JPH0556441B2 - - Google Patents

Description

[Detailed description of the invention]

[Industrial Field of Application] The present invention relates to a method of controlling the exposure of each imaging device when combining the fields of view of a plurality of imaging devices, such as a television camera, to obtain one composite field of view, and more specifically, For example, when measuring the shape of a hot-rolled thick plate by capturing a composite image using multiple imaging devices, exposure is individually controlled depending on the level of the video signal of each imaging device. This paper proposes an exposure control method for multiple imaging devices to capture only the images necessary for measurement without clearly capturing backgrounds that are unnecessary for measurement. . [Prior art] A method of joining multiple images to obtain one composite image is a method often used, for example, in aerial surveying or remote sensing using artificial satellites. The exposure when the image is taken is generally configured so that the amount of light received by the imaging device at the time the image is taken is always constant depending on the characteristics of the imaging device. Therefore, in such a composite image, differences in image density, differences in tone, etc. are often clearly visible at the joints. By the way, the inventors of the present application have proposed a "plate shape measuring device" (Japanese Patent Application No. 147248/1982) that uses a composite image obtained by a plurality of imaging devices as described above. Specifically, nine television cameras arranged in a 3 x 3 matrix take self-luminous near-infrared images of a hot-rolled thick plate and create one composite image. Image processing is performed to detect the position of each edge of the plank, which ultimately measures the shape of the plank (planar shape, more specifically external dimensions and linearity), and shape measurement. Based on a plurality of two-dimensional imaging devices arranged so as to image each part of the target board, image signals obtained by the imaging devices, and signals obtained by changing the level of the image signal and delaying the image signal. a slicer that creates a signal indicating the rising and falling points of an image signal, and the rising edge and the rising edge based on the output signal of the slicer
The present invention includes an edge position detection circuit that associates a falling point with a position of a captured image, and an arithmetic unit that associates and synthesizes position information obtained by the edge position detection circuit with the imaging field of view of each two-dimensional imaging device. Features. However, the exposure control, specifically the aperture control method, of the 3x3 matrix-arranged television camera used by the inventors in the above-mentioned "plate shape measuring device" is different for each imaging device as described above. Since the conventional method of controlling exposure according to the amount of light received was followed, the following problems existed. Exposure control, ie, aperture control, for each of the nine television cameras is to appropriately open and close the aperture so that the peak level of the imaged video signal is constant. For this reason, when capturing a near-infrared image of a hot-rolled thick plate on a relatively low-temperature background, the hot-rolled thick plate has a high imaging signal level (usually expressed as white in television images). ), so the aperture of the television camera is narrowed down so that the image of the plate being hot-rolled is within its field of view, and conversely, the aperture of the television camera is that the image of the plate being hot-rolled is not within its field of view. is close to full throttle. In a TV camera where the aperture is close to fully open, rolling line components such as rollers and side guides reflect near-infrared rays from the thick plate being hot-rolled, and the image is captured as a near-infrared image. Therefore, when such image signals are analyzed, the image of the roller or side guide may be mistakenly recognized as the image of the thick plate being hot rolled, which is the actual object of measurement, resulting in erroneous measurements. become. Therefore, in order to avoid the above-mentioned erroneous measurements, it is necessary to remove images that are not the actual measurement target through software processing when analyzing the image signal, and then perform the original analysis processing to determine whether the image signal is being measured during hot rolling. Because we were measuring the shape of a thick plate, it took a long time to process the data, and the reality was that the software processing to remove image signals that were not part of the measurement target was not always perfect. It was hot. Therefore, for example, a method can be considered in which a television camera whose entire field of view is occupied by a thick plate being hot rolled is used as a standard and applied to other television cameras with the same aperture control. Possible problems include that a considerable amount of time is required for processing to select a television camera. [Object of the Invention] The present invention has been made in view of the above-mentioned circumstances, and is aimed at determining the appropriateness of the reference imaging device when combining images captured by a plurality of imaging devices to obtain one composite image. Measure the level of the imaging signal of each other imaging device at the same exposure as the exposure, compare these with the reference level set for each imaging device, and adjust each imaging device to its appropriate exposure according to this result. or by taking images with the same exposure as the reference imaging device, unnecessary objects other than the object to be imaged are not captured at the time of image capturing, and software-based methods are used to remove the above-mentioned unnecessary objects. The purpose of this invention is to propose an exposure control method for multiple imaging devices that eliminates the need for processing and thereby allows the original image processing to be performed more accurately in a shorter time. [Structure of the Invention] The exposure control method for multiple imaging devices according to the present invention includes:
In an exposure control method for obtaining a composite image by combining images of a bright object on a dark background, each imaged by a plurality of imaging devices having adjacent fields of view, an appropriately defined standard imaging method is used. The level of the video signal of each other imaging device is measured at the same exposure as the proper exposure of the device, and the level of the video signal of each imaging device other than the reference imaging device is determined as the standard set for each imaging device. The imaging device whose video signal level is higher than the reference level will take an image at its appropriate exposure, and the imaging device whose video signal level is lower than the reference level will be the imaging device that becomes the reference level. The feature is that the image is captured with the same exposure as . Conversely, if there is a dark object on a bright background, measure the level of the video signal of each other imaging device at the same exposure as the appropriate exposure of the standard imaging device. Then, the level of the video signal of each imaging device other than the reference imaging device is compared with the reference level set for each imaging device, and if the level of the video signal is lower than the reference level, the imaging device is An imaging device that captures images with proper exposure and whose video signal level is higher than the reference level is characterized in that it captures images with the same exposure as the reference imaging device. [Examples] The present invention will be described in detail below based on drawings showing examples thereof. FIG. 1 is a schematic diagram showing the general configuration of a plate shape measuring device using the exposure control method for multiple imaging devices according to the present invention. A reversible rolling mill M is installed in a rolling line in which a large number of rollers R, R, . . . are arranged side by side. This reversible rolling mill M rolls a hot-rolled thick steel plate S that is moved on rollers R, R, . . . in both directions indicated by the white arrows. A camera house 1 is provided above one side of the reversible rolling mill M. The camera house 1 has nine television cameras 11 arranged in a 3x3 matrix as described later.
19 (see FIG. 2), each camera 11 to 19 is controlled by a camera control section 4,
In addition, the near-infrared images of the hot-rolled thick steel plate S taken by the cameras 11 to 19 are sent to the camera control unit 4 and subjected to preliminary processing such as noise removal and image distortion correction, and then input to the arithmetic unit 3. be done. The computing device 3 analyzes the image signals of each of the nine cameras 11 to 19, detects the edge position of the hot rolled thick steel plate S within the field of view of each camera 11 to 19, and synthesizes the results to determine the hot rolled thickness. The entire shape of the steel plate S is measured. The shape of the hot rolled thick steel plate S thus measured is measured, and its characteristic elements are provided to the control device 2. The control device 2 controls the driving devices (not shown) of the reversible rolling mill M, rollers R, R, etc., via the rolling mill control device 5, based on the characteristic elements of the shape of the hot-rolled thick steel plate S given from the arithmetic device 3. ) etc., thereby controlling hot rolled thick steel plate S
The shape is rolled into a specified shape. Figure 2 shows each camera 11 to 1 in camera house 1.
9 is a schematic diagram showing the relative position of the imaging field of view 111 to 119 and the hot-rolled thick steel plate S to be imaged. In the camera house 1, nine cameras 11 to 19 are arranged in a 3×3 matrix as shown in FIG. The cameras 11 to 14 and 16 to 19 other than the camera 15 located at the center are centrally located vertically above the field of view 115 of the central camera 15. The camera 15 is located at the center and enables imaging of the tilted field of view.
The fields of view 111 to 1 are arranged around the field of view 115 of
14, 116 to 119 are set. As a result, nine cameras 11 to 12 are placed in a relatively small area compared to the size of the hot rolled thick steel plate S to be measured.
It is now possible to place 19. Then, the hot rolled thick steel plate S is attached to each camera 11 to
Composite visual field 110 from visual fields 111 to 119 of 19
, i.e., the field of view 11 of the central camera 15
The cameras 11 to 19 take an image of the hot rolled thick steel plate S, that is, measure its shape, with the cameras 11 to 19 positioned such that all of the hot rolled thick steel plate S are occupied by the hot rolled thick steel plate S. In addition, the composite field of view 11 of each camera 11 to 19
The composite field of view 11 is arranged so that the edge of 0 and the edge of the hot rolled thick steel plate S are not parallel but intersect at a slight angle.
0 is set. This is for each camera 11-19
This is because if the direction of the scanning line coincides with the direction of the edge of the hot-rolled thick steel plate S, detection becomes somewhat difficult. FIG. 3 is a schematic diagram showing the configuration of the aperture control mechanism of each camera 11-19. In the figure a, the central camera 15 is shown, and in the same figure b, the other cameras 11 to 14, 16 to
19 aperture control mechanisms are shown. Central camera 15 connects other cameras 11 to 14, 16
This is a camera that serves as a reference for the aperture control of cameras 15 to 19, and is configured to perform optimal aperture control based on the characteristics of the image sensor of the camera 15. That is, the level IL ₅ of the video signal captured by the camera 15
is given to the light reception level control circuit 15E,
The correction signal FC is applied from the light reception level control circuit 15E to the motor 15M that drives the aperture of the lens system 15l to open and close so that the level IL ₅ of this video signal becomes a predetermined value.
The correction signal FC for driving the aperture of the camera 15 to open and close is also given to the camera control section 4. Figure 3b shows other cameras 11-14, 16-19.
Although the aperture control mechanism shown in FIG. And camera 11
(or 12 to 14, 16 to 19) of the video signal level IL ₁ (or IL ₂ to IL ₄ , IL ₆ to _{IL 9} ) is determined by the comparator 11
C (or 12C to 14C, 16C to 19C), and this comparator 11C (or 12C
~14C, 16C~19C), the standard value K ₁ (or
K ₂ to _{K 4} and K ₆ to _{K 9} ) are given, respectively.
The reference value K1 _etc. is based on the electrical characteristics of each camera 11 etc.
Alternatively, they are determined as appropriate depending on the arrangement of rolling line members such as rollers and side guides within the field of view. Comparator 11C (or 12C to 14C,
The output corresponding to the comparison result of 16C to 19C) is output from the light reception level control circuit 11E (or 12E to 14E, 1
6E-19E). Light reception level control circuit 11E (or 12E to 14
E, 16E to 19E), the above-mentioned central camera 1
A correction signal FC for the aperture of No. 5 is given from the camera control section 4. Then, the above-mentioned comparator 11C (or 12C to 1
4C, 16C to 19C), the received light level control circuit 11E (or 12E to 14E, 1
Motor 11M (or 12M to 14M, 16M to 19
The correction signal FC given to camera 11
(or 12-14, 16-19) The configuration is such that the camera control unit 4 can switch between individual aperture control and the same aperture control as the central camera 15. The method of the present invention is carried out using nine television cameras installed in the plate shape measuring device for hot-rolled thick steel plate S configured as described above. The method of the present invention will be explained in accordance with a flowchart of aperture control of each camera 11 to 19 by the control section 4. First, at the time indicated by the arrow a in the time chart of FIG. It is stopped at the position occupied by the steel plate S, and aperture control of the central camera 15 is started. This is done in the lens system 15l so that the level _IL5 of the image signal captured by the camera 15 becomes a predetermined value.
The aperture is controlled to an aperture value _F5 by a correction signal FC given from the light reception level control circuit 15E. The exposure in this state is the appropriate exposure for the central camera 15, which is the reference imaging device, and is the exposure that allows the edge of the object to be imaged to be identified. Specifically, this is an exposure obtained by a known automatic exposure technique in which the image peak signal is lowered by a certain value than the saturation limit of the camera. By the way, in this case, as shown in FIG. 2, since the entire field of view 115 of the camera 15 is normally occupied by the hot rolled thick steel plate S, the level of the video signal is
IL ₅ has become considerably high, and therefore the aperture of the lens system 15l has to be narrowed down considerably, and its aperture value F ₅ has become a considerably large value. Field of view 11
In this case, the aperture of the camera 15 will be considerably opened compared to the above case, and therefore the aperture value _F5 will be a considerably small value. Also, at the point of arrow a in Fig. 4, the central camera 15
The apertures of the lens systems 11l-14l, 16l-19l of the cameras 11-14, 16-19 other than the above are controlled to the same aperture value _F5 as the aperture of the central camera 15. In this control, a correction signal FC instructing the aperture value _F5 is sent from the light receiving level control circuit 15E of the central camera 15 to the central camera 15 via the camera control unit 4.
This is performed by providing the received light level control circuits 11E to 14E and 16E to 19E of the cameras 11 to 14 and 16 to 19 other than the above. In this way, the cameras 11 to 11 other than the central camera 15
The aperture value of 14, 16 to 19 is set to F5, which is the same as that of the central camera ₁₅ , and at the point of arrow b in Fig. 4,
Level of each video signal IL ₁ ~ IL ₄ , IL ₆ ~ _{IL 9}
are applied to comparators 11C to 14C and 16C to 19C, respectively. At this time, as shown in FIG.
Since the proportion of the hot rolled thick steel plate S in the
IL ₁ and IL ₃ are considerably low values. And each of these cameras 11-14, 16-
The levels IL ₁ -IL ₄ , IL ₆ -IL ₉ of the 19 image signals are compared with respective reference values K ₁ -K ₄ , K ₆ -K ₉ .
Each of the reference values K ₁ to K ₄ and K ₆ to _{K 9} is based on the characteristics of the cameras 11 to 14 and 16 to 19 to which they are set,
Or the field of view 111-114, 116-11
For example, if the level IL ₁ of the video signal is lower than the reference value K ₁ of the camera 11 due to the state of the background within the camera 9, that is, the number and arrangement of rollers, side guides, etc., the individual aperture control by the camera 11 is performed. , the respective reference values K ₁ to K ₄ , K ₆ to values that allow unnecessary objects such as rollers and side guides to be clearly imaged.
_K9 is prescribed. Each camera 11 to 14, 1 other than the central camera 15
Actual aperture control in steps 6 to 19 is performed as follows. Levels of video signals IL ₁ to _{IL 4} , IL ₆ to IL 9 of each camera 11 to 14, 16 to ₁₉ and their respective reference values K ₁
~ _K4 , _K6 ~ _K9 are each comparison circuit 11C~14C,
16C to 19C, and the video signal level IL ₁ (or IL ₂ to IL ₄ , IL ₆ to _{IL 9} ) is equal to or higher than the reference value K ₁ (or K ₂ to K ₄ , K ₆ to _{K 9} ) 11 (or 12 to 14, 16 to 19) are respectively individually connected to light receiving level control circuits 11E (or 12E to 14E,
16E to 19E), the aperture is controlled, and conversely, the video signal level IL ₁ (or IL ₂ to IL ₄ , IL ₆ to
Camera 11 (or 12-14, 16-1) whose _IL9 ) is smaller than the reference value _K1 (or _K2 - _K4 , _K6 - _K9 )
9), the aperture value is controlled to be the same as the aperture value _F5 of the central camera 15 given via the camera control unit 14, and actual imaging is performed. Note that the central camera 15 has its own light receiving level control circuit 15E.
Imaging is performed using aperture control. Therefore, in a state as shown in FIG. 2, that is, in a state where the hot rolled thick steel plate S occupies the entire field of view 115 of the central camera 15, the proportion of the hot rolled thick steel plate S occupying the field of view 111, 113 is very large. small camera 1
When imaging is performed with its own individual aperture control such as Nos. 1 and 13, it is possible to measure not only the hot-rolled thick steel plate S that is the object of imaging as described above, but also the shape of the side guides, rollers, etc. However, in the method of the present invention, the aperture values of the cameras 11 to 14 and 16 to 19 are set such that even unnecessary objects such as side guides and rollers are clearly imaged. If the control is made to be small to the extent that there is a possibility that the image will be captured, the imaging is performed at the same aperture value as the aperture value _F5 of the central camera 15 whose entire field of view is occupied by the hot rolled thick steel plate S that is the object of imaging. Therefore, although it is inappropriate for the entire field of view, it is suitable for the hot-rolled thick steel plate S that is the object of imaging.In other words, although the hot-rolled thick steel plate S is clearly imaged, The aperture value is such that unnecessary objects such as guides are hardly imaged. The aperture control of each of the cameras 11 to 14 and 16 to 19 is completed at the point indicated by the arrow c in Fig. 4, and the imaging of the hot-rolled thick steel plate S and the shape measurement by image processing are performed as indicated by the arrow d in Fig. 4. It is done by the time of the mark. This makes it possible to remove unnecessary objects such as rollers and side guides from the captured image signal when measuring the shape of hot rolled thick steel plate S.

〔effect〕

As detailed above, according to the method of the present invention, the imaging results of the imaging target by the plurality of imaging devices are kept almost constant, regardless of the proportion of the field of view occupied by the imaging target whose brightness is considerably different from the background. At the same time, it is also possible to almost erase the background, so when combining images captured by these multiple imaging devices and analyzing them, it is necessary to erase the background using software processing. Since this is no longer necessary, the time for data processing is reduced and its accuracy is improved. In the above embodiment, a case was described in which a bright object on a dark background, such as a self-luminous infrared image of a hot-rolled thick steel plate, was imaged, but in the opposite case,
That is, the present invention is also applicable to the case of a dark object to be imaged on a bright background, and in this case, the comparison result between the level of the video signal of each imaging device and the reference value may be reversed. Moreover, it is of course applicable not only to infrared images but also to ordinary visible light images.

[Brief explanation of the drawing]

Fig. 1 is a schematic diagram showing the configuration of a plate shape measuring device for hot-rolled thick steel plates, which is an example of the object of the method of the present invention, and Fig. 2 shows the nine television cameras, their fields of view, and the objects to be imaged. FIG. 3A is a block diagram showing the aperture control mechanism of the central camera serving as a reference; FIG. 3B is a block diagram showing the aperture control mechanism of other cameras; FIG.
The figure is a timing chart of aperture control. Figure 5 is a flowchart of aperture control of each television camera by the camera control unit. Figure 6 is an explanatory diagram showing the relationship between the field of view position of the central camera and the measurement position of a hot rolled thick steel plate. It is. 1...Camera house 11-19...TV camera 11C-14C, 16C-19C...Comparison circuit 11E-19E...Light reception level control circuit
11l~19l... Lens system 111~119...
...TV camera field of view S...Hot-rolled thick steel plate.

Claims (1)

[Scope of Claims] 1. In an exposure control method when a composite image is obtained by combining images each having a bright object on a dark background, each imaged by a plurality of imaging devices having adjacent fields of view, the method comprises: The level of the video signal of each other imaging device is measured at the same exposure as the determined proper exposure of the reference imaging device, and the level of the video signal of each imaging device other than the reference imaging device is measured for each imaging device. Compared with the reference level set for each device, the imaging device whose image signal level is higher than the reference level will take an image at the appropriate exposure, and the imaging device whose video signal level is lower than the reference level. An exposure control method for a plurality of imaging devices, characterized in that images are taken with the same exposure as the reference imaging device. 2. In the exposure control method for obtaining a composite image by combining images of a dark object on a bright background, each captured by a plurality of imaging devices having adjacent fields of view, the standard shall be determined as appropriate. The level of the video signal of each other imaging device is measured at the same exposure as the proper exposure of the imaging device, and the level of the video signal of each imaging device other than the reference imaging device is set for each imaging device. When compared with the reference level, the imaging devices whose video signal level is lower than the reference level take images at their respective appropriate exposures, and the imaging devices whose video signal level is higher than the reference level take images that become the reference level. An exposure control method for multiple imaging devices, characterized in that images are taken with the same exposure as the devices.