JPH0779429A - Image monitoring device - Google Patents

Abstract

(57) [Abstract] [Purpose] To provide an image monitoring device capable of avoiding erroneous detection of a moving object when there is a change in environment such as sunshine conditions or when the device is activated. [Structure] A moving object extraction unit 13 for extracting a moving object by taking a time-series difference between a plurality of images captured in a captured image memory 11, a background image captured in a background image memory 15, and a predetermined image in the captured image memory 11. It has a moving object extraction unit 17 that extracts a moving object by taking a background difference from an image captured after a lapse of time, and when the apparatus is activated, the moving object extraction unit 13 is selected and this moving object extraction unit 13 When the moving object is no longer extracted, the contents of the background image memory 15 are updated to the image captured at that time, and the moving object extraction unit 17
Select. When the moving object extraction unit 17 no longer extracts the moving object, the contents of the background image memory 15 are updated with the image captured at that time.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image monitoring apparatus for detecting a moving object within a monitoring area.

[0002]

2. Description of the Related Art As an image monitoring apparatus of this type, for example,
The following are known. That is, the image in the monitoring area is captured by the ITV camera and converted into an electric signal. The image signal from the ITV camera is converted into digital data and then sequentially input to the image memory. In this way, a change area is cut out by performing difference calculation and binarization calculation processing on a plurality of images input in time series, and a plurality of images input in time series are extracted. The change area is cut out by performing difference calculation and binarization calculation processing with the image captured as the background image. Then, a rectangle circumscribing this change area is obtained, and if the size of the rectangle is larger than a certain value (threshold value of the number of pixels), it is determined that a moving object is detected, and an alarm device or the like is operated. It is a thing.

[0003]

In the conventional image monitoring apparatus described above, a background image is created in advance as a reference image, and the difference between the image captured at a fixed time interval (background update timing) and the background image ( The moving object is detected from the background difference. In this case, there are the following background image update methods. That is, if a moving object is not detected based on the background difference between the image captured according to the background update timing and the background image, the image captured at the background update timing is updated as the background image. A second background update method that updates the weighted average of consecutive images captured at the update timing as a background image, and the weighting of consecutive images captured at the background update timing unless a moving object is detected by the background difference. A third background updating method has been proposed in which the average is updated as a background image.

However, in the first background updating method, if the sunshine changes when a moving object is detected,
Since the background screen is not updated, there is a problem that an abnormality is always detected after the sunshine changes. Also,
In the second background updating method, when there is a moving object in the image captured at the background updating timing, after this image is updated as the background image, the moving object is detected due to the influence of the moving object in the background image. There was a problem that the accuracy was lowered. Furthermore, the third background updating method solves the problem of the second background updating method, but if there is a change in the sunshine while a moving object is being detected, the image before the sunshine change is displayed. Since the background image is the weighted average of the images after the change of the sunshine, there is a problem that the actual background and the background image are slightly different from each other, so that the false detection of the moving object is likely to occur.

If the background image contains a moving object, the result of the background difference is an abnormal state in which the moving object is always detected. Therefore, the background image should not contain the moving object. In particular, it cannot be guaranteed that there is no moving object in the image captured as the initial background image by the first, second, and third background updating methods when the device is activated. That is, in the first and third background updating methods, the moving object may be always detected and the background image may not be updated. In the second background updating method, the moving object may be continuously detected for a while at the time of starting the device. There is. Therefore,
There is a problem that the device must be started when there is no moving object, or the background image must be captured by the operation of the operator.

Therefore, an object of the present invention is to provide an image monitoring apparatus capable of avoiding erroneous detection of a moving object when there is a change in environment such as sunshine conditions or when the apparatus is activated.

[0007]

SUMMARY OF THE INVENTION An image monitoring apparatus of the present invention is an image pickup means for picking up an image in a monitoring area, and an image picked up by this image pickup means is continuously fetched at predetermined time intervals and digitized. And a converting means for continuously converting the digitized image by the converting means, and performing a predetermined arithmetic process between the plurality of acquired images to extract a moving object in the monitoring area. A moving object extracting means, a background image in the monitoring area digitized by the converting means, and a moving object image including the background image are respectively captured, and predetermined arithmetic processing is performed between the both captured images. Thus, the second moving object extracting means for extracting the moving object in the monitoring area and the first moving object extracting means at the time of starting the apparatus are selected. When the moving object in the moving object extraction means detects that it is no longer being extracted, and updates the background image at that time captured image, the second
When it is detected that the moving object is no longer extracted by the first control means for selecting the moving object extraction means and the second moving object extraction means selected by the first control means,
The second that updates the background image to the image captured at that time
And the control means.

Further, the image monitoring apparatus of the present invention comprises an image pickup means for picking up an image in the monitoring area, and a conversion means for continuously fetching the images picked up by the image pickup means at predetermined time intervals and digitizing them. First moving object extracting means for extracting a moving object in the monitoring area by successively capturing images digitized by the converting means and performing a predetermined calculation process between the captured plurality of images. When,
The background image in the monitoring area digitized by the converting means and the moving object image including the background image are respectively captured, and a predetermined calculation process is performed between the captured images, whereby the monitoring area Second moving object extracting means for extracting a moving object in, the selecting means for selecting one of the first moving object extracting means and the second moving object extracting means, and the selecting means for selecting the moving object. When the second moving object extraction means is selected,
If it is determined that the background image is not suitable, the first control means for controlling the selecting means to select the first moving object extracting means, and the first movement selected by the first controlling means. When the object extracting unit detects that the moving object is no longer extracted, the background image is updated to the image captured at that time, and the selecting unit is controlled to select the second moving object extracting unit. And the control means.

Further, the image monitoring apparatus of the present invention comprises an image pickup means for picking up an image in the monitoring area, and a conversion means for continuously fetching the images picked up by the image pickup means at predetermined time intervals and digitizing them. First moving object extracting means for extracting a moving object in the monitoring area by successively capturing images digitized by the converting means and performing a predetermined calculation process between the captured plurality of images. When,
The background image in the monitoring area digitized by the converting means and the moving object image including the background image are respectively captured, and a predetermined calculation process is performed between the captured images, whereby the monitoring area Second moving object extracting means for extracting the moving object in the above, and the first moving object extracting means and the second moving object extracting means detect that the moving object is not continuously extracted for a predetermined time or longer. And the first moving object extracting means when the apparatus is started, and the moving object is not continuously extracted for a predetermined time or more by the selected first moving object extracting means. When detected, the background image is updated to the image captured at that time, and the second image is updated.
The detecting means means that the moving object is not continuously extracted for a predetermined time or longer by the first control means for selecting the moving object extracting means and the second moving object extracting means selected by the first controlling means. And a second control means for updating the background image with the image captured at that time.

Further, the image monitoring apparatus of the present invention comprises an image pickup means for picking up an image in the monitoring area, and a conversion means for continuously taking in the images picked up by the image pickup means at predetermined time intervals and digitizing them. First moving object extracting means for extracting a moving object in the monitoring area by successively capturing images digitized by the converting means and performing a predetermined calculation process between the captured plurality of images. When,
The background image in the monitoring area digitized by the converting means and the moving object image including the background image are respectively captured, and a predetermined calculation process is performed between the captured images, whereby the monitoring area Second moving object extracting means for extracting a moving object in, the selecting means for selecting one of the first moving object extracting means and the second moving object extracting means, and the selecting means. Storage means for storing the moving object extracted by the first or second moving object extracting means.

[0011]

When the apparatus is activated, the moving object is detected by the difference (time-series difference) between images continuously captured at relatively short time intervals, and when the moving object is no longer detected by the time-series difference. By switching to detection of the moving object based on the background difference, it is possible to avoid erroneous detection of the moving object due to the presence of the moving object when the apparatus is started and the background image includes the moving object.

Further, when there is a change in the sunshine condition or the like during the detection of the moving object by the background difference, the mode is switched to the detection of the moving object by the time series difference, and the background difference is detected when the moving object is no longer detected by the time series difference. By switching to the detection of the moving object by the method, erroneous detection of the moving object after the environmental change can be avoided when there is a change in the environment such as the sunshine condition.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 2 shows an image monitoring apparatus according to an embodiment of the present invention. That is, for example, the ITV camera 1 as an image capturing unit captures an image in the monitoring area 6 and converts it into an electric signal. The image signal captured by the ITV camera 1 is sent to the processing device 3 and the display device 4 via the transmission line 2. The display device 4 displays the images captured by the ITV camera 1, and the processing device 3 continuously captures the images captured by the ITV camera 1 and performs image processing and determination for detecting a moving object. As a result, when a moving object is detected, moving object detection is displayed on the screen of the display device 4, and an alarm sound is emitted by the alarm device 5.

FIG. 1 schematically shows a main part of the above-mentioned image monitoring apparatus. The same parts as those in FIG. 2 will be described with the same reference numerals. In the figure, the image signal from the ITV camera 1 is converted into a digital signal by the A / D converter 10 at regular time intervals, and is captured in the captured image memory 11 as image data. The image data continuously captured at the constant time intervals is sent to the changing area extracting unit 12 and also to the changing area extracting unit 16.

The change area extraction unit 12 extracts the change area of the image by taking a time-series difference between the respective image data in the order in which they are taken. That is, as shown in FIGS. 3 and 4, image data captured at time t = t _i-1 (see FIG. 3a)
And image data captured at time t = t _i (FIG. 3b).
And the difference image (see FIG. 4a) is calculated for each pixel, and the obtained difference image (see FIG. 4a) is set to a certain value (threshold value).
The change region 30 is extracted as a binarized image (see FIG. 4b) by binarizing with.

FIG. 5 and FIG. 6 show this state by the brightness value of each pixel. In FIG. 5A, time t = t _i-1.
FIG. 5B shows the image data captured at time t = t _i by the brightness value in pixel units. 5 (a) and FIG. 5 (b), a difference such as luminance is calculated in pixel units to generate difference data (see FIG. 6a), and a pixel whose difference data is equal to or more than a predetermined threshold value is “1”, Pixels below a predetermined threshold are binarized as "0", and as a result, differential binarized data (see FIG. 6b) in which the pixels in the change region are detected as "1" are obtained.

The moving object extracting unit 13 is a changing area extracting unit 1.
A rectangle circumscribing the change area extracted in 2 is obtained, and a moving object is extracted. An example thereof will be described with reference to FIGS. 7, 8 and 9. In this example, a moving object is extracted using projection. First, the change area 30 extracted as a binarized image is projected on the x axis (horizontal axis) (see FIG. 7). Next, a projection is taken on the y-axis (vertical axis) at a position where the projection value is larger than a certain fixed value (threshold value of the number of pixels) (see FIG. 8). Next, if the change area of the x and y axes is expanded for the area having a certain projection value (threshold value of the number of pixels) or more, a rectangle surrounding the change area is obtained, and the moving object 31 is obtained. (See Figure 9).

It is also possible to extract a moving object using labeling, an example of which will be described with reference to FIGS. 10, 11 and 12. This is done by assigning a serial number (labeling) to a closed area (see FIG. 10) in which the value of the change area 30 extracted as a binarized image is “1” and obtaining the area and distance of each partial area. A grouped area is extracted, and a shape, an area having a value of “1”, and the like are calculated in the area to obtain a rectangle surrounding the changed area (see FIG. 11). If the size of the rectangle thus obtained is larger than a certain fixed value (a threshold value of the number of pixels), it means that the moving object 31 is extracted (see FIG. 12).

The background update control unit 14 controls the image data sent from the A / D converter 10 as background image data into the background image memory 15 in accordance with the preset background update timing. However, when the signal indicating that the moving object is extracted is received from the moving object extraction unit 13, the background update control is not performed.

The background image memory 15 includes a background update control unit 1
Under the control of No. 4, the image data sent from the A / D converter 10 is taken in as background image data. The changed area extraction unit 16 obtains a background difference between the background image data captured in the background image memory 15 and the image data newly captured in the captured image memory 11 for a predetermined time to determine the changed area of the image. Extract. That is, as shown in FIGS. 13 and 14, a background image (FIG. 13a) stored in the background image memory 15 and image data (FIG. 13b) captured in the captured image memory 11 at time t = t _i are displayed. , The difference image such as the brightness is calculated for each pixel, and the obtained difference image (FIG. 14A) is binarized by a certain constant value (threshold value) to change the region 30 into a difference binary image (FIG. 14B). ).

The moving object extraction unit 17 includes a changed area extraction unit 1
A rectangle circumscribing the change area extracted in 6 is obtained, and a moving object is extracted. The extraction method is the same as that of the moving object extraction unit 13 described above, and thus the description thereof is omitted.

The background confirmation unit 18 includes information on the changing area extracted by the background difference in the changing area extracting unit 16, information on the moving object extracted by the moving object extracting unit 13 by the time series difference, and moving object extracting unit 17. The detection of a change in sunshine conditions and the like, the detection of an object that has entered the monitoring area and stopped halfway (stopped object), etc. are performed from the information of the moving object extracted by the background difference.

That is, when the sunshine condition changes, the brightness value of the captured image generally rises and falls. Therefore, when background subtraction is performed, a changed area appears on the entire screen. Therefore,
The number of pixels in the change area is counted by the change area extraction unit 16 based on the background difference, and if the count value exceeds a predetermined value, it is considered that the sunshine condition has changed. If the moving object extracting unit 13 based on the time series difference does not extract the moving object and the moving object extracting unit 17 based on the background difference extracts the moving object, the object is regarded as a stationary object, and the background checking unit 1
The position and size of the stationary object are stored in a stationary object storage area (not shown) in FIG.

The moving object selecting section 19 receives a signal from the background checking section 18 indicating that a change in sunshine conditions or the like has been detected or a stationary object has been extracted, and when the apparatus is activated, When the moving object extracted by the extracting unit 13 is selected and the moving object extracting unit 13 detects that the moving object is no longer extracted, the moving object extracted by the moving object extracting unit 17 is selected, and the selected moving object is selected. The moving object storage unit 20 determines that a moving object that should originally be detected is detected.
And the alarm device 5 is started.

The display control unit 21 uses the captured image memory 11
When a signal indicating that a moving object to be originally detected is detected from the moving object selection unit 19 while controlling the display device 4 to display the image captured by the captured image memory 11, the captured image is stored in the captured image memory 11. The image is displayed on the display device 4, and a figure, a character or the like for notifying the detection of the moving object is displayed.

A specific example of updating the background image by the background update control unit 14 will be described with reference to FIGS. 15, 16 and 17. The input image (FIG. 1) at the time of starting the apparatus (t = t ₀ ).
5a), the moving object 30 is included in the image.
After a while (t = t _i ) after activation of the present apparatus, if the moving object 30 disappears from the monitoring area in the image (see FIG. 15b) input according to the background update timing, t = t. _The image input to _i is captured as a background screen in the background screen memory 15. After capturing the background image at t = t _i , the background screen is updated at the background update timing of t = t _j−1 , and the image input at the background update timing of t = t _j (see FIG. 16b). Again contained the moving object 30. Therefore, the background image updated at t = t _j-1 (see FIG. 16a) is not updated at t = t _j . afterwards,
The image input at the background update timing of t = t _k-1 (see FIG.
7a), the moving object 30 is included, so t = t
_{With k-1} , the background screen is not updated. Since the moving object 30 is not included in the image input at the next background update timing of t = t _k (see FIG. 17b), the image input at t = t _k is taken into the background screen memory 15 as the background screen. .

Next, the processing for detecting a moving object will be described in detail with reference to the flowcharts shown in FIGS. In the detection processing described below, the moving object extraction unit 17
A flag to be set when selecting the moving object extracted in step S1 is set in the moving object selection unit 19 and is called a background flag. When the moving object is detected by the moving object extracted by the moving object extracting unit 13, the background flag is set when the moving object is not extracted M times consecutively. At this time, the number of times the moving object is not continuously extracted is counted by, for example, the counter N of the software provided in the moving object selection unit 19.

Further, when the moving object is detected by the moving object extracted by the moving object extraction unit 17, if the moving object is not extracted K times consecutively, the background image is loaded into the background image memory 15. I will. At this time, the number of times the moving object is not continuously extracted is, for example, the counter J of the software provided in the background update control unit 14.
It is supposed to be counted by.

First, when the apparatus is started, the values of M and K are set, and the background flag and the counters N and J are reset (step S1). Next, the image is taken from the ITV camera 1, converted into a digital signal by the A / D converter 10, and taken into the taken-in image memory 11 as an initial image, and the background update control unit 14 causes the initial image of the background image to be stored in the background image memory. 15 (step S2).

Then, the changing area extracting unit 12 extracts a changing area of the image by obtaining a time-series difference between the respective image data in the order of being taken into the taken-in image memory 11. At the same time, in the changed area extraction unit 16, the background image data captured in the background image memory 15 and the captured image memory 11 are stored.
After a predetermined time, the background difference between the newly acquired image data and the newly acquired image data is calculated, and the changed area of the image is extracted. The moving object extracting unit 13 extracts a moving object by obtaining a rectangle circumscribing the changing region extracted by the changing region extracting unit 12 (step S3). At the same time, the moving object extracting unit 17 also extracts a moving object by obtaining a rectangle circumscribing the changing region extracted by the changing region extracting unit 16 (step S4).

Next, in step S5, it is checked whether or not the background flag is set. If the background flag is reset, the moving object is detected by the moving object extraction unit 13. Therefore, the process proceeds to step S6, and it is checked whether or not the moving object is extracted by the moving object extraction unit 13. If a moving object is extracted as a result of this check, the process proceeds to step S7, and the detected moving object is stored in the moving object storage unit 20.
The counter N is reset in step S8.

In step S6, if the changing object is not extracted due to the time series difference, the process proceeds to step S9, the counter N is incremented, and further, in step S10, it is determined whether or not the value of the counter N is smaller than the set value M. Check that. As a result of this check, if the value of the counter N is smaller than the set value M, steps S2 to S6 are repeated.

If the moving object is not extracted again, the counter N is incremented (step S
9) Since the counter N is reset when the moving object is extracted (step S8), the moving object is detected by the moving object extracted by the time-series difference until the moving object is not extracted M times in succession. To do.

When the value of the counter N becomes equal to the set value M in step S10 and the moving object is no longer extracted M times in succession, the captured image is captured as the background image in the background image memory 15 (step S11). , Counter N
Is reset (step S12) and the background flag is set (step S13).

After that, while the background flag is set, the moving object is detected by the moving object extracted by the background difference. That is, as a result of the check in step S5, it is checked whether or not a moving object is extracted.
As a result of this check, if the background flag is set, the process proceeds to step S14. If the moving object extraction unit 17 does not extract the moving object, the counter J is incremented (step S15) and the value of the counter J is set. Value K
If smaller than (step S16), step S2
~ Step S5 and step S14 are repeated. And
When the moving object is not extracted again, the counter J is incremented (step S15), but when the moving object is extracted, the counter J is reset (step S15).
19), the background image is not updated until the moving object is no longer extracted K times in succession. When the moving object is no longer extracted K times consecutively (step S16), the captured image is captured as the background image in the background image memory 15,
The background image is updated (step S17), and the counter J
Is reset (step S18).

Next, when a moving object is extracted by the background difference (step S14), the counter J is first reset (step S19). Next, when a change in the sunshine condition is detected (step S20), the background flag is reset (step S26), and thereafter, the detection of the moving object is extracted by the time series difference until the background flag is set again. It is performed by a moving object (steps S2 to S13).

If no change in the sunshine conditions is detected in step S20, the process proceeds to step S21 and steps S21 to S2.
At 4, the process for the stationary object is performed. That is, the stationary object is considered to have been detected when the moving object is extracted by the background difference and when the moving object is not extracted by the time-series difference, so when the stationary object is detected for the first time, The detected position and size of the stopped object are stored in a not-shown stopped object storage area in the background confirmation unit 18 (steps S23 to S24).

After that, when the moving object is extracted by the background difference, it is checked whether or not the moving object is the same as the stopped object stored in the stopped object storage area (step S21). The stopped object storage area is cleared (step S22), the background flag is reset (step S26), and thereafter, until the background flag is set again, the detection of the moving object is detected by the time series difference. (Steps S2 to S13). If the detected moving object is not the same as the stopped object stored in the stopped object storage area, it is checked whether or not the moving object is the stopped object (step S23). The position and size of the stopped object detected in the area are stored (step S24) and stored in the moving object storage unit 20 (step S25). If the result of the check in step 23 is that the extracted moving object is not a stopped object, the process proceeds to step S25 as it is, and the extracted moving object is stored in the moving object storage unit 20.

FIG. 20 shows the switching timing of the detection result of the moving object and the updating timing of the background screen when the apparatus is started and when a change in the sunshine condition or the like is detected. First, when the apparatus is activated at time t ₀ ,
The detection result of the moving object based on the time series difference is selected. Then, when the moving object is no longer detected (time t ₁ ).
Then, the detection result of the moving object based on the time-series difference is switched to the detection result of the moving object based on the background difference, and at the same time, the image captured at time t ₁ is used as the background image.

[0040] In the image capture timing of time t _m is the moving object is not detected, the image captured at time t _m and a background image, even at time t _{m + 2,} the moving object is not detected, the time t _{m + The} image captured in ₂ is used as the background image.

When the background confirmation section 18 detects a change in the sunshine condition at time t _n , the detection result of the moving object based on the background difference is switched to the detection result of the moving object based on the time series difference.
At this time, the background image is not updated because the moving object and the change in the sunshine condition are detected.

When the moving object is no longer detected at time t _{n + 2} , the detection result of the moving object based on the time-series difference is switched to the extraction result of the moving object based on the background difference, and at the same time, it is captured at time t _{n + 2} . Let the image be the background image.

As described above, according to the above embodiment, the moving object is detected by the time-series difference when the apparatus is activated, and the moving object is detected by the background difference when the moving object is no longer detected by the time-series difference. By switching to detection of an object and updating the background image, it is possible to avoid erroneous detection of the moving object due to the presence of the moving object when the apparatus is started and the background image includes the moving object.

Further, when a moving object is detected by the background difference, it is judged whether or not there is a change in the sunshine condition or a stationary object is detected from the detected moving object, and the mode is switched to the detection of the moving object by the time series difference. By switching to the detection of the moving object based on the background difference when the moving object is not detected by the difference, erroneous detection of the moving object after the change of the sunshine can be avoided when the sunshine condition or the like changes.

In the moving object detection processing method in the above embodiment, the moving object extraction processing based on the time series difference and the moving object extraction processing based on the background difference are simultaneously performed, and either one of the extraction processing is performed. I had selected processing,
The present invention is not limited to this, and either one of the moving object extraction processing by the time series difference or the moving object extraction processing by the background difference is selected, and only the selected extraction processing is operated. There is no need to explain what is good.

[0046]

As described in detail above, according to the present invention, when there is a change in environment such as sunshine conditions or when the apparatus is started,
An image monitoring device capable of avoiding erroneous detection of a moving object can be provided.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a main part of an image monitoring apparatus according to an embodiment of the present invention.

FIG. 2 is a diagram schematically showing an image monitoring device according to an embodiment of the present invention.

FIG. 3 is a diagram illustrating a method of extracting a change area based on a time series difference.

FIG. 4 is a diagram illustrating a method of extracting a change region based on a time series difference.

FIG. 5 is a diagram illustrating a method of extracting a change area based on a time series difference.

FIG. 6 is a diagram illustrating a method of extracting a change area based on a time series difference.

FIG. 7 is a diagram illustrating a method of extracting a moving object.

FIG. 8 is a diagram illustrating a method of extracting a moving object.

FIG. 9 is a diagram illustrating a method of extracting a moving object.

FIG. 10 is a diagram illustrating a method of extracting a moving object.

FIG. 11 is a diagram illustrating a method of extracting a moving object.

FIG. 12 is a diagram illustrating a method of extracting a moving object.

FIG. 13 is a diagram illustrating a method of extracting a change area based on a background difference.

FIG. 14 is a diagram illustrating a method of extracting a change area based on a background difference.

FIG. 15 is a diagram illustrating a method of updating a background image.

FIG. 16 is a diagram illustrating a background image updating method.

FIG. 17 is a diagram illustrating a background image updating method.

FIG. 18 is a flowchart illustrating a moving object detection process.

FIG. 19 is a flowchart illustrating a moving object detection process.

FIG. 20 is a timing chart showing the switching timing of the detection result of the moving object and the update timing of the background screen.

[Explanation of symbols]

1 ... ITV camera (imaging means), 2 ... Transmission path, 3 ... Processing device, 4 ... Display device, 5 ... Alarm device, 6 ... Monitoring area, 1
0 ... A / D converter, 11 ... Image memory, 12 ... Change area extraction section (time series difference), 13 ... Moving object extraction section (time series difference), 14 ... Background update control section, 15 ... Background image memory, 16 ... Change area extraction part (background difference), 17 ... Moving object extraction part (background difference), 18 ... Background confirmation part, 19 ... Moving object selection part, 20 ... Moving object storage part, 21 ... Display control part, 30 ... Change area, 31 ... Moving object.

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical display location G06T 7/20 G08B 25/00 510 M 7323-5G

Claims (4)

[Claims] 1. An image pickup means for picking up an image in a monitoring area, a conversion means for continuously picking up an image picked up by the image pickup means at a predetermined time interval, and digitizing the image, and a digitalizing means for the converting means. First moving object extracting means for extracting a moving object in the monitoring area by performing a predetermined arithmetic processing between the plurality of acquired images, and digitizing by the converting means. The captured background image in the monitoring area and the moving object image including the background image are respectively captured, and a predetermined arithmetic process is performed between the captured images to extract the moving object in the monitoring area. The second moving object extracting means and the first moving object extracting means are selected when the apparatus is activated, and the moving object is extracted by the selected first moving object extracting means. When it is detected that it is no longer present, the background image is updated to the image captured at that time, and the first control means for selecting the second moving object extraction means and the first control means are selected. When the second moving object extracting unit detects that the moving object is no longer extracted, the second moving image extracting unit updates the background image with the image captured at that time. apparatus. 2. An image pickup means for picking up an image in the monitoring area, a converting means for continuously taking in an image picked up by the image pickup means at a predetermined time interval and digitizing the image, and a digitalizing means for the converting means. First moving object extracting means for extracting a moving object in the monitoring area by performing a predetermined arithmetic processing between the plurality of acquired images, and digitizing by the converting means. The captured background image in the monitoring area and the moving object image including the background image are respectively captured, and a predetermined arithmetic process is performed between the captured images to extract the moving object in the monitoring area. A second moving object extracting means; a selecting means for selecting one of the first moving object extracting means and the second moving object extracting means; When the second mobile object extraction means is selected in the unit, when the background image is determined not to be appropriate,
A first control unit that controls the selection unit to select the first moving object extraction unit, and a moving object is not extracted by the first moving object extraction unit selected by the first control unit. And a second control means for controlling the selecting means so as to select the second moving object extracting means while updating the background image to the image captured at that time. Characteristic image monitoring device. 3. An image pickup means for picking up an image in the monitoring area, a converting means for continuously taking in an image picked up by the image pickup means at a predetermined time interval and digitizing the image, and a digitalizing means for the converting means. First moving object extracting means for extracting a moving object in the monitoring area by performing a predetermined arithmetic processing between the plurality of acquired images, and digitizing by the converting means. The captured background image in the monitoring area and the moving object image including the background image are respectively captured, and a predetermined arithmetic process is performed between the captured images to extract the moving object in the monitoring area. The moving object has not been continuously extracted for a predetermined time or longer by the second moving object extraction means and the first moving object extraction means and the second moving object extraction means. The first moving object extracting means is selected when the apparatus is activated, and the moving object is not continuously extracted for a predetermined time or longer by the selected first moving object extracting means. When detected by the detection means, the background image is updated to the image captured at that time, and the first control means for selecting the second moving object extraction means and the first control means are selected. Second control means for updating the background image to the image captured at that time when the detection means detects that the moving object is not continuously extracted for a predetermined time or longer by the second moving object extraction means, An image monitoring apparatus comprising: 4. An image pickup means for picking up an image in a monitoring area, a converting means for continuously taking in an image picked up by the image pickup means at a predetermined time interval and digitizing the image, and a digitalizing means for the converting means. First moving object extracting means for extracting a moving object in the monitoring area by performing a predetermined arithmetic processing between the plurality of acquired images, and digitizing by the converting means. The captured background image in the monitoring area and the moving object image including the background image are respectively captured, and a predetermined arithmetic process is performed between the captured images to extract the moving object in the monitoring area. A second moving object extracting means; a selecting means for selecting one of the first moving object extracting means and the second moving object extracting means; Image monitoring apparatus characterized by comprising storage means for storing the moving object extracted by the first or second moving object extraction means is selected by the means.