定支架上；其中，多个CCD相机的固定位置位于同一圆环上，工作状态下，圆环的中心位于隧道的圆心位置，以保证多个CCD相机与隧道内衬面的距离一致，并且多个CCD相机的整体拍 摄角度覆盖隧道内衬面的范围。
在其中一个实施例中，CCD相机与辅助光源依次间隔固定于固定支架上。
在其中一个实施例中，相邻的CCD相机之间的拍摄角度部分重叠。
在其中一个实施例中，固定支架包括底架、固定块、多个支撑杆，底架与固定块固定连接，支撑杆的一端与固定块固定连接，支撑杆的另一端与CCD相机或者辅助光源固定连接。
在其中一个实施例中，辅助光源为LED面阵光源。
本发明还提供了一种隧道巡检系统，包括用于钢轨上行进的探伤移动平台，还包括设置于探伤移动平台上的计算机、显示器、同步编码器、数据采集器、以及如上提供的隧道巡检图像采集装置；其中，固定支架设于探伤移动平台上，工作状态下，多个CCD相机所在的圆环中心位于隧道的圆心位置，以保证多个CCD相机与隧道内衬面的距离一致；同步编码器分别与CCD相机、辅助光源电连接，用以根据探伤移动平台的位移量，每隔固定距离发送同步脉冲信号触发多个CCD相机、多个辅助光源同步工作；数据采集器分别与多个CCD相机、同步编码器电连接，用以采集CCD相机拍摄的隧道内衬图像，并结合同步编码器对隧道内衬图像进行编号；计算机分别与数据采集器、显示器电连接，用以对隧道内衬图像进行图像识别及病害点标注，同时，对病害点标注之后的隧道内衬图像进行拼接并实时显示。
在其中一个实施例中，还包括定位装置，定位装置与计算机电连接，计算机与辅助光源电连接，定位装置用以隧道内RFID标签的读取与写入，其中，读取RFID标签内的定位信息，并反馈给计算机进行探伤移动平台的里程校正；读取RFID标签内的隧道结构信息，并反馈给计算机进行辅助光源的亮度与曝光参数调整，以改变图像景深效果从而获取清晰图像；将计算机反馈的病害点信息写入RFID标签进行病害点位置打标。
在其中一个实施例中，计算机、显示器、同步编码器、数据采集器、隧道巡检图像采集装置、定位装置分别可拆卸式设置于探伤移动平台上。
本发明还提供了一种隧道巡检方法，包括以下步骤：
S1：探伤移动平台进入隧道区域，将探伤移动平台上的多个CCD相机所在的圆环中心调整为隧道的圆心位置，以保证多个CCD相机与隧道内衬面的距离一致，然后启动隧道巡检；
S2：根据探伤移动平台的位移量，每隔固定距离发送同步脉冲周期信号触发多个CCD相机、多个用于CCD相机拍摄曝光的辅助光源同步工作，进行隧道内衬图像的拍摄；
S3：对隧道内衬图像进行实时采集，并结合同步脉冲周期信号对隧道内衬图像进行编号；
S4：对编号之后的隧道内衬图像进行图像识别、病害点标注和在线报警，同时，对病害点标注之后的隧道内衬图像进行拼接并实时显示。
在其中一个实施例中，在步骤S1至S4执行过程中，还包括以下步骤：读取隧道内的RFID标签，根据RFID标签的隧道结构信息对辅助光源进行亮度与曝光参数调整，以改变图像景深效果从而获取清晰图像。
在其中一个实施例中，步骤S4还包括以下步骤：若图像识别检测出病害点，则将病害点信息写入隧道内的RFID标签。
在其中一个实施例中，步骤S2具体包括以下步骤：
S21：根据探伤移动平台的位移量产生具有方向与位移量的脉冲信号；
S22：对脉冲信号进行整合及处理，得到同步脉冲周期信号，并判断当前探伤移动平台的行走方向以及脉冲数量获取探伤移动平台的里程，同时，对同步脉冲周期信号进行多分频输出，触发CCD相机及辅助光源同步工作，进行隧道内衬图像的拍摄。
在其中一个实施例中，在步骤S1至S4执行过程中，还包括以下步骤：读取隧道内的RFID标签，根据RFID标签内的定位信息对探伤移动平台进行里程校正。
在其中一个实施例中，步骤S4中，图像识别具体包括以下步骤：
A1：接收隧道内衬图像并存储；
A2：对隧道内衬图像的重复区进行识别与图像剪裁；
A3：对剪裁后的隧道内衬图像进行图像过滤，并对图像过滤得到隧道内衬图像中的杂质因素进行平滑处理；
A4：对杂质因素的像素灰度值进行整体抑制，并对整体抑制后的隧道内衬图像进行初步识别，将疑似病害点的像素灰度值进行整体加强；
A5：通过疑似病害点与理论模型进行相似度对比，判断疑似病害点是否为病害点，并反馈病害点信息。
本发明与现有技术相比具有以下的优点和积极效果：
1)本发明的隧道巡检图像采集装置中，设置多个CCD相机，且多个CCD相机的固定位置位于同一圆环上，工作状态下，圆环的中心位于隧道的圆心位置，如此布置可以保证CCD工业相机镜头与隧道内衬面的距离保持一致，在拍摄过程中无需调整焦距等参数，采集图像不易发生变形，每个镜头拍摄的图像实际大小一致，大大减小了图像处理的难度，便于图像的实时处理，提高图像处理时效性，同时，由于采用同一圆环的设置方式，可大大提高检测范围，最大可达到360°，便于覆盖隧道内衬面的范围，可完成全线隧道断面(如圆形、矩形、马蹄形等)不同视距的清晰拍照，达到了图像易处理、检测范围大的技术效果；
2)本发明的隧道巡检图像采集装置中，CCD相机与辅助光源依次间隔设于固定支架上，每一辅助光源可供两个CCD相机的曝光，提高了辅助光源的利用率，降低了成本，同时在相同的拍摄要求下减小了CCD相机拍摄之间的曝光差异性，进一步提高了图像之间的一致性，提高图像处理时效性；
3)本发明的隧道巡检图像采集装置中，相邻的CCD相机之间的拍摄角度部分重叠，不仅减小了出现漏检区域的可能性，而且利用图像重叠的部分，可降低图像拼接的难度，便于图像的实时处理，提高图像处理时效性；
4)本发明的隧道巡检图像采集装置中，固定支架结构简单轻量化，易于运输存储，而且降低了成本，同时采用LED面阵光源，能够在隧道壁面形成亮度均匀的光带，在相同的拍摄要求下减小了CCD相机拍摄之间的曝光差异性，进一步提高了图像之间的一致性，提高图像处理时效性；
5)本发明的隧道巡检图像采集装置通过设置在探伤移动平台上，并配合计算机、显示器、同步编码器、数据采集器，进行隧道的同步拍摄、病害识别、实时拼接与显示，其中，基于隧道巡检图像采集装置采集的图像易处理性，提高了隧道巡检系统的识别速度与拼接速度，从而可实现现场的实时显示，用户可通过实时显示的内容进行巡检的现场处理或者及时通知有关部门进行处理，大大提高了隧道巡检时效性，达到了效率高、检测时间短、精确度高、不容易发生漏检的技术效果；
6)本发明的隧道巡检系统中，采用可拆卸式的模块化设计，便于可单独存储、运输，提高了系统的便捷性；
7)本发明的隧道巡检方法中，将所述探伤移动平台上的多个CCD相机所在的圆环中心调整为隧道的圆心位置，以保证多个所述CCD相机与隧道内衬面的距离一致，如此，在拍摄过程中无需调整焦距等参数，采集图像不易发生变形，每个镜头拍摄的图像实际大小一致，大大减小了图像处理的难度，便于图像的实时处理，提高了隧道巡检系统的识别速度与拼接速度，从而可实现现场的实时显示，用户可通过实时显示的内容进行巡检的现场处理或者及时通知有关部门进行处理，大大提高了隧道巡检时效性，达到了效率高、检测时间短、精确度高、不容易发生漏检的技术效果；
8)本发明的隧道巡检方法中，读取与写入RFID标签的操作，可进行病害点的标记、辅助光源调整、探伤移动平台的里程校正，方便了后期的复检，以及自动改变光源亮度、曝光参数，从而改变图像景深效果拍摄出清晰图像，同时里程校正也提高了病害点定位的准确性，达到了病害点可追溯、巡检定位准、拍摄图像清晰的技术效果；
9)本发明的隧道巡检方法中，通过图像裁剪减少识别区域、图像过滤减少干扰识别的杂质因素、图像增强抑制杂质因素并识别及突出疑似病害点、通过相似度对比确认是否为病害点，达到了识别速度快、识别准确度高的技术效果。
附图说明
通过阅读下文优选实施方式的详细描述，各种其他的优点和益处对于本领域普通技术人员将变得清楚明了。附图仅用于示出优选实施方式的目的，而并不认为是对本发明的限制。
图1为本发明的一种隧道巡检图像采集装置的立体结构图；
图2为本发明的一种隧道巡检图像采集装置的平面结构图；
图3为本发明的一种隧道巡检系统的整体结构示意图；
图4为本发明的一种隧道巡检系统的安装位置示意图；
图5为本发明的一种隧道巡检系统的架构框图；
图6为本发明的一种隧道巡检系统的同步触发流程图；
图7为本发明的一种隧道巡检系统的病害识别流程图；
图8为本发明的一种隧道检测方法的整体流程图。
附图标记说明：
1-隧道巡检图像采集装置；11-CCD相机；12-辅助光源；13-固定支架；131-支撑杆；132-固定块；133-底架；2-探伤移动平台；31-计算机；32-显示器；33-同步编码器；34-数据采集器；35-定位装置。
具体实施方式
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对照附图说明本发明的具体实施方式。显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图，并获得其他的实施方式。
为使图面简洁，各图中只示意性地表示出了与本发明相关的部分，它们并不代表其作为产品的实际结构。另外，以使图面简洁便于理解，在有些图中具有相同结构或功能的部件，仅示意性地绘示了其中的一个，或仅标出了其中的一个。在本文中，“一个”不仅表示“仅此一个”，也可以表示“多于一个”的情形。
以下结合附图和具体实施例对本发明提出的一种隧道巡检图像采集装置1及隧道巡检系统作进一步详细说明。
实施例1
参看图1，本申请提供了一种隧道巡检图像采集装置，应用于隧道内衬面的图像采集，包括：多个CCD相机11、多个用于CCD相机11拍摄曝光的辅助光源12、固定支架13，CCD相机11与辅助光源12固定于固定支架13上；其中，多个CCD相机11的固定位置位于同一圆环上，工作状态下，圆环的中心位于隧道的圆心位置，以保证多个CCD相机11与隧道内衬面的距离一致，并且多个CCD相机11的整体拍摄角度覆盖隧道内衬面的范围。
现具体对本实施例进行详细说明：
本实施例的隧道巡检图像采集装置可以安装于任何平台上，通过固定支架13进行连接安装，以进行隧道巡检的图像采集。
参看图1和图2，本实施例采用了6台CCD相机11和7个辅助光源12，辅助光源12为频闪面阵光源，通过固定支架13在同一圆环上分布，该布局的优势在于辅助光源12与CCD相机11在同一支架上紧密排列，整体结构 简单。其中，本实施例的每个辅助光源12与CCD相机11间隔22.5°，6个CCD相机11通道，每个CCD相机11环向拍摄角度为48°，以可完全覆盖270°隧道(有部分重叠区域)，辅助光源12采用LED面阵光源，间隔45°排布，能够在隧道壁面形成亮度均匀的光带。上述角度设置、数量设置、设备选择仅是本申请一具体的技术方案，同样地，可以根据隧道的结构、相机的环向拍摄角度等调整上述设计，最大可实现360°的拍摄覆盖，在其中一个实施例中，考虑钢轨的存在及成本，本申请最佳选择270°的整体拍摄角度，即可完成全线隧道断面(如圆形、矩形、马蹄形等)不同视距的清晰拍照。
本实施例设置多个CCD相机11，且多个CCD相机11的固定位置位于同一圆环上，工作状态下，圆环的中心位于隧道的圆心位置，如此布置可以保证CCD工业相机镜头与隧道内衬面的距离保持一致，在拍摄过程中无需调整焦距等参数，采集图像不易发生变形，每个镜头拍摄的图像实际大小一致，大大减小了图像处理的难度，便于图像的实时处理，提高图像处理时效性，同时，由于采用同一圆环的设置方式，可大大提高检测范围，最大可达到360°，便于覆盖隧道内衬面的范围，可完成全线隧道断面不同视距的清晰拍照，达到了图像易处理、检测范围大的技术效果。
具体地，参看图1，本实施例的固定支架13包括底架133、固定块132、多个支撑杆131，底架133与固定块132固定连接，支撑杆131的一端与固定块132固定连接，支撑杆131的另一端与CCD相机11或者辅助光源12固定连接。同样地，固定支架13还可以为圆环型支撑板，而CCD相机11与辅助光源12安装于支撑板上，再辅助的支架进行整体的固定支撑。本实施例的固定支架13结构简单轻量化，易于运输存储，而且降低了成本，同时采用LED面阵光源，能够在隧道壁面形成亮度均匀的光带，在相同的拍摄要求下减小了CCD相机11拍摄之间的曝光差异性，进一步提高了图像之间的一致性，提高图像处理时效性。
较优地，参看图2，CCD相机11与辅助光源12依次间隔固定于固定支架13上。本实施例如此设置，每一辅助光源12可供两个CCD相机11的曝光，提高了辅助光源12的利用率，降低了成本，在相同的拍摄要求下，减小了CCD相机11拍摄之间的曝光差异性，提高了图像之间的一致性，从而提 高图像处理时效性，同时采用LED面阵光源能够在隧道壁面形成亮度均匀的光带，也是为了减小了CCD相机11拍摄之间的曝光差异性，进一步提高了图像之间的一致性，从而提高图像处理时效性。
较优地，本实施例的相邻的CCD相机11之间的拍摄角度部分重叠。本发明相邻的CCD相机11之间的拍摄角度部分重叠，不仅减小了出现漏检区域的可能性，而且利用图像重叠的部分，可降低图像拼接的难度，便于图像的实时处理，提高图像处理时效性。
实施例2
参看图3，本申请提供了一种基于实施例1的隧道巡检系统，包括用于钢轨上行进的探伤移动平台2，还包括设置于探伤移动平台2上的计算机31、显示器32、同步编码器33、数据采集器34、以及如实施例1的隧道巡检图像采集装置1；
其中，固定支架13设于探伤移动平台2上，工作状态下，多个CCD相机11所在的圆环中心位于隧道的圆心位置，以保证多个CCD相机11与隧道内衬面的距离一致；同步编码器33分别与CCD相机11、辅助光源12电连接，用以根据探伤移动平台2的位移量，每隔固定距离发送同步脉冲信号触发多个CCD相机11、多个辅助光源12同步工作；数据采集器34分别与多个CCD相机11、同步编码器33电连接，用以采集CCD相机11拍摄的隧道内衬图像，并结合同步编码器33对隧道内衬图像进行编号；计算机31分别与数据采集器34、显示器32电连接，用以对隧道内衬图像进行图像识别及病害点标注，同时，对病害点标注之后的隧道内衬图像进行拼接并实时显示。
现具体对本实施例进行详细说明：
参看图4和图5，本实施例的探伤移动平台2为双轨探伤车，可在钢轨上移动，其中，隧道巡检图像采集装置1固定安装于车体尾部并突出车体，安装好的隧道巡检图像采集装置1，多个CCD相机11所在的圆环中心位于隧道的圆心位置，此种布置方式可以保证CCD工业相机镜头与隧道内衬面的距离保持一致。双轨探伤车的前部设有计算机31和显示器32，以用于操控整个双轨探伤车以及隧道巡检系统，用户可以在现场实时调整隧道巡检系统的相关参数，以及可以现场实时查看隧道巡检系统拍摄的照片，以现场确认隧道是否有病害点，若有则可及时通知相关部门进行检修等操作。
本实施例的同步编码器33可安装于探伤移动平台2的移动轴承上，根据轴承的旋转产生具有方向和位移量的脉冲信号，以进行CCD相机11与光源的同步触发，可采用行驶固定的距离同步触发一次拍摄的方式，具体地，参看图6，根据车体的移动产生A+、B+、A-、B-信号,并将产生的信号传输至信号处理器，信号处理器对同步编码器33产生的脉冲信号进行整合及处理，判断当前搭载智能隧道巡检系统的双轨探伤车的行走方向以及脉冲数量，判断相应编码值并上传至计算机31以用于图片存储，并通过合适的多分频传输，触发CCD面阵相机及光源控制器同步工作，CCD面阵相机及光源控制器接收到相应脉冲信号，CCD面阵相机触发工作用于采集隧道内衬图像数据，光源控制器接受脉冲信号后控制光源的开关频率同步CCD面阵相机拍摄频率，保证CCD面阵相机拍摄时隧道内足够明亮。
本实施例的数据采集器34可为交换机，交换机可以为多路转一路的交换机，可实现数据汇总、简化相机到计算机31通讯连接线路。计算机31接收到数据采集器34传送的隧道内衬图像后，通过计算机31进行病害点图像识别，其中，通过前期大量现场采集隧道内部图像，大量分析统计人行站台、电气线缆管道、隧道管片接缝、消防通道指示灯、电器盒、消防紧急电话、列车站台、消防管道、楼梯等正常图像与问题图像内容形成的区域灰度差异，为隧道巡检各类智能识别做基础技术准备。
参看图7，病害点图像识别具体过程为：第一步：系统通过频闪面阵LED光源和面阵CCD工业相机的配合，连续采集整个隧道270°隧道内衬图像，并保存下来，对图像重复区域进行图像剪裁；第二步：通过图像过滤将隧道管片、站台等的一些随机分布的杂质进行过滤，使得隧道管片、站台等表面更加平滑，有助于提高隧道智能算法的效率；第三步：图像增强是对拍摄到的普通隧道管片、站台等表面图像像素灰度值进行整体抑制，对类似裂纹、渗漏水等特殊点整体加强，有利于隧道病害识别算法快速筛选出疑似病害点；第四步：通过疑似伤损点与理论模型对比相似度，相似度符合理论模型的则进行异常报警，生成问题点报告。
参看图3，较优地，还包括定位装置35，定位装置35与计算机31电连接，计算机31与辅助光源12电连接，定位装置35用以隧道内RFID标签的读取与写入，其中，读取RFID标签内的定位信息，并反馈给计算机31进行 探伤移动平台2的里程校正；读取RFID标签内的隧道结构信息，并反馈给计算机31进行辅助光源12的亮度与曝光参数调整，以改变图像景深效果从而获取清晰图像；将计算机31反馈的病害点信息写入RFID标签进行病害点位置打标。
具体地，定位装置35设有车载读写器，通过定位系统及时读取标签信息，用以太网、蓝牙、Zigbee、WLAN或者RS232、RS485等通讯方式将标签信息传入计算机31并在后台生成台账，优选地，如果该标签处是整公里里程标签、系统会自动触发里程校正，以减少里程校正频繁性，若该处经过智能识别有病害则计算机31会自动打标，即控制车载读写器在RFID标签中写入病害信息，以便于后期复查。若行驶经过隧道结构尺寸变化地段，可以通过标签提前通知巡检系统，巡检系统到达该位置自动改变光源亮度、曝光参数，从而改变图像景深效果，拍摄清晰图像。优选地，车载读写器基于RFID技术。本实施例的隧道巡检系统通过定位装置35读取或写入RFID标签，可进行病害点的标记、探伤移动平台2的里程校正，方便了后期的复检，同时里程校正也提高了病害点定位的准确性，达到了巡检定位准的技术效果。
较优地，计算机31、显示器32、同步编码器33、数据采集器34、隧道巡检图像采集装置1、定位装置35分别可拆卸式设置于探伤移动平台2上。本实施例的隧道巡检系统采用可拆卸式的模块化设计，便于可单独存储、运输，提高了系统的便捷性。
现对本实施例的工作过程进行说明：
隧道巡检系统整体工作流程如下：
S1：车辆行驶进入隧道区域，开始检测；
S2：同步编码器33随着双轨探伤车前进自触发同步脉冲周期信号，并同时将信号发送给CCD工业相机以及光源控制器，使其同步工作，并在行驶固定的距离后完成一次图像采集；
S3：双轨探伤车行驶过程中，通过RFID定位系统进行整公里打标，在隧道尺寸变化处通过RFID标签提前通知计算机31，隧道巡检系统到达该位置自动改变光源亮度、曝光参数，从而改变图像景深效果，拍摄清晰图像
S4：数据采集器34对图像实时进行采集，并结合同步编码器33对相应的图像进行编号处理后通过各种通讯方式：以太网、蓝牙、Zigbee、WLAN 或者RS232、RS485等通讯方式上传给计算机31；
S5：在计算机31内智能识别系统通过局部图像灰度差异性特征实现渗漏水识别，通过线性灰度差异性特征实现管线脱落等自动识别功能。对于识别得到的病害点，通过RFID定位装置35在RFID标签处写入相应信息，便于后期复查；
S6：计算机31将经过识别并标注病害点的隧道图像调用，并通过算法将同一时刻6个CCD相机11获取的图像进行拼接以更加直观的展示完整的隧道内衬图像。
本实施例的隧道巡检图像采集装置1通过设置在探伤移动平台2上，并配合计算机31、显示器32、同步编码器33、数据采集器34，进行隧道的同步拍摄、病害识别、实时拼接与显示，其中，基于实施例1的隧道巡检图像采集装置1采集的图像易处理性，提高了隧道巡检系统的识别速度与拼接速度，从而可实现现场的实时显示，用户可通过实时显示的内容进行巡检的现场处理或者及时通知有关部门进行处理，大大提高了隧道巡检时效性，达到了效率高、检测时间短、精确度高、不容易发生漏检的技术效果。
实施例3
参看图8，本申请提供了一种隧道巡检方法，包括以下步骤：
S1：探伤移动平台进入隧道区域，将探伤移动平台上的多个CCD相机所在的圆环中心调整为隧道的圆心位置，以保证多个CCD相机与隧道内衬面的距离一致，然后启动隧道巡检；
S2：根据探伤移动平台的位移量，每隔固定距离发送同步脉冲周期信号触发多个CCD相机、多个用于CCD相机拍摄曝光的辅助光源同步工作，进行隧道内衬图像的拍摄；
S3：对隧道内衬图像进行实时采集，并结合同步脉冲周期信号对隧道内衬图像进行编号；
S4：对编号之后的隧道内衬图像进行图像识别、病害点标注和在线报警，同时，对病害点标注之后的隧道内衬图像进行拼接并实时显示。
现对本实施例进行详细说明：
参看图1、图3和图4，本实施例的探伤移动平台2可以为双轨探伤车，可在钢轨上移动，双轨探伤车上安装有隧道巡检系统，用以执行本实施例的 隧道巡检方法，隧道巡检系统包括：计算机31、显示器32、同步编码器33、数据采集器34、定位装置35、隧道巡检图像采集装置1，隧道巡检图像采集装置1包括CCD相机11、辅助光源12、固定支架13，固定支架又包括支撑杆131、固定块132、底架133。本实施例的隧道巡检方法的可执行体包括但不限于上述的双轨探伤车。
参看图5，本实施例的步骤S1为：探伤移动平台进入隧道区域，将探伤移动平台上的多个CCD相机所在的圆环中心调整为隧道的圆心位置，以保证多个CCD相机与隧道内衬面的距离一致，然后启动隧道巡检。
本实施例将所述探伤移动平台上的多个CCD相机所在的圆环中心调整为隧道的圆心位置，以保证多个所述CCD相机与隧道内衬面的距离一致，如此，在拍摄过程中无需调整焦距等参数，采集图像不易发生变形，每个镜头拍摄的图像实际大小一致，大大减小了图像处理的难度，便于图像的实时处理，提高了隧道巡检系统的识别速度与拼接速度，从而可实现现场的实时显示，用户可通过实时显示的内容进行巡检的现场处理或者及时通知有关部门进行处理，大大提高了隧道巡检时效性，达到了效率高、检测时间短、精确度高、不容易发生漏检的技术效果。
参看图6，本实施例的步骤S2为：同步编码器随着双轨探伤车前进自触发同步脉冲周期信号，并同时将信号发送给CCD相机以及辅助光源，使其同步工作，并每行驶固定的距离后完成一次图像采集；
步骤S2具体包括以下步骤：S21：在双规探伤车移动时，编码器根据车体的位移量产生具有方向与位移量的A+、B+、A-、B-信号，并将产生的信号传输至信号处理器；S22：信号处理器对同步编码器产生的脉冲信号进行整合及处理，判断当前搭载隧道巡检系统的双轨探伤车的行走方向以及脉冲数量，判断相应编码值并上传至计算机，并通过合适的多分频传输，触发CCD面阵相机及光源控制器工作；S23：CCD相机及辅助光源的控制器接收到相应脉冲信号，CCD相机触发工作用于采集隧道内衬图像数据，光源控制器接受脉冲信号后控制光源的开关频率同步CCD面阵相机拍摄频率，保证CCD面阵相机拍摄时隧道内足够明亮。
本实施例的步骤S3为：数据采集器对图像实时进行采集，并结合同步编码器对相应的图像进行编号处理后通过各种通讯方式：以太网、蓝牙、 Zigbee、WLAN或者RS232、RS485等通讯方式上传给计算机；
本实施例的步骤S4为：计算机对编号之后的隧道内衬图像进行图像识别、病害点标注和在线报警，同时，对病害点标注之后的隧道内衬图像进行拼接并实时显示。具体地，计算机通过局部图像灰度差异性特征实现渗漏水识别，通过线性灰度差异性特征实现管线脱落等自动识别功能，根据识别的结果进行病害点标记以及在线报警，在线报警可通过显示器进行提示报警，也可通过指示灯和/或声音等进行报警。计算机将经过识别并标注病害点的隧道内衬图像调用，并通过算法将同一时刻CCD相机获取的图像进行拼接以更加直观的展示完整的隧道内衬图像。
其中，通过前期大量现场采集隧道内部图像，大量分析统计人行站台、电气线缆管道、隧道管片接缝、消防通道指示灯、电器盒、消防紧急电话、列车站台、消防管道、楼梯等正常图像与问题图像内容形成的区域灰度差异，为隧道巡检的图像识别做基础技术准备。具体地，参看图7，图像识别具体包括以下步骤：A1：通过频闪面阵LED光源的辅助光源和CCD面阵相机的配合，连续采集整个隧道内衬图像，并保存下来；A2：对隧道内衬图像的重复区进行识别与图像剪裁；A3：通过图像过滤将隧道管片、站台等的一些随机分布的杂质进行过滤，使的隧道管片、站台等表面更加平滑，有助于提高隧道智能算法的效率；A4：图像增强：对拍摄到的普通隧道管片、站台等表面图像像素灰度值进行整体抑制，对类似裂纹、渗漏水等特殊点整体加强，有利于隧道病害识别算法快速筛选出疑似病害点；A5：通过疑似病害点与理论模型进行相似度对比，相似度符合理论模型的判断疑似病害点为病害点，并反馈病害点信息。
本实施例通过图像裁剪减少识别区域、图像过滤减少干扰识别的杂质因素、图像增强抑制杂质因素并识别及突出疑似病害点、通过相似度对比确认是否为病害点，达到了识别速度快、识别准确度高的技术效果。
较优地，步骤S4还包括以下步骤：对于识别得到的病害点，通过定位装置在RFID标签处写入相应信息，便于后期复查，定位装置可基于RFID技术。
较优地，在步骤S1至S4执行过程中，即双轨探伤车行驶过程中，通过定位装置进行整公里打标，在隧道尺寸变化处通过RFID标签提前通知隧道 巡检系统，隧道巡检系统到达该位置自动改变光源亮度、曝光参数，从而改变图像景深效果，拍摄清晰图像。具体地，定位装置通过车载阅读器进行读取标签信息，用以太网、蓝牙、Zigbee、WLAN或者RS232、RS485等通讯方式将标签信息传入计算机并在后台生成台账，如果该标签处是整公里里程标签、系统会自动触发里程校正，若该处经过智能识别有病害则巡检系统会自动打标。若行驶经过隧道结构尺寸变化地段，可以通过RFID标签提前通知巡检系统，巡检系统到达该位置自动改变光源亮度、曝光参数，从而改变图像景深效果，拍摄清晰图像。
本实施例的读取与写入RFID标签的操作，可进行病害点的标记、辅助光源调整、探伤移动平台的里程校正，方便了后期的复检，以及自动改变光源亮度、曝光参数，从而改变图像景深效果拍摄出清晰图像，同时里程校正也提高了病害点定位的准确性，达到了病害点可追溯、巡检定位准、图像清晰的技术效果。
上面结合附图对本发明的实施方式作了详细说明，但是本发明并不限于上述实施方式。即使对本发明作出各种变化，倘若这些变化属于本发明权利要求及其等同技术的范围之内，则仍落入在本发明的保护范围之中。

Claims (14)

  1. 一种隧道巡检图像采集装置，应用于隧道内衬面的图像采集，其特征在于，包括：多个CCD相机、多个用于所述CCD相机拍摄曝光的辅助光源、固定支架，所述CCD相机与所述辅助光源固定于所述固定支架上；其中，
    多个所述CCD相机的固定位置位于同一圆环上，工作状态下，所述圆环的中心位于隧道的圆心位置，以保证多个所述CCD相机与隧道内衬面的距离一致，并且多个所述CCD相机的整体拍摄角度覆盖隧道内衬面的范围。
  2. 根据权利要求1所述的隧道巡检图像采集装置，其特征在于，所述CCD相机与所述辅助光源依次间隔固定于所述固定支架上。
  3. 根据权利要求1所述的隧道巡检图像采集装置，其特征在于，相邻的所述CCD相机之间的拍摄角度部分重叠。
  4. 根据权利要求1至3任意一项所述的隧道巡检图像采集装置，其特征在于，所述固定支架包括底架、固定块、多个支撑杆，所述底架与所述固定块固定连接，所述支撑杆的一端与所述固定块固定连接，所述支撑杆的另一端与所述CCD相机或者所述辅助光源固定连接。
  5. 根据权利要求1至3任意一项所述的隧道巡检图像采集装置，其特征在于，所述辅助光源为LED面阵光源。
  6. 一种隧道巡检系统，包括用于钢轨上行进的探伤移动平台，其特征在于，还包括设置于所述探伤移动平台上的计算机、显示器、同步编码器、数据采集器、以及如权利要求1至5任意一项所述的隧道巡检图像采集装置；其中，
    所述固定支架设于所述探伤移动平台上，工作状态下，多个所述CCD相机所在的圆环中心位于所述隧道的圆心位置，以保证多个所述CCD相机与隧道内衬面的距离一致；
    所述同步编码器分别与所述CCD相机、所述辅助光源电连接，用以根据所述探伤移动平台的位移量，每隔固定距离发送同步脉冲信号触发多个所述CCD相机、多个所述辅助光源同步工作；
    所述数据采集器分别与多个所述CCD相机、所述同步编码器电连接， 用以采集所述CCD相机拍摄的隧道内衬图像，并结合所述同步编码器对所述隧道内衬图像进行编号；
    所述计算机分别与所述数据采集器、所述显示器电连接，用以对所述隧道内衬图像进行图像识别及病害点标注，同时，对所述病害点标注之后的所述隧道内衬图像进行拼接并实时显示。
  7. 根据权利要求6所述的隧道巡检系统，其特征在于，还包括定位装置，所述定位装置与所述计算机电连接，所述计算机与所述辅助光源电连接，所述定位装置用以隧道内RFID标签的读取与写入，其中，
    读取所述RFID标签内的定位信息，并反馈给所述计算机进行所述探伤移动平台的里程校正；
    读取所述RFID标签内的隧道结构信息，并反馈给所述计算机进行所述辅助光源的亮度与曝光参数调整，以改变图像景深效果从而获取清晰图像；
    将所述计算机反馈的病害点信息写入所述RFID标签进行病害点位置打标。
  8. 根据权利要求7所述的隧道巡检系统，其特征在于，所述计算机、所述显示器、所述同步编码器、所述数据采集器、所述隧道巡检图像采集装置、所述定位装置分别可拆卸式设置于所述探伤移动平台上。
  9. 一种隧道巡检方法，其特征在于，包括以下步骤：
    S1：探伤移动平台进入隧道区域，将所述探伤移动平台上的多个CCD相机所在的圆环中心调整为隧道的圆心位置，以保证多个所述CCD相机与隧道内衬面的距离一致，然后启动隧道巡检；
    S2：根据所述探伤移动平台的位移量，每隔固定距离发送同步脉冲周期信号触发多个所述CCD相机、多个用于所述CCD相机拍摄曝光的辅助光源同步工作，进行隧道内衬图像的拍摄；
    S3：对所述隧道内衬图像进行实时采集，并结合所述同步脉冲周期信号对所述隧道内衬图像进行编号；
    S4：对所述编号之后的所述隧道内衬图像进行图像识别、病害点标注和在线报警，进行图像识别及病害点标注和在线报警同时，对所述病害点标注之后的所述隧道内衬图像进行拼接并实时显示。
  10. 根据权利要求9所述的隧道巡检方法，其特征在于，在所述步骤S1 至S4执行过程中，还包括以下步骤：读取隧道内的RFID标签，根据所述RFID标签的隧道结构信息对所述辅助光源进行亮度与曝光参数调整，以改变图像景深效果从而获取清晰图像。
  11. 根据权利要求9所述的隧道巡检方法，其特征在于，所述步骤S4还包括以下步骤：若所述图像识别检测出病害点，则将病害点信息写入隧道内的RFID标签。
  12. 根据权利要求9至11任意一项所述的隧道巡检方法，其特征在于，所述步骤S2具体包括以下步骤：
    S21：根据所述探伤移动平台的位移量产生具有方向与位移量的脉冲信号；
    S22：对所述脉冲信号进行整合及处理，得到所述同步脉冲周期信号，并判断当前所述探伤移动平台的行走方向以及脉冲数量获取所述探伤移动平台的里程，同时，对所述同步脉冲周期信号进行多分频输出，触发所述CCD相机及辅助光源同步工作，进行隧道内衬图像的拍摄。
  13. 根据权利要求12所述的隧道巡检方法，其特征在于，在所述步骤S1至S4执行过程中，还包括以下步骤：读取隧道内的RFID标签，根据所述RFID标签内的定位信息对所述探伤移动平台进行里程校正。
  14. 根据权利要求9至11任意一项所述的隧道巡检方法，其特征在于，所述步骤S4中，所述图像识别具体包括以下步骤：
    A1：接收所述隧道内衬图像并存储；
    A2：对所述隧道内衬图像的重复区进行识别与图像剪裁；
    A3：对剪裁后的所述隧道内衬图像进行图像过滤，并对图像过滤得到所述隧道内衬图像中的杂质因素进行平滑处理；
    A4：对所述杂质因素的像素灰度值进行整体抑制，并对所述整体抑制后的所述隧道内衬图像进行初步识别，将疑似病害点的像素灰度值进行整体加强；
    A5：通过所述疑似病害点与理论模型进行相似度对比，判断所述疑似病害点是否为病害点，并反馈病害点信息。
PCT/CN2020/117088 2019-10-08 2020-09-23 隧道巡检图像采集装置、隧道巡检系统及隧道巡检方法 Ceased WO2021068746A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US17/767,437 US12045966B2 (en) 2019-10-08 2020-09-23 Tunnel-inspection image acquisition device, tunnel-inspection system, and tunnel-inspection method
EP20873699.1A EP4043693B1 (en) 2019-10-08 2020-09-23 Image acquisition device for tunnel inspection, tunnel inspection system, and tunnel inspection method

Applications Claiming Priority (6)

Application Number Priority Date Filing Date Title
CN201921669806.4U CN210720194U (zh) 2019-10-08 2019-10-08 一种隧道巡检图像采集装置及隧道巡检系统
CN201910949526.7A CN110593957B (zh) 2019-10-08 2019-10-08 一种隧道巡检方法
CN201910949526.7 2019-10-08
CN201910949507.4A CN110487816A (zh) 2019-10-08 2019-10-08 一种隧道巡检图像采集装置及隧道巡检系统
CN201910949507.4 2019-10-08
CN201921669806.4 2019-10-08

Publications (1)

Publication Number Publication Date
WO2021068746A1 true WO2021068746A1 (zh) 2021-04-15

Family

ID=75437005

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/117088 Ceased WO2021068746A1 (zh) 2019-10-08 2020-09-23 隧道巡检图像采集装置、隧道巡检系统及隧道巡检方法

Country Status (3)

Country Link
US (1) US12045966B2 (zh)
EP (1) EP4043693B1 (zh)
WO (1) WO2021068746A1 (zh)

Cited By (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113984288A (zh) * 2021-10-22 2022-01-28 中国能源建设集团江苏省电力设计院有限公司 一种电缆隧道衬砌渗漏水检测装置及方法
CN113989117A (zh) * 2021-10-28 2022-01-28 武汉光谷卓越科技股份有限公司 基于相机阵列的隧道图像动态拼接方法、装置及电子设备
CN114166848A (zh) * 2021-11-24 2022-03-11 中国铁道科学研究院集团有限公司铁道建筑研究所 一种隧道检测车及其检测方法
CN114517710A (zh) * 2021-12-20 2022-05-20 江苏三恒科技股份有限公司 一种应用于矿井的挂轨式巡检机器人系统
CN114697508A (zh) * 2022-05-05 2022-07-01 中国煤炭科工集团太原研究院有限公司 一种矿用钢带孔定位摄像装置及钢带孔自动定位和识别方法
CN114777693A (zh) * 2022-05-05 2022-07-22 杭州耕德电子有限公司 一种使用光来检测产品平面度的检测装置及检测方法
CN114827421A (zh) * 2022-05-05 2022-07-29 中国煤炭科工集团太原研究院有限公司 一种紧凑型自防护矿用定位摄像装置及钢带孔自动定位和识别方法
CN114913473A (zh) * 2022-03-21 2022-08-16 中国科学院光电技术研究所 一种轻量化单体式成像的接触网安全巡检仪
CN114910484A (zh) * 2022-07-19 2022-08-16 山西省信息产业技术研究院有限公司 一种隧道病害检点标记方法、系统、工控单元及介质
CN115242951A (zh) * 2022-07-27 2022-10-25 国家海洋环境监测中心 一种基于图像扫描系统的浮游动物快速检测方法
CN115900687A (zh) * 2023-01-06 2023-04-04 西安华创马科智能控制系统有限公司 液压支架机器人轨道定位方法及装置
CN116132814A (zh) * 2022-04-01 2023-05-16 港珠澳大桥管理局 海底沉管拼接结构信息采集设备、采集方法、装置和设备
CN116718236A (zh) * 2023-08-10 2023-09-08 山东联合能源管道输送有限公司 一种基于计算机视觉的检测系统及使用方法
CN116838906A (zh) * 2023-09-04 2023-10-03 山东省地质科学研究院 一种地质剖面摄影成像装置及控制方法
CN117468941A (zh) * 2023-12-28 2024-01-30 四川省铁路建设有限公司 基于智能自检台车的隧道缺陷检测方法及自检台车
CN117830856A (zh) * 2024-01-24 2024-04-05 青岛华正信息技术股份有限公司 一种园区环境监测图像智能优化方法及装置
CN119151880A (zh) * 2024-08-26 2024-12-17 山东金宇信息科技集团有限公司 一种公路隧道安全隐患检测方法、设备及介质
CN121074036A (zh) * 2025-11-06 2025-12-05 宁波市电力设计院有限公司 一种用于电缆沟的监测方法和监测系统

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN119887690B (zh) * 2024-12-27 2026-04-21 中国铁道科学研究院集团有限公司城市轨道交通中心 一种二阶段隧道表观通信信号线缆脱落检测方法及系统

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20120086963A (ko) * 2011-01-27 2012-08-06 조동혁 투명창을 부착한 케이스 내부에서 이동하면서 감시하는 원격 감시카메라 시스템
CN104567708A (zh) * 2015-01-06 2015-04-29 浙江工业大学 基于主动式全景视觉的隧道全断面高速动态健康检测装置与方法
CN105510351A (zh) * 2016-01-28 2016-04-20 北京工业大学 一种隧道检测车
CN109115434A (zh) * 2018-06-27 2019-01-01 杭州国翌科技有限公司 一种隧道健康监测系统及方法
CN109246364A (zh) * 2018-10-26 2019-01-18 华为技术有限公司 摄像装置调整方法及相关设备
CN110487816A (zh) * 2019-10-08 2019-11-22 上海市东方海事工程技术有限公司 一种隧道巡检图像采集装置及隧道巡检系统
CN110593957A (zh) * 2019-10-08 2019-12-20 上海市东方海事工程技术有限公司 一种隧道巡检方法
CN210720194U (zh) * 2019-10-08 2020-06-09 上海市东方海事工程技术有限公司 一种隧道巡检图像采集装置及隧道巡检系统

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3276621B2 (ja) * 1999-11-11 2002-04-22 計測検査株式会社 トンネルの内部壁面のひび割れ検出方法
CN104019742A (zh) * 2014-06-05 2014-09-03 武汉武大卓越科技有限责任公司 隧道衬砌裂缝快速检测方法
JP6444086B2 (ja) * 2014-07-25 2018-12-26 西日本高速道路エンジニアリング四国株式会社 トンネル覆工面調査システムおよびトンネル覆工面調査システムに用いる車両
JP6373111B2 (ja) * 2014-07-25 2018-08-15 西日本高速道路エンジニアリング四国株式会社 トンネル覆工面調査システムおよびトンネル覆工面調査システムに用いる車両
EP3442213B1 (en) * 2017-08-09 2020-12-09 Ricoh Company, Ltd. Structure wall imaging device, vehicle, and structure wall imaging method
CN208283293U (zh) * 2018-01-10 2018-12-25 南京火眼猴信息科技有限公司 用于隧道检测图像捕获装置的补光装置
EP3748341A4 (en) * 2018-02-02 2021-02-24 Fujifilm Corporation IMAGE PROCESSING DEVICE AND IMAGE PROCESSING METHOD
CN108593654B (zh) * 2018-03-28 2020-08-25 北京交通大学 隧道图像采集系统及方法
CN108615233B (zh) * 2018-04-17 2021-11-23 成都唐源电气股份有限公司 一种吊弦图像处理方法、基于线阵相机的吊弦抓拍系统
JP7330710B2 (ja) * 2018-04-26 2023-08-22 キヤノン株式会社 情報処理装置、情報処理方法及びプログラム
JP7279438B2 (ja) * 2019-03-19 2023-05-23 株式会社リコー 撮像装置、車両及び撮像方法
WO2021014754A1 (ja) * 2019-07-23 2021-01-28 富士フイルム株式会社 ひび割れ評価装置、ひび割れ評価方法、及びひび割れ評価プログラム

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20120086963A (ko) * 2011-01-27 2012-08-06 조동혁 투명창을 부착한 케이스 내부에서 이동하면서 감시하는 원격 감시카메라 시스템
CN104567708A (zh) * 2015-01-06 2015-04-29 浙江工业大学 基于主动式全景视觉的隧道全断面高速动态健康检测装置与方法
CN105510351A (zh) * 2016-01-28 2016-04-20 北京工业大学 一种隧道检测车
CN109115434A (zh) * 2018-06-27 2019-01-01 杭州国翌科技有限公司 一种隧道健康监测系统及方法
CN109246364A (zh) * 2018-10-26 2019-01-18 华为技术有限公司 摄像装置调整方法及相关设备
CN110487816A (zh) * 2019-10-08 2019-11-22 上海市东方海事工程技术有限公司 一种隧道巡检图像采集装置及隧道巡检系统
CN110593957A (zh) * 2019-10-08 2019-12-20 上海市东方海事工程技术有限公司 一种隧道巡检方法
CN210720194U (zh) * 2019-10-08 2020-06-09 上海市东方海事工程技术有限公司 一种隧道巡检图像采集装置及隧道巡检系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4043693A4 *

Cited By (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113984288B (zh) * 2021-10-22 2024-05-03 中国能源建设集团江苏省电力设计院有限公司 一种电缆隧道衬砌渗漏水检测装置及方法
CN113984288A (zh) * 2021-10-22 2022-01-28 中国能源建设集团江苏省电力设计院有限公司 一种电缆隧道衬砌渗漏水检测装置及方法
CN113989117A (zh) * 2021-10-28 2022-01-28 武汉光谷卓越科技股份有限公司 基于相机阵列的隧道图像动态拼接方法、装置及电子设备
CN114166848A (zh) * 2021-11-24 2022-03-11 中国铁道科学研究院集团有限公司铁道建筑研究所 一种隧道检测车及其检测方法
CN114517710B (zh) * 2021-12-20 2023-01-03 江苏三恒科技股份有限公司 一种应用于矿井的挂轨式巡检机器人系统
CN114517710A (zh) * 2021-12-20 2022-05-20 江苏三恒科技股份有限公司 一种应用于矿井的挂轨式巡检机器人系统
CN114913473B (zh) * 2022-03-21 2023-08-15 中国科学院光电技术研究所 一种轻量化单体式成像的接触网安全巡检仪
CN114913473A (zh) * 2022-03-21 2022-08-16 中国科学院光电技术研究所 一种轻量化单体式成像的接触网安全巡检仪
CN116132814B (zh) * 2022-04-01 2023-12-22 港珠澳大桥管理局 海底沉管拼接结构信息采集设备、采集方法、装置和设备
CN116132814A (zh) * 2022-04-01 2023-05-16 港珠澳大桥管理局 海底沉管拼接结构信息采集设备、采集方法、装置和设备
CN114827421A (zh) * 2022-05-05 2022-07-29 中国煤炭科工集团太原研究院有限公司 一种紧凑型自防护矿用定位摄像装置及钢带孔自动定位和识别方法
CN114777693A (zh) * 2022-05-05 2022-07-22 杭州耕德电子有限公司 一种使用光来检测产品平面度的检测装置及检测方法
CN114827421B (zh) * 2022-05-05 2024-06-07 中国煤炭科工集团太原研究院有限公司 矿用自防护定位摄像装置及钢带孔自动定位和识别方法
CN114697508A (zh) * 2022-05-05 2022-07-01 中国煤炭科工集团太原研究院有限公司 一种矿用钢带孔定位摄像装置及钢带孔自动定位和识别方法
CN114777693B (zh) * 2022-05-05 2024-03-22 杭州耕德电子有限公司 一种使用光来检测产品平面度的检测装置及检测方法
CN114910484A (zh) * 2022-07-19 2022-08-16 山西省信息产业技术研究院有限公司 一种隧道病害检点标记方法、系统、工控单元及介质
CN114910484B (zh) * 2022-07-19 2022-11-04 山西省信息产业技术研究院有限公司 一种隧道病害检点标记方法、系统、工控单元及介质
CN115242951A (zh) * 2022-07-27 2022-10-25 国家海洋环境监测中心 一种基于图像扫描系统的浮游动物快速检测方法
CN115900687A (zh) * 2023-01-06 2023-04-04 西安华创马科智能控制系统有限公司 液压支架机器人轨道定位方法及装置
CN116718236B (zh) * 2023-08-10 2023-11-07 山东联合能源管道输送有限公司 一种基于计算机视觉的检测系统及使用方法
CN116718236A (zh) * 2023-08-10 2023-09-08 山东联合能源管道输送有限公司 一种基于计算机视觉的检测系统及使用方法
CN116838906B (zh) * 2023-09-04 2024-01-09 山东省地质科学研究院 一种地质剖面摄影成像装置及控制方法
CN116838906A (zh) * 2023-09-04 2023-10-03 山东省地质科学研究院 一种地质剖面摄影成像装置及控制方法
CN117468941A (zh) * 2023-12-28 2024-01-30 四川省铁路建设有限公司 基于智能自检台车的隧道缺陷检测方法及自检台车
CN117468941B (zh) * 2023-12-28 2024-03-12 四川省铁路建设有限公司 基于智能自检台车的隧道缺陷检测方法及自检台车
CN117830856A (zh) * 2024-01-24 2024-04-05 青岛华正信息技术股份有限公司 一种园区环境监测图像智能优化方法及装置
CN117830856B (zh) * 2024-01-24 2024-06-04 青岛华正信息技术股份有限公司 一种园区环境监测图像智能优化方法及装置
CN119151880A (zh) * 2024-08-26 2024-12-17 山东金宇信息科技集团有限公司 一种公路隧道安全隐患检测方法、设备及介质
CN121074036A (zh) * 2025-11-06 2025-12-05 宁波市电力设计院有限公司 一种用于电缆沟的监测方法和监测系统

Also Published As

Publication number Publication date
US12045966B2 (en) 2024-07-23
EP4043693A1 (en) 2022-08-17
EP4043693B1 (en) 2025-12-24
EP4043693A4 (en) 2023-10-11
US20240070831A1 (en) 2024-02-29

Similar Documents

Publication Publication Date Title
WO2021068746A1 (zh) 隧道巡检图像采集装置、隧道巡检系统及隧道巡检方法
CN110487816A (zh) 一种隧道巡检图像采集装置及隧道巡检系统
CN210720194U (zh) 一种隧道巡检图像采集装置及隧道巡检系统
CN110593957B (zh) 一种隧道巡检方法
KR101111569B1 (ko) 궤도차량을 이용한 철도시설물 감시시스템 및 방법
CN207991478U (zh) 铁路隧道衬砌表面病害快速检测系统
CN102564335B (zh) 一种大型隧道变形测量方法
CN104914108B (zh) 基于机器视觉的高速公路隧道检测车系统
CN108593654B (zh) 隧道图像采集系统及方法
CN202098297U (zh) 接触网关键悬挂自动巡查装置
CN103901040A (zh) 一种基于机器视觉的三维网状物缺陷在线检测系统
CN103217111A (zh) 一种非接触式接触线几何参数检测方法
CN206563560U (zh) 一种基于线激光扫描的手机外壳质量在线检测装置
CN203310772U (zh) 一种大型钢管焊缝缺陷双域检测装置
CN101424529B (zh) 列车扫描系统
WO2012057592A2 (ko) 궤도차량을 이용한 철도시설물 감시시스템 및 방법
CN111583200A (zh) 大跨越输电线路预警系统及方法
CN104237257A (zh) 一种用于运营地铁隧道结构病害综合快速检测装置
CN102589808A (zh) 大型隧道渗水点测量方法
CN109186546A (zh) 隧道病害特征检测系统和方法
Hsieh et al. Offline Deep-learning-based Defective Track Fastener Detection and Inspection System.
CN114670899A (zh) 一种轨道检测系统用的图像采集装置
CN111208146A (zh) 一种隧道线缆检测系统及检测方法
CN208872293U (zh) 隧道病害特征检测系统
CN103434268B (zh) 用于印刷设备的检测系统

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20873699

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 17767437

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2020873699

Country of ref document: EP

Effective date: 20220509

WWG Wipo information: grant in national office

Ref document number: 2020873699

Country of ref document: EP