JP2016134804A - Imaging range abnormality discrimination device, imaging range abnormality discrimination method and computer program for imaging range abnormality discrimination - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an imaging range abnormality discrimination device capable of appropriately discriminating whether an abnormality that at least a part of an imaging range of an imaging part is constantly shielded is occurring or not.SOLUTION: An imaging range abnormality discrimination device comprises: a storage part 32 for storing a reference image representing an imaging range of an imaging part 2 in a state where the imaging range is not shielded constantly; a statistic pixel generation part 332 for calculating a statistical representative value of luminance for each of a plurality of images, at a same position, obtained by imaging the imaging range for a predetermined period by the imaging part 2, and generating a statistic image with the statistical representative value of luminance of each of the pixels as luminance of the corresponding pixel; and an abnormality discrimination part 333 for comparing the statistic image with the reference image to discriminate whether the abnormality that at least a part of the imaging range of the imaging part is constantly shielded is occurring or not.SELECTED DRAWING: Figure 3

Description

  The present invention relates to an imaging range abnormality determination device, an imaging range abnormality determination method, and an imaging range abnormality determination computer program that determine whether an abnormality has occurred in an imaging range of an imaging unit, for example.

  A surveillance camera may be installed indoors or outdoors for traffic monitoring, intruder monitoring, disaster monitoring, and the like. When the camera is installed outdoors, over time, the camera lens may become dirty or the roadside trees may grow, so that the shooting range of the camera is permanently blocked, and the camera may not perform its monitoring function sufficiently. There is. For this reason, when the shooting range of the camera is constantly blocked, it is desirable to immediately notify the system administrator of this fact and quickly remove the cause of blocking the shooting range of the camera.

  For example, there has been proposed a vehicle monitoring system that assumes an image when an original monitoring target image is obstructed due to the presence of an attachment on the imaging system lens surface or the like as an abnormal state image (for example, a patent) Reference 1). In an image with attached objects, the amount of texture in the screen is less than normal and the amount of texture movement is small, while in normal images, the amount of texture is large, and background objects and movements in the texture are large. The texture on the object has the property of moving. According to this characteristic, the vehicle monitoring system disclosed in Patent Literature 1 detects an abnormal state by evaluating a pixel number ratio of moving pixels among pixels having a texture in the in-screen monitoring target area.

JP 2003-312408 A

  However, in the vehicle monitoring system disclosed in Patent Document 1, when the monitoring camera is fixedly installed outdoors, if there is no moving object within the shooting range of the camera, the imaging system lens surface or the like is used. Even if deposits are present, the texture does not change. For this reason, the vehicle monitoring system disclosed in Patent Document 1 cannot properly detect an abnormal state due to the presence of an adhering substance when the monitoring camera is fixedly installed.

  In view of this, the present specification provides, as one aspect, an imaging range abnormality determination device that can appropriately determine whether or not an abnormality has occurred in which at least a part of the imaging range of the imaging unit is constantly blocked. For the purpose.

  According to one embodiment, an imaging range abnormality determination device is provided. The imaging range abnormality determination device includes a storage unit that stores a reference image representing a shooting range in a state where the shooting range of the shooting unit is not obstructed constantly, and the shooting unit captures the shooting range over a predetermined period. Statistical image generation for obtaining a statistical representative value of luminance for each pixel at the same position in a plurality of images obtained by the above, and generating a statistical image with the statistical representative value of luminance of each pixel as the luminance of the corresponding pixel And an abnormality determining unit that compares the statistical image with the reference image and determines whether or not an abnormality that permanently blocks at least a part of the imaging range of the imaging unit has occurred.

The objects and advantages of the invention will be realized and attained by means of the elements and combinations particularly pointed out in the appended claims.
It should be understood that both the foregoing general description and the following detailed description are exemplary and explanatory and are not restrictive of the invention as claimed.

  The imaging range abnormality determination device disclosed in this specification can appropriately determine whether or not an abnormality has occurred in which at least a part of the imaging range of the imaging unit is constantly blocked.

1 is a schematic configuration diagram of a monitoring system according to one embodiment. (A) is a figure which shows an example of the image produced | generated by the imaging | photography part when the whole imaging | photography range is not obstruct | occluded steadily, (b) is a part of imaging | photography range being interrupted steadily. It is a figure which shows an example of the image produced | generated by the imaging | photography part when it exists. It is a functional block diagram of a control part. (A)-(d) is a figure which shows an example of the sample image in case the whole imaging | photography range of an imaging | photography part is not obstruct | occluded permanently. (E) is a figure which shows an example of the statistical image produced | generated using the sample image shown by (a)-(d). In (a) to (d), after the sample images shown in FIGS. 4 (a) to (d) are acquired, the roadside tree grows and a part of the photographing range of the photographing unit is constantly blocked. It is a figure which shows an example of the sample image in a case. (E) is a figure which shows an example of the statistical image produced | generated using the sample image shown by (a)-(d). It is a figure which shows the operation | movement flowchart of a reference image generation process. It is a figure which shows the operation | movement flowchart of imaging | photography range abnormality determination processing.

Hereinafter, a monitoring system according to an embodiment will be described with reference to the drawings.
This monitoring system includes a photographing unit and a monitoring device. The monitoring device obtains a statistical representative value of luminance for each pixel at the same position of a plurality of images obtained by photographing the same photographing range over a predetermined period by the photographing unit. Then, the monitoring device generates a statistical image in which the statistical representative value of the luminance of each pixel is the luminance of the corresponding pixel. The monitoring device compares the statistical image with a reference image representing a shooting range in a state where the shooting range is not constantly obstructed, and whether or not an abnormality in which at least a part of the shooting range is constantly blocked occurs. Determine whether.

  FIG. 1 is a schematic configuration diagram of a monitoring system according to one embodiment. As shown in FIG. 1, the monitoring system 1 includes a photographing unit 2, a monitoring device 3, and a display device 4. The imaging unit 2 and the monitoring device 3 are connected via a network.

  For example, the photographing unit 2 is fixedly installed outdoors, and a surveillance worker such as a guard guards a road and a photographing range including a road in order to monitor a moving object such as a person or a vehicle on the road. Generate an image. For this purpose, the imaging unit 2 includes a two-dimensional detector composed of an array of photoelectric conversion elements having sensitivity to visible light, such as CCD or C-MOS, and an image of an imaging range including a road on the two-dimensional detector. And an imaging optical system for imaging. The imaging unit 2 captures the same imaging range at a certain imaging cycle (for example, 1/30 second), and generates an image of, for example, 640 × 480 pixels that captures the imaging range. In the present embodiment, the luminance of each pixel included in the image is represented by 256 gradations and takes a numerical value in the range of 0 to 255. The photographing unit 2 outputs the image to the monitoring device 3 every time an image is generated.

  2A and 2B are diagrams illustrating an example of an image generated by the photographing unit. The imaging unit 2 is installed so that the entire imaging range is not constantly obstructed before monitoring of a road and a moving object such as a person and a vehicle on the road is started (hereinafter referred to as “before operation”). In this case, the imaging unit 2 generates the image 200 shown in FIG. 2A, and the monitoring system 1 can sufficiently perform the monitoring function. Thereafter, dirt is attached to the lens of the photographing unit 2 or a roadside tree grows, and the photographing unit 2 generates an image 201 shown in FIG. As shown in the image 201, since at least a part of the photographing range of the photographing unit 2 is constantly blocked by the dirt 210 attached to the lens of the photographing unit 2 and the grown roadside tree 211, the monitoring system 1 The monitoring function cannot be sufficiently performed.

  The monitoring device 3 is, for example, a stationary computer installed indoors. Each time the monitoring device 3 receives an image from the photographing unit 2, the monitoring device 3 displays the image to the display device 4 in order to allow the monitoring worker to monitor the road and moving objects such as people and vehicles on the road. Display. In addition, the monitoring device 3 is an example of a shooting range abnormality determination device, and based on the image received from the shooting unit 2, is there an abnormality in which at least a part of the shooting range of the shooting unit 2 is constantly blocked? Determine whether or not. Then, when an abnormality has occurred, the monitoring device 3 displays a warning message to that effect on the display device 4 and notifies the monitoring worker, for example.

  The monitoring device 3 includes an interface unit 31, a storage unit 32, and a control unit 33. The interface unit 31 and the storage unit 32 are each connected to the control unit 33 via a bus.

  The interface unit 31 includes an interface circuit for connecting the monitoring device 3 to the network. The interface unit 31 receives an image from the photographing unit 2 via the network and passes the image to the control unit 33.

  The storage unit 32 includes, for example, a semiconductor memory such as an electrically rewritable nonvolatile memory and a volatile memory. And the memory | storage part 32 memorize | stores the various programs for controlling the monitoring apparatus 3, the temporary calculation result by the control part 33, etc. FIG. The various programs include an imaging range abnormality determination computer program for determining whether or not an abnormality that permanently blocks at least a part of the imaging range of the imaging unit 2 has occurred. Note that the storage unit 32 may store an image displayed on the display device 4 so that a monitoring worker, an administrator of the monitoring system 1 and the like can check the information later.

  The control unit 33 includes one or a plurality of processors (not shown) and their peripheral circuits. And the control part 33 controls the monitoring apparatus 3 whole by running the various programs memorize | stored in the memory | storage part 32 on the processor. For example, the control unit 33 executes an imaging range abnormality determination computer program stored in the storage unit 32 on the processor, thereby causing an abnormality in which at least a part of the imaging range of the imaging unit 2 is constantly blocked. It is determined whether or not.

  The display device 4 includes, for example, a liquid crystal display or an organic electronics luminescence display, and displays an image received from the photographing unit 2. When an abnormality occurs in which at least a part of the photographing range of the photographing unit 2 is constantly blocked, the display device 4 displays, for example, a warning message to that effect on the display device 4 and notifies the monitoring worker. To do.

  FIG. 3 is a functional block diagram of the control unit 33. As illustrated in FIG. 3, the control unit 33 includes a sample image acquisition unit 331, a statistical image generation unit 332, and an abnormality determination unit 333.

  The sample image acquisition unit 331 stores an image for each predetermined sampling period in the storage unit 32 as a sample image among images received from the imaging unit 2 during a preset predetermined sampling period.

  For example, when the administrator of the monitoring system 1 instructs the generation of a reference image through an operation unit (not shown) before operation, the sample image acquisition unit 331 starts acquiring the sample image. The administrator of the monitoring system 1 confirms that the entire photographing range of the photographing unit 2 is not constantly obstructed by dirt attached to the lens of the photographing unit 2 and a grown roadside tree. If the entire shooting range of the shooting unit 2 is not constantly blocked, the administrator of the monitoring system 1 determines that the shooting range of the shooting unit 2 is in a normal state, and generates a reference image. Can be indicated.

  The sample image acquisition unit 331 stores the image received from the imaging unit 2 in the storage unit 32 as a sample image for each sampling period. Then, the sample image acquisition unit 331 ends the acquisition of the sample image when a predetermined sampling period has elapsed since the generation of the reference image is instructed by the administrator of the monitoring system 1. In the present embodiment, the predetermined sampling period is set to 24 hours. The sampling period is set to 30 seconds. Therefore, a total of 2880 sample images are stored in the storage unit 32.

  On the other hand, during monitoring of roads and moving objects such as people and vehicles on the roads (hereinafter referred to as “in operation”), the sample image acquisition unit 331 determines whether or not it is the execution timing of the imaging range abnormality determination process. The execution timing of the imaging range abnormality determination process may be scheduled in advance so as to arrive regularly (for example, every week or every month). The sample image acquisition unit 331 also stores the image received from the imaging unit 2 as a sample image in the storage unit 32 for each sampling period during a predetermined sampling period at the execution timing of the imaging range abnormality determination process as before the operation. Remember.

  The statistical image generation unit 332 obtains a statistical representative value of luminance for each pixel at the same position of the plurality of sample images stored in the storage unit 32, and determines the statistical representative value of luminance of each pixel as a corresponding pixel. A statistical image with the brightness of is generated. In the present embodiment, the statistical image generation unit 332 calculates an average value of luminance as a statistical representative value of luminance.

  For example, the statistical image generation unit 332 reads a plurality of sample images stored during the sampling period from the storage unit 32 when the sampling period has elapsed since the start of acquisition of the sample image. The statistical image generation unit 332 calculates an average value of luminance for each pixel at the same position of the plurality of read sample images. The statistical image generation unit 332 generates a statistical image in which the average value of the luminance of each pixel is used as the luminance of the corresponding pixel. Therefore, the statistical image is an image of 640 × 480 pixels, like the sample image, and the luminance of each pixel is represented by 256 gradations.

  If the statistical image generation unit 332 is not yet in operation, the statistical image generation unit 332 stores the generated statistical image in the storage unit 32 as a reference image representing a shooting range in a state where the shooting range of the shooting unit 2 is not obstructed. On the other hand, the statistical image generation unit 332 stores the generated statistical image in the storage unit 32 as the current statistical image representing the state of the current shooting range when the shooting range abnormality determination process is being executed.

  4A to 4D show examples of sample images when the entire photographing range of the photographing unit is not obstructed constantly. FIG. 4 (e) shows an example of a statistical image generated using the sample images shown in FIGS. 4 (a) to 4 (d). On the other hand, in FIGS. 5A to 5D, the roadside tree grows after the sample images shown in FIGS. 4A to 4D are acquired, and a part of the shooting range of the shooting unit is constantly set. An example of the sample image in the case of being interrupted is shown. FIG.5 (e) shows an example of the statistical image produced | generated using the sample image shown by Fig.5 (a)-(d).

  The sample image 400 shown in FIG. 4A and the sample image 500 shown in FIG. 5A show a daytime road in which the vehicle is not traveling. The sample image 401 shown in FIG. 4B and the sample image 501 shown in FIG. 5B show an evening road where the vehicle is not traveling. In the sample image 402 shown in FIG. 4C, the evening road on which the vehicle 412 is traveling is shown. In the sample image 502 shown in FIG. 5C, the evening road on which the vehicle 512 is traveling is shown. Each is reflected. In the sample image 403 shown in FIG. 4D and the sample image 503 shown in FIG. 5D, a night road where the vehicle is not traveling is shown.

  The statistical image 404 shown in FIG. 4 (e) generates, as the luminance of the corresponding pixel, the average value of the luminance of the pixels at the same position of a plurality of sample images including the sample images 400 to 403 acquired during the sampling period. Is done. The statistical image 504 shown in FIG. 5E generates an average value of the luminances of the pixels at the same position of a plurality of sample images including the sample images 500 to 503 acquired during the sampling period as the luminance of the corresponding pixel. Is done.

  The change in the brightness of each pixel in the sample images 400 to 403 and the change in the brightness of each pixel in the sample images 500 to 503 are mainly caused by a change in the brightness of the day. In this embodiment, in order to reduce the influence on the brightness of each pixel on the statistical image due to such a change in brightness of the day, the sampling period is a change in brightness due to ambient light in the shooting range. One example of the cycle is 24 hours, that is, one day. Thereby, the brightness | luminance of each pixel of the range in which the fixed object other than the roadside tree 511 which has always blocked | interrupted a part of imaging | photography range of the imaging | photography part 2 is reflected between the statistical image 404 and the statistical image 504, respectively. In, it becomes substantially the same.

  In the sample image 402, the luminance of each pixel in the range in which the vehicle 412 is shown varies from the luminance of each pixel in the corresponding range of the sample image 401. Similarly, in the sample image 502, the luminance of each pixel in the range where the vehicle 512 is shown changes from the luminance of each pixel in the corresponding range of the sample image 501. Such moving objects such as the vehicles 412 and 512 are not always projected on the sample image and are not projected at the same position on the sample image. In other words, the moving objects such as the vehicles 412 and 512 are merely temporarily blocking the shooting range of the shooting unit 2, unlike the roadside tree 511 that constantly blocks a part of the shooting range of the shooting unit 2. In order to appropriately determine whether or not an abnormality has occurred in which at least a part of the shooting range of the shooting unit 2 is constantly blocked, a moving object that only temporarily blocks the shooting range of the shooting unit 2 is used. It is preferable that the brightness of each pixel on the statistical images 404 and 504 is not affected. Therefore, in the present embodiment, the number of sample images acquired in a predetermined sampling period is determined so that the luminance of each pixel on the statistical image is not affected by the variation in luminance of the corresponding pixel due to the moving object.

  That is, the standard deviation of the average luminance of the pixels at the same position of the sample image is inversely proportional to the square root of the number of sample images. For this reason, the standard deviation of the average value of the brightness | luminance of the pixel in the same position of a sample image becomes so small that the number of sample images used when producing | generating a statistical image is large. In the present embodiment, the luminance of each pixel of the sample image is represented by 256 gradations. Therefore, the statistical image generation unit 332 can sufficiently reduce the standard deviation of the average value of the luminance of the pixels at the same position of the sample image to, for example, one gradation or less by using 2500 or more sample images. Therefore, in this embodiment, the sampling period is set to 30 seconds, for example. In this case, since the sampling period is 24 hours, the number of sample images used for generating the statistical image is 2880. For this reason, in the statistical images 404 and 504, the influence on the brightness by moving objects such as the vehicles 412 and 512 that only temporarily block the shooting range of the shooting unit 2 is sufficiently reduced.

  On the other hand, the roadside tree 511 that constantly blocks the shooting range of the shooting unit 2 is different from moving objects such as the vehicles 412 and 512 that only temporarily block the shooting range, as shown in FIGS. The sample images 500 to 503 shown in FIG. And the brightness | luminance of each pixel of the range in which the street tree 511 of the sample images 500-503 shown by FIG. 5 (a)-(d) is reflected is sample image 400-shown by FIG. 5 (a)-(d). The luminance of each pixel in the corresponding range 403 is different. For this reason, also in the statistical image 504 shown in FIG. 5E, the luminance of each pixel in the range where the street tree 511 appears is equal to the luminance of each corresponding pixel in the statistical image 404 shown in FIG. It will be different.

  Thus, the luminance of each pixel in the statistical image 504 in the range where the shooting range of the imaging unit 2 is not constantly blocked by the street tree 511 is substantially equal to the luminance of each corresponding pixel in the statistical image 404. On the other hand, the luminance of each pixel in the statistical image 504 in the range where the photographing range of the photographing unit 2 is constantly blocked by the street tree 511 is different from the luminance of each corresponding pixel in the statistical image 404.

  The abnormality determination unit 333 compares the current statistical image with the reference image, and determines whether or not an abnormality in which at least a part of the imaging range of the imaging unit 2 is constantly blocked has occurred.

  For example, the abnormality determining unit 333 obtains a luminance difference value between corresponding pixels of the current statistical image and the reference image for each pixel. Then, the abnormality determination unit 333 obtains the number of pixels for which the luminance difference value is equal to or greater than the first threshold value. As described above, the luminance of each pixel in the range where the photographing range of the photographing unit 2 is not constantly obstructed is substantially equal to the luminance of the corresponding pixel in the range where the photographing range is not constantly obstructed. On the other hand, the luminance of each pixel in the range where the imaging range of the imaging unit 2 is constantly blocked is different from the luminance of the corresponding pixel in the range where the imaging range is not constantly blocked. For this reason, in the range where the imaging range of the imaging unit 2 is constantly blocked, the corresponding pixels of the current statistical image and the reference image are compared to the range where the imaging range of the imaging unit 2 is not constantly blocked. The difference value of the brightness between them increases. The abnormality determination unit 333 may constantly block the shooting range of the shooting unit 2 by comparing the difference value of the luminance between corresponding pixels of the current statistical image and the reference image with the first threshold value. A range of pixels can be extracted. Therefore, the first threshold value may be set to a value of about 10 to 50, for example, 20, for example.

  When at least a part of the shooting range of the shooting unit 2 is constantly blocked, compared to the case where the entire shooting range is not blocked, the corresponding pixel and brightness of the reference image among the pixels of the current statistical image The number of different pixels increases. That is, the number of pixels whose luminance difference value is equal to or greater than the first threshold value increases. Therefore, in the present embodiment, the abnormality determination unit 333 determines whether or not the number of pixels whose luminance difference value is greater than or equal to the first threshold is greater than or equal to the second threshold.

  The abnormality determination unit 333 is in a normal state where the entire imaging range of the imaging unit 2 is not obstructed constantly if the number of pixels whose luminance difference value is greater than or equal to the first threshold is less than the second threshold. It is determined that On the other hand, the abnormality determination unit 333 constantly blocks at least a part of the imaging range of the imaging unit 2 if the number of pixels whose luminance difference value is greater than or equal to the first threshold is greater than or equal to the second threshold. It is determined that an abnormality has occurred. Thereby, the abnormality determination unit 333 can appropriately determine whether or not an abnormality that constantly blocks at least a part of the imaging range of the imaging unit 2 has occurred. If there is an abnormality in which at least a part of the photographing range of the photographing unit 2 is constantly blocked, the abnormality determining unit 333, for example, superimposes a warning message indicating that on the image displayed on the display device 4. To display to the monitoring worker.

  Note that the abnormality determination unit 333 indicates that an abnormality in which at least a part of the photographing range of the photographing unit 2 is constantly blocked is generated by, for example, a warning sound output from a speaker (not shown). You may be notified. In addition, the abnormality determination unit 333 notifies the monitoring worker that the abnormality that the at least part of the imaging range of the imaging unit 2 is constantly blocked has occurred by, for example, lighting a lamp (not shown). You may be notified. Furthermore, the abnormality determination unit 333 transmits a notification that an abnormality has occurred in which at least a part of the photographing range of the photographing unit 2 is constantly blocked to, for example, a central management system connected via a network. Thus, the administrator of the monitoring system 1 may be notified. The abnormality determination unit 333 indicates that an abnormality in which at least a part of the photographing range of the photographing unit 2 is constantly interrupted is displayed on the display device 4, output to the speaker, lamp lighting, and central management. A part of transmission to the system or the like may be combined and notified. Alternatively, the abnormality determination unit 333 indicates that an abnormality in which at least a part of the imaging range of the imaging unit 2 is constantly interrupted is displayed on the display device 4, output to the speaker, lighting of the lamp, and You may notify by performing all the transmissions etc. to a central management system.

  FIG. 6 is an operation flowchart of reference image generation processing executed by the control unit 33. For example, the control unit 33 executes the reference image generation process in response to an instruction to generate a reference image by the administrator of the monitoring system 1 before operation.

  First, the sample image acquisition unit 331 stores the image received from the imaging unit 2 in the storage unit 32 as a sample image for each sampling period (step S101).

  The sample image acquisition unit 331 determines whether the sampling period has elapsed since the start of acquisition of the sample image (step S102). If the sampling period has not elapsed (No at Step S102), the sample image acquisition unit 331 repeats the processes after Step S101.

  On the other hand, when the sampling period has elapsed (step S102—Yes), the statistical image generation unit 332 generates a statistical image using a plurality of sample images stored in the storage unit 32 during the sampling period (step S103).

  Then, the statistical image generation unit 332 stores the generated statistical image as a reference image in the storage unit 32 (step S104), and ends the reference image generation process.

  FIG. 7 is an operation flowchart of an imaging range abnormality determination process executed by the control unit 33. During operation, the control unit 33 executes the imaging range abnormality determination process every time the execution timing of the imaging range abnormality determination process scheduled in advance arrives.

  First, the sample image acquisition unit 331 stores the image received from the imaging unit 2 in the storage unit 32 as a sample image for each sampling period (step S201).

  The sample image acquisition unit 331 determines whether the sampling period has elapsed since the start of acquisition of the sample image (step S202). If the sampling period has not elapsed (No at Step S202), the sample image acquisition unit 331 repeats the processes after Step S201.

  On the other hand, when the sampling period has elapsed (step S202—Yes), the statistical image generation unit 332 generates a current statistical image using a plurality of sample images stored in the storage unit 32 during the sampling period (step S203). ). Then, the statistical image generation unit 332 stores the current statistical image in the storage unit 32 (step S204).

  The abnormality determining unit 333 obtains a luminance difference value between corresponding pixels of the current statistical image and the reference image for each pixel (step S205).

  Then, the abnormality determining unit 333 obtains the number of pixels for which the luminance difference value is equal to or greater than the first threshold (step S206).

  The abnormality determination unit 333 determines whether or not the number of pixels whose luminance difference value is greater than or equal to the first threshold is greater than or equal to the second threshold (step S207). If the number of pixels for which the luminance difference value is equal to or greater than the first threshold is less than the second threshold (No in step S207), the abnormality determination unit 333 constantly blocks the entire imaging range of the imaging unit 2. It is determined that the state is not normal, and the abnormality determination process is terminated.

  On the other hand, when the number of pixels whose luminance difference value is greater than or equal to the first threshold is greater than or equal to the second threshold (Yes in step S207), the abnormality determination unit 333 determines that at least a part of the imaging range of the imaging unit 2 is It is determined that an abnormality that is constantly obstructed has occurred. In this case, the abnormality determination unit 333 causes the display device 4 to display a warning message indicating that an abnormality that permanently blocks at least a part of the imaging range of the imaging unit 2 has occurred (step S208). Thereafter, the control unit 33 ends the abnormality determination process.

  As described above, this monitoring apparatus obtains a statistical representative value of luminance for each pixel at the same position of a plurality of images obtained by the imaging unit imaging the same imaging range over a predetermined period. Then, the monitoring device generates a statistical image in which the statistical representative value of the luminance of each pixel is the luminance of the corresponding pixel. The monitoring device compares the statistical image with a reference image representing a shooting range in a state where the shooting range is not constantly obstructed, and whether or not an abnormality in which at least a part of the shooting range is constantly blocked occurs. Determine whether. Thereby, the monitoring apparatus can appropriately determine whether or not an abnormality has occurred in which at least a part of the imaging range of the imaging unit is constantly blocked.

  According to the modification, the statistical representative value of the luminance for each pixel at the same position of the plurality of sample images obtained as the luminance of the pixel of the statistical image is the median luminance or the mode of luminance. May be. In this case, since the variation in the luminance of the corresponding pixel between the sample images and the influence of the presence of the sample image having greatly different luminance on the pixel value of the statistical image can be reduced, the sampling period is more than that in the above embodiment. May be long.

  According to another modification, the monitoring device 3 may store a plurality of reference images having different environmental conditions in the storage unit 32. For example, the monitoring device 3 may generate a plurality of statistical images for each season and store them in the storage unit 32 as reference images. The average value of brightness for each pixel at the same position in multiple sample images acquired in summer is the average value of the brightness for each pixel at the same position in multiple sample images acquired in winter when the sunshine time is shorter than in summer. Higher than. For this reason, the luminance of each pixel of the statistical image generated in the summer is higher than the luminance of each pixel of the statistical image generated in the winter. Furthermore, the monitoring device 3 may generate a plurality of statistical images for each weather and store them as reference images in the storage unit 32, or generate a reference image for each season and each weather and store them in the storage unit 32. May be. The average value of luminance for each pixel at the same position of a plurality of sample images acquired during clear weather is the average value of the luminance for each pixel at the same position of a plurality of sample images acquired during cloudy weather, rainy weather, and snowy weather. Higher than. For this reason, the brightness of each pixel of the statistical image acquired during fine weather is higher than the brightness of each pixel of the statistical image acquired during cloudy weather, rainy weather, and snowy weather.

  For example, the statistical image generation unit 332 generates a statistical image for each season and / or weather, and stores each statistical image in the storage unit 32 as a reference image. Then, the abnormality determination unit 333 compares the current statistical image with each of a plurality of reference images stored in the storage unit 32 and having different environmental conditions. The abnormality determination unit 333 indicates that an abnormality has occurred when a comparison result between the current statistical image and at least one of the plurality of reference images satisfies a condition that at least a part of the shooting range is not constantly blocked. What is necessary is just to determine that it is not. For example, the abnormality determination unit 333 obtains a luminance difference value between corresponding pixels of the statistical image and the reference image for each of the plurality of reference images, and the luminance difference value is equal to or greater than the first threshold value. It is determined whether or not the number of pixels that are equal to or greater than the second threshold value. If there is at least one reference image in which the number of pixels whose luminance difference value is greater than or equal to the first threshold value is less than the second threshold value, the abnormality determination unit 333 causes the entire imaging range of the imaging unit 2 to be constant. What is necessary is just to determine with it being the normal state which is not obstruct | occluded. On the other hand, for any reference image, the abnormality determination unit 333 has at least a part of the imaging range of the imaging unit 2 when the number of pixels whose luminance difference value is equal to or greater than the first threshold is equal to or greater than the second threshold. What is necessary is just to determine with the abnormality which is obstruct | occluded steadily. As a result, the abnormality determination unit 333 has an abnormality in which at least a part of the imaging range of the imaging unit 2 is constantly blocked even when the luminance of each pixel on the reference image changes due to a change in the environmental condition of the imaging range. It can be determined more accurately whether or not it has occurred.

  Or the abnormality determination part 333 reads the reference image corresponding to the environmental condition at the time of the production | generation of the present statistical image from the memory | storage part 32 which memorize | stores the some reference image for every environmental condition, and compares with the present statistical image. May be. For example, the abnormality determination unit 333 may specify a season from date information or the like set in the control unit 33, read a reference image of the specified season from the storage unit 32, and compare it with the current statistical image. In addition, the abnormality determination unit 333 acquires weather information from a server device that manages weather information, for example, via a network, reads a weather reference image specified from the acquired weather information from the storage unit 32, and displays the current statistical image. May be compared. Alternatively, a thermometer, a hygrometer, or the like connected to the monitoring device 3 via a network is installed on the outer surface or the vicinity of the monitoring camera 2. The abnormality determination unit 333 may receive the temperature measured by the thermometer and the humidity measured by the hygrometer via the network, and may specify the weather from the received temperature and humidity. And the abnormality determination part 333 should just read the reference image of the identified weather from the memory | storage part 32, and should compare with the present statistical image. As a result, the abnormality determination unit 333 can use the reference image representing the state of the shooting range according to the current environmental conditions for detecting an abnormality of the shooting range, so that there is an abnormality in which at least a part of the shooting range is constantly blocked. It can be determined more accurately whether or not it has occurred.

Furthermore, according to another modification, each time the sample image acquisition unit 331 acquires a sample image, the statistical image generation unit 332 sequentially changes the luminance of each pixel of the sample image to the luminance of the corresponding pixel of the intermediate image. A statistical image may be generated by weighted addition. For example, the statistical image generation unit 332 sets the sample image acquired at the beginning of the sampling period as an intermediate image. Thereafter, each time the sample image acquisition unit 331 acquires a sample image, the statistical image generation unit 332 weights and adds the luminance of each pixel of the sample image to the luminance of the corresponding pixel of the intermediate image using a forgetting factor. . The luminance of the pixel at the position (x, y) of the sample image is I (x, y), the luminance of the pixel at the position (x, y) of the intermediate image is Ac (x, y), and the forgetting factor is α (for example, 0.01 ). In this case, the luminance Ac ′ (x, y) of the pixel of the updated intermediate image at the position (x, y) is expressed by the following equation.

  The statistical image generation unit 332 may use the intermediate image stored in the storage unit 32 as the statistical image when the sampling period has elapsed. In this way, since the monitoring device 3 does not have to store a plurality of sample images in the storage unit 32, the storage amount of the storage unit 32 can be reduced.

  According to yet another modification, the abnormality determination unit 333 applies to a pixel whose luminance difference value between corresponding pixels of the current statistical image and the reference image is equal to or greater than a first threshold (hereinafter referred to as a target pixel). By performing the labeling process, the pixels of interest that are adjacent to each other may be grouped. Then, for each group of target pixels, the abnormality determination unit 333 compares the number of target pixels included in the group with the second threshold, and if there is a group in which the number of target pixels is equal to or greater than the second threshold, It may be determined that an abnormality that constantly blocks at least a part of the imaging range has occurred. Accordingly, the abnormality determination unit 333 can appropriately detect an abnormality in the shooting range when an object that constantly blocks a clustered area in the shooting range of the imaging unit 2 appears in the shooting range.

  According to still another modified example, the abnormality determination unit 333 divides the current statistical image and the reference image into a plurality of regions, compares corresponding regions of the current statistical image and the reference image, It may be determined whether or not an abnormality has occurred in which at least part of the shooting range of 2 is constantly blocked. For example, the abnormality determination unit 333 may calculate the sum of squared differences (Sum of Squared Difference, SSD) of luminance differences between corresponding pixels in the corresponding region between the current statistical image and the reference image, or the luminance difference value. The sum of absolute values (Sum of Absolute Difference, SAD) is obtained. The abnormality determination unit 333 may determine that there is a difference between corresponding areas of the current statistical image and the reference image if the SSD or SAD is equal to or greater than a predetermined threshold. If the number of different areas is equal to or greater than a predetermined number (for example, an integer of 1 or more), the abnormality determination unit 333 can determine that an abnormality in which the imaging range is constantly blocked occurs in the different areas. .

  In addition, the abnormality determination unit 333 generates a normalized cross-correlation (NCC) or a zero average normalized cross-correlation (Zero−) of the luminance of pixels in a corresponding region between the current statistical image and the reference image. Mean Normalized Cross Correlation (ZNCC) may be obtained. The abnormality determination unit 333 may determine that there is a difference between corresponding areas of the current statistical image and the reference image if NCC or ZNCC is equal to or less than a predetermined threshold. If the number of different areas is equal to or greater than a predetermined number (for example, an integer equal to or greater than 1), the abnormality determination unit 333 indicates that an abnormality has occurred in which a portion of the imaging range corresponding to the different areas is constantly blocked. What is necessary is just to judge. In these modifications, the abnormality determination unit 333 not only determines whether or not an abnormality in which at least a part of the shooting range is constantly blocked occurs, but also includes a portion of the shooting range that is constantly blocked. Can be identified.

  Alternatively, the abnormality determination unit 333 compares the current statistical image with the reference image using feature points such as edges and corners extracted from each of the current statistical image and the reference image, and determines the shooting range of the imaging unit 2. It may be determined whether or not an abnormality that at least partially obscures has occurred. For example, if the number of feature points whose positions differ between the current statistical image and the reference image is equal to or greater than a predetermined number, the abnormality determination unit 333 has an abnormality in which a part of the imaging range of the imaging unit 2 is constantly blocked. You may determine that it has occurred.

  According to still another modification, the reference image has a shooting range of the shooting unit 2 blocked by dirt attached to the lens of the shooting unit 2, stationary objects such as a roadside tree, and moving objects such as people and vehicles. It may be a single image that captures the shooting range in a state that is not. The reference image may be subjected to image processing for adjusting the luminance so that the luminance of each pixel becomes an average value for one day. The reference image may be generated using fewer sample images than the current statistical image generated in the shooting range abnormality determination process.

  A computer program for determining an imaging range abnormality, which realizes the functions of the control units according to the above-described embodiments and modifications, is a computer-readable portable recording medium such as a semiconductor memory, a magnetic recording medium, or an optical recording medium. May be provided in a recorded form.

  All examples and specific terms listed herein are intended for instructional purposes to help the reader understand the concepts contributed by the inventor to the present invention and the promotion of the technology. It should be construed that it is not limited to the construction of any example herein, such specific examples and conditions, with respect to showing the superiority and inferiority of the present invention. Although embodiments of the present invention have been described in detail, it should be understood that various changes, substitutions and modifications can be made thereto without departing from the spirit and scope of the present invention.

DESCRIPTION OF SYMBOLS 1 Monitoring system 2 Imaging | photography part 3 Monitoring apparatus 4 Display apparatus 31 Interface part 32 Memory | storage part 33 Control part 331 Sample image acquisition part 332 Statistical image generation part 333 Abnormality determination part

Claims (7)

A storage unit for storing a reference image representing the imaging range in a state where the imaging range of the imaging unit is not obstructed constantly;
For each pixel at the same position of a plurality of images obtained by shooting the shooting range over a predetermined period, the shooting unit obtains a statistical representative value of the brightness, the statistical representative value of the brightness of each pixel, A statistical image generation unit that generates a statistical image with the brightness of the corresponding pixel;
An abnormality determination unit that compares the statistical image with the reference image and determines whether or not an abnormality in which at least a part of the photographing range is constantly blocked occurs;
An imaging range abnormality determination device having The abnormality determination unit
For each pixel, obtain a difference value of luminance between corresponding pixels of the statistical image and the reference image,
Find the number of pixels for which the difference value of the brightness is equal to or greater than a first threshold
When the number of pixels with the luminance difference value equal to or greater than the first threshold is equal to or greater than the second threshold, it is determined that an abnormality in which at least a part of the imaging range is constantly blocked occurs.
The imaging range abnormality determination device according to claim 1.   The imaging range abnormality determination device according to claim 1, wherein the predetermined period is a cycle of a change in brightness due to ambient light in the imaging range, and the statistical representative value of the luminance is an average value of luminance. The number of the plurality of images is the number of the standard deviation of the average luminance of the pixels at the same position of the plurality of images is less than one gradation of the luminance of the pixels in the statistical image.
The imaging range abnormality determination device according to any one of claims 1 to 3. The storage unit stores a plurality of the reference images representing the shooting range in a state where the shooting range is not constantly blocked under different environmental conditions,
The abnormality determination unit compares the statistical image with each of the plurality of reference images, and at least a part of the imaging range is constantly blocked for at least one of the plurality of reference images by the comparison. If it is determined that the abnormality has not occurred, it is determined that the abnormality has not occurred.
The imaging range abnormality determination device according to any one of claims 1 to 4. For each pixel at the same position of a plurality of images obtained by photographing the photographing range over a predetermined period, the photographing unit obtains a statistical representative value of the brightness, and corresponds the statistical representative value of the brightness of each pixel. Generate a statistical image with pixel brightness,
Whether the statistical image is compared with a reference image representing the imaging range in a state where the imaging range is not constantly obstructed, and whether or not there is an abnormality in which at least a part of the imaging range is constantly obstructed To determine,
An imaging range abnormality determination method. For each pixel at the same position of a plurality of images obtained by photographing the photographing range over a predetermined period, the photographing unit obtains a statistical representative value of the brightness, and corresponds the statistical representative value of the brightness of each pixel. Generate a statistical image with pixel brightness,
Whether the statistical image is compared with a reference image representing the imaging range in a state where the imaging range is not constantly obstructed, and whether or not there is an abnormality in which at least a part of the imaging range is constantly obstructed To determine,
A computer program for determining an imaging range abnormality for causing a computer to execute the operation.