JP2021034763A - Image processing device and image processing program - Google Patents

Abstract

To allow a high accuracy determination in an image processing device determining whether or not a monitoring space is in a monitoring difficult state, by comparing a reference image with an object image, even in the case where an environmental change which does not cause a problem in a monitoring state of the monitoring space is generated.SOLUTION: Reliability setting means 54 generates a reference image 44 as a photographing image 40 photographed when a mobile body of a detection object is not existed in a monitoring space, and a reliability image 48 in which reliability is set in each image region 60 on the basis of similarity of a stable image 46 on the basis of the other photographing image 40. Monitoring situation determination means 56 calculates an evaluation value indicating a similarity level with the object image and the reliability image 48 on the basis of the similarity level of the object image and the reliability image 48, which are the photographing image 40 of the determination object for each image region 60 after the setting of the weight so as to be larger as the reliability is higher and to be smaller as the reliability is lower in each image region 60, and determines that the monitoring space is in a monitoring difficult state when the evaluation value is less than a prescribed value.SELECTED DRAWING: Figure 2

Description

The present invention relates to an image processing apparatus and an image processing program.

Conventionally, an image processing device that detects a moving object from a captured image by comparing a captured image captured by a photographing unit such as a camera with a background image prepared in advance based on a past captured image is known. .. In such an image processing device, the background image is updated as needed in order to cope with fluctuations in the brightness of the shooting area (surveillance space) such as the movement of the sun and the lighting of lighting.

In order for intruders to pretend to be fluctuating in the shooting environment and easily invade the building, there is a plan to place a shield in advance so that the shooting section cannot see the intrusion route, and employees arbitrarily carry luggage in front of the shooting section. In some cases, a shield such as is placed, and the background image is updated by the image of the shield. That is, the background image may include a shield, making it difficult to detect (monitor) an intruder as a moving object.

In order to detect such a situation that is difficult to monitor, the current monitoring space is conventionally stored by storing a reference image that cannot be automatically updated and does not include a moving object, and comparing the reference image with the target image to be judged. Is determined whether or not the situation is difficult to monitor (for example, Patent Document 1).

Patent No. 3896193

When determining whether the monitoring space is in a difficult situation to monitor by comparing the reference image and the target image, when the environment of the monitoring space changes that does not cause a problem in the monitoring situation, there is a problem in the monitoring situation. Even though there was no such thing, it was sometimes misjudged that there was a problem with the monitoring status. Environmental changes in the monitoring space that do not cause problems in the monitoring status here are, for example, small items that are in the monitoring space (that is, included in the reference image) and that do not affect the monitoring status even if they move. For example, moving stationery or small cardboard boxes.

An object of the present invention is to change the environment of the monitoring space so as not to cause a problem in the monitoring status in an image processing device that determines whether the monitoring space is in a difficult situation to monitor by comparing the reference image and the target image. Even in such a case, it is possible to make a highly accurate determination.

The present invention is an image processing device that monitors whether or not a moving body exists in the monitoring space based on a captured image obtained by photographing the monitoring space, and the moving body in the captured image exists in the monitoring space. A reliability setting means for setting the reliability for each image area of the reference image according to the similarity between the reference image captured when the reference image is not used and the comparison image based on the captured image different from the reference image. The degree of difference between the reference image and the target image, which is the captured image to be determined, for each image area is obtained, and the monitoring space is based on the weight according to the degree of difference for each image area and the reliability. It is provided with a monitoring status determining means for determining whether or not the presence of a moving object can be monitored, and is characterized in that the weight of the high-reliability image region is larger than the set low-reliability image region. It is an image processing device.

Desirably, the monitoring status determining means obtains the degree of difference based on the comparison of the edges extracted from the reference image and the target image, respectively.

Desirably, the reliability setting means is characterized in that, in addition to the determination of similarity, the reliability is set according to the feature amount of the edge extracted from each of the image regions of the reference image.

Desirably, the monitoring status determining means calculates an evaluation value indicating the degree of disagreement between the reference image and the target image from the weight and the degree of difference for each image region, and based on the evaluation value, the said It is characterized in that it is determined whether or not the monitoring space is in a situation where the presence of the moving body can be monitored.

Desirably, the reliability setting means cumulatively updates the reliability for each image region according to the similarity obtained by sequentially comparing the reference image and the plurality of comparative images. Among the image regions, the higher the current reliability, the smaller the amount of decrease in the reliability when the reference image and the comparison image are dissimilar.

Desirably, the monitoring status determining means is homogeneous in the edge extracted from the reference image with respect to the weight, in which the edge does not exist in the reference image or the number of pixels corresponding to the edge is less than a predetermined number. It is characterized in that it is determined whether or not the monitoring space is in a state where the presence of the moving body can be monitored based on the weight having a larger weight in the peripheral neighborhood region than the region other than the peripheral neighborhood region and the degree of difference. And.

Desirably, the monitoring status determining means determines that the monitoring space cannot monitor the presence of the moving body when the average of the reliabilitys of the plurality of image regions is less than a predetermined value. ..

Desirably, the reliability setting means cumulatively updates the reliability for each image region according to the similarity obtained by sequentially comparing the reference image and the plurality of comparison images. The monitoring status determining means is in a situation where the monitoring space cannot monitor the presence of the moving body when the average of the reliabilitys of a plurality of the image regions tends to decrease in the cumulative update of the reliability. It is characterized in that it is determined.

Further, the present invention is an image processing device that monitors a computer based on a captured image of a surveillance space to see if a moving body exists in the monitoring space, and the moving body in the captured image is the moving body. The reliability is set for each image area of the reference image according to the similarity between the reference image captured when it does not exist in the monitoring space and the comparison image based on the captured image different from the reference image. The degree of difference between the reliability setting means and the target image which is the captured image to be determined is obtained for each image area, and based on the degree of difference for each image area and the weight according to the reliability. The monitoring space functions as an image processing device including a monitoring status determining means for determining whether or not the presence of the moving object can be monitored, and the reliability is higher than the set low reliability image region. It is an image processing program characterized by increasing the weight of an image area.

According to the present invention, in an image processing device that determines whether the monitoring space is in a difficult situation to monitor by comparing the reference image and the target image, an environmental change in the monitoring space that does not cause a problem in the monitoring situation occurs. Even in such a case, it is possible to make a highly accurate determination.

It is a block diagram of the security system which concerns on this embodiment. It is a block diagram of the image sensor which concerns on this embodiment. It is a figure which shows the example of the reference image. It is a figure which shows the example of a stable image. It is a figure which shows the example of the image area. It is the first figure which shows the example of the target image. It is the 2nd figure which shows the example of the target image. It is the first figure which shows the example of another target image. FIG. 2 is a second diagram showing an example of another target image. It is a flowchart which shows the flow of the reliability setting process. It is a flowchart which shows the flow of the monitoring status determination processing.

Hereinafter, embodiments of the present invention will be described.

FIG. 1 is a schematic configuration diagram of a security system 10 according to the present embodiment. The security system 10 is a security center connected to each security device 14 via a security device 14 installed in each monitored property 12 such as a store, an office, an apartment, a warehouse, or a house, and a communication network 16 such as a public telephone line. The device 18 and the user device 20 are included. Further, in the security system 10, whether there is a moving body (personal image in the present embodiment) to be detected from the monitoring space in the monitoring target property 12 based on the captured image obtained by capturing the monitoring space of the monitoring target property 12. It includes one or more image sensors 22 (image processing devices) for monitoring (detecting) whether or not, and a recording device 24 for recording a captured image captured by the image sensor 22. The image sensor 22 and the recording device 24 are communicably connected to the security device 14.

When the security device 14 receives an alarm signal from the image sensor 22 connected to itself via the premises LAN or the like, the security device 14 detects the alarm signal, the identification signal of the security device 14 itself, the monitored property 12, or an abnormality. The identification signal of the image sensor 22 is transmitted to the security center device 18. Therefore, the security device 14 has a communication interface for communicating with the image sensor 22, a communication interface for communicating with the security center device 18 and the user device 20, and a control unit for controlling them.

The security device 14 may be capable of operating in a plurality of security modes. For example, it may be possible to operate in a security set mode in which the monitoring space is monitored and a security release mode in which the monitoring space is not monitored.

The security center device 18 is composed of a so-called computer, and includes a communication interface for communicating with the security device 14 via the communication network 16, a display device such as a liquid crystal display, and a notification unit composed of a buzzer, an LED, or the like. .. When the security center device 18 receives an alarm signal from the security device 14 via the communication network 16, the security center device 18 notifies the monitored property 12 in which the security device 14 that has transmitted the alarm signal is installed, and the content of the detected abnormality. Notify the observer through the display device.

The user device 20 is also composed of a so-called computer, and remotely operates a communication interface for communicating with the security device 14 via the communication network 16, a display device such as a liquid crystal display, and the security device 14 such as a keyboard and a mouse. It has a user interface for inputting operation commands for. When the user device 20 observes the pre-registered monitored property 12 via the user interface, the user device 20 currently photographs the security device 14 installed in the registered monitored property 12. Various image request signals requesting that the captured image inside or the captured image recorded in the recording device 24 be transmitted to the user device 20 are transmitted. Then, when the captured image is received from the security device 14, the user device 20 displays the requested captured image on the display unit.

The recording device 24 is a recording medium that can be attached to and detached from the recording device 24, such as a magnetic disk device such as an HDD, a magnetic tape such as a DAT, and an optical recording medium such as a DVD-RAM, and data by accessing the recording medium. It consists of a device that reads and writes. The recording device 24 receives the captured image captured by the image sensor 22 from the security device 14, and records the captured image in association with the capture time.

FIG. 2 is a schematic configuration diagram of an image sensor 22 as an image processing device.

The communication unit 30 is an input / output interface for transmitting and receiving various setting signals and control signals between the image sensor 22 and the security device 14 via a communication network such as a premises LAN, and various types such as Ethernet (registered trademark). It is composed of the communication interface circuit of the above and the driver software that drives them. Specifically, when a moving body (intruder) is detected by the signal processing unit 50 described later, the communication unit 30 outputs an intrusion alarm signal indicating that the intruder has been detected to the security device 14.

The photographing unit 32 photographs the monitoring space in the monitored property 12 and acquires a photographed image. The photographing unit 32 according to the present embodiment is a fixed camera, and photographs a predetermined monitoring space at a constant angle of view.

Further, the photographing unit 32 has two shooting modes, a daytime mode in which the surveillance space is photographed with visible light and a nighttime mode in which the surveillance space is irradiated with near-infrared light illumination, and a plurality of photographing modes can be switched. The captured image can be acquired by the method of. Specifically, in the daytime mode, an IR cut filter is inserted into the optical path of the imaging optical system so as to cut near-infrared light, and the two-dimensional detector of the photographing unit 32 acquires a visible light image. On the other hand, in the night mode, the IR cut filter is removed from the optical path of the imaging optical system, and the two-dimensional detector acquires a near-infrared image. A method in which a visible camera that acquires a visible light image and a near-infrared camera that acquires a near-infrared light image are separately provided, and the sensitivity wavelength range of the photoelectric converter is changed by changing the voltage applied to the organic thin film having a laminated structure. In the daytime mode, a visible image may be taken, and in the nighttime mode, a near-infrared image may be taken, depending on a method of controlling the light. In this way, the photographing unit 32 includes a photoelectric converter having sensitivity to visible light, infrared light, and the like, an imaging optical system that forms an image of the monitoring space on the photoelectric converter, and an imaging optical system. It is configured to include an IR cut filter (infrared cut filter) provided in the camera, an infrared image sensor for detecting the temperature of an object to be imaged, and the like.

In the present embodiment, the photographing unit 32 photographs at regular time intervals (for example, 1/5 second) to acquire a photographed image, but the photographing interval of the photographing unit 32 is not limited to this. The acquired captured image is stored in the storage unit 38. In the present embodiment, the photographing unit 32 is included in the image sensor 22, but the photographing unit 32 may be provided outside the image sensor 22.

The illuminance sensor 34 includes, for example, a photodiode. The illuminance sensor 34 detects the illuminance (brightness) of the monitoring space and outputs an illuminance signal indicating the detected illuminance to the signal processing unit 50 described later. As the illuminance sensor 34, various known sensors can be used.

The illumination unit 36 includes, for example, a near-infrared LED. The illumination unit 36 irradiates near-infrared light toward the monitoring space when the photographing mode of the photographing unit 32 is the night mode. On the other hand, when the shooting mode of the shooting unit 32 is the daytime mode, the lighting unit 36 is turned off.

The storage unit 38 is composed of a semiconductor memory, a magnetic disk (HDD), an optical disk drive such as a CD-ROM / DVD-RAM, and a recording medium thereof. The storage unit 38 stores an image processing program for operating each unit of the image sensor 22. Further, as shown in FIG. 2, the storage unit 38 stores the captured image 40 acquired by the photographing unit 32. Further, the storage unit 38 stores the background image 42, the reference image 44, the stable image 46, and the reliability image 48 described below.

The captured image 40 is an image acquired by the photographing unit 32 sequentially photographing the monitoring space. The storage unit 38 stores captured images 40 for several tens of frames (images), and when a new captured image 40 is captured, the most acquired of the plurality of captured images 40 stored in the storage unit 38. The latest captured image 40 is stored after deleting the captured image 40 having an old time.

The background image 42 is an image that does not include an image corresponding to a moving body, and is created or selected by the signal processing unit 50 based on one or a plurality of captured images 40 stored in the storage unit 38. For example, the signal processing unit 50 obtains the difference between two captured images 40 adjacent to each other in time series, and obtains the average value of the absolute values of the brightness differences between the corresponding pixels between the two images. When there are no moving objects in both images, the average value of the absolute values of the brightness differences between the two images becomes considerably small. Therefore, the signal processing unit 50 selects one of the two captured images 40 whose average value is smaller than the predetermined reference as the background image 42.

The background image 42 is updated in order to cope with fluctuations in the brightness of the monitoring space due to fluctuations in the lighting state, diurnal fluctuations in the sun, and the like. In the present embodiment, the difference between the average luminance value of the plurality of pixels included in the newly acquired captured image 40 and the average luminance value of the plurality of pixels included in the already stored background image 42 is greater than a predetermined value. The background image 42 is updated when the image becomes larger. The predetermined setting is preset by an administrator or the like. Although various "predetermined values" appear in the present specification, they are also preset by an administrator or the like. It should be noted that it is determined whether or not to update the background image 42 by comparing the exposure amount when the new photographed image 40 is photographed with the exposure amount of the background image 42 already stored, instead of comparing the average brightness values. You may. Further, it may be determined whether or not to update the background image 42 based on the illuminance of the monitoring space detected by the illuminance sensor 34. Alternatively, the background image 42 may be updated at regular intervals (for example, every 10 minutes).

A plurality of background images 42 may be stored in the storage unit 38, or one background image 42 may be stored in the storage unit 38.

The reference image 44 is an image taken when the moving object to be detected does not exist in the monitoring space among the captured images 40 captured by the photographing unit 32. That is, the reference image 44 is a captured image 40 that does not include an image corresponding to a moving body. When the administrator or the like visually confirms the plurality of captured images 40, the reference image 44 is selected from among the plurality of captured images 40. Alternatively, since there are often no moving objects in the monitoring space when the image sensor 22 is started up (when the monitoring of the monitoring space is started), the captured image 40 taken when the image sensor 22 is started up is used as the reference image 44. You may try to. Alternatively, the frame-to-frame difference may be performed within a predetermined time after the image sensor 22 is started up, and the image with no difference, that is, the image in which the moving body does not exist may be used as the reference image 44. FIG. 3 shows a reference image 44 according to this embodiment.

The stable image 46 as a comparative image is an image created or selected based on a plurality of captured images 40. The details of the stable image 46 will be described together with the description of the processing of the stable image extraction means 52 described later. Further, the reliability image 48 is an image created based on the reference image 44. The details of the reliability image 48 will be described together with the description of the process of the reliability setting means 54 described later.

In the present embodiment, the captured image 40, the background image 42, the reference image 44, the stable image 46, and the reliability image 48 are stored in the storage unit 38 in the image sensor 22, but these images are stored in the storage unit 38. A storage unit provided outside the image sensor 22 may be stored in a storage unit accessible from the image sensor 22.

The signal processing unit 50 has a built-in microprocessor unit, a memory such as a ROM / RAM, and peripheral circuits thereof, and executes various signal processing in the image sensor 22. As shown in FIG. 2, the signal processing unit 50 exerts the functions of the stable image extraction means 52, the reliability setting means 54, the monitoring status determination means 56, and the moving object detecting means 58. By exerting these means, the signal processing unit 50 determines with high accuracy whether or not the monitoring space is in a state where the moving body can be monitored, and then the moving body is detected from the captured image 40. Hereinafter, each means included in the signal processing unit 50 will be described.

The stable image extraction means 52 extracts the stable image 46 based on the plurality of captured images 40 and stores it in the storage unit 38. In the present embodiment, the stable image extraction means 52 has a small difference in the result of the difference extraction process (hereinafter referred to as “background subtraction process”) between the captured image 40 and the background image 42 (the difference amount is large). There is little difference (difference amount) in the result of the difference extraction process (hereinafter referred to as "frame-to-frame difference process") between the captured image 40 and the other captured images 40 captured before and after the captured image 40. Is less than a predetermined amount) The image is extracted as a stable image 46. Desirably, among the plurality of captured images 40, the captured image 40 that satisfies the above conditions and has no illumination fluctuation is stabilized in the comparison with the background image 42 and the comparison with other frame images. It may be extracted as an image 46. Whether or not the illumination fluctuation has occurred can be determined based on the average luminance value of both images to be compared (the average luminance value of a plurality of pixels included in each image) or the difference in the exposure amount.

FIG. 4 shows an example of the stable image 46 in this embodiment. Compared to the reference image 44 shown in FIG. 3, the shadow of the pillar due to the sunlight is cast, the object in front of the pillar is moving, and the pen on the desk is gone, that is, Although the environment of the monitoring space is changing, the background image 42 is sequentially updated as described above, and an image close to the image shown in FIG. 4 can be stored as the background image 42. An image having a difference from the reference image 44 can also be extracted as a stable image 46.

Further, the stable image extraction means 52 performs inter-frame difference processing between the plurality of captured images 40 captured during a predetermined time, and as a result, the change is small continuously for a predetermined time (the amount of change is a predetermined amount). If it is less than), any one of the plurality of captured images 40 may be used as the stable image 46.

Further, the stable image extraction means 52 does not use one captured image 40 as a stable image 46, but divides the captured image 40 into a plurality of image areas, and performs background difference processing or inter-frames from one captured image 40. A process of extracting only the image region determined to have little change in the difference processing is executed for the plurality of captured images 40, and the stable image 46 is obtained by combining the image regions with little change extracted from the plurality of captured images 40. You may create it.

The stable image 46 is used for the reliability setting process in the reliability setting means 54 described later. As will be described later, the stable image extracting means 52 extracts a new stable image 46 every time the reliability (described later) is updated by the reliability setting means 54.

The reliability setting means 54 sets the reliability of each image area of the reference image 44 based on the reference image 44 and the stable image 46. Here, the image area is a certain area in the predetermined reference image 44. In FIG. 3, an example of the image area 60 is shown. Further, FIG. 4 also shows the image region 60 in the stable image 46. In the examples of FIGS. 3 and 4, the rectangular area separated by the alternate long and short dash line is the image area 60. One image area may be composed of a plurality of pixels or may be one pixel. Further, when the image area is composed of a plurality of pixels, the image area 60 does not have to be rectangular.

In the present embodiment, first, the reliability setting means 54 sets the initial reliability of the entire image area 60 to the median value. For example, if the reliability takes a value from 0 to 255, the reliability setting means 54 sets the initial reliability of the entire image area 60 to 128. The reliability does not have to be a numerical value. As an example, reliability may be indicated by a plurality of ranks. For example, the reliability may be represented by five ranks from rank A, which is the highest reliability rank, to rank E, which is the lowest reliability rank. Further, the initial reliability of the entire image area 60 may be the lowest value (0 in this embodiment) or the highest value.

The reliability setting means 54 may extract an edge from the reference image 44 and set the initial reliability of each image area 60 according to the edge feature amount of the edge extracted from the reference image 44. Here, the edge means a pixel in which the difference in pixel value between the adjacent pixels is equal to or larger than a predetermined value, and the edge feature amount (edge feature amount) is the difference between the pixel values. Means size. For example, the larger the edge feature amount of the reference image 44 included in the image area 60, the higher the initial reliability may be set. Further, the object included in the reference image 44 may be detected from the shape of the extracted edge, and the initial reliability of the image region 60 in which a larger object is detected may be set higher.

Next, the reliability setting means 54 compares the reference image 44 and the stable image 46, and whether or not the reference image 44 and the stable image 46 are similar for each image area 60, specifically, is similar. Determine if they are or are different. In the present embodiment, normalization correlation is performed for each pixel between the reference image 44 and the stable image 46 for each image area 60, and when the correlation value is equal to or higher than a preset first threshold value, the image area 60 is set. If it is determined that they are similar and the correlation value is less than the preset second threshold value, it is determined that the image area 60 is different, and the correlation value is equal to or more than the second threshold value and less than the second threshold value. In this case, it is determined that the image area 60 is in an intermediate state that does not correspond to any similarity or difference. The normalization correlation here is a process of calculating a correlation value based on whether or not the pixel values (luminance value and color value) of the corresponding pixels in the reference image 44 and the stable image 46 are similar. is there. The correlation value may be calculated by considering the pixel values of pixels in the vicinity of the pixel for which the correlation value is to be obtained, instead of comparing the pixel values of each pixel. Further, in the present embodiment, the first threshold value and the second threshold value are different values, but the first threshold value and the second threshold value may be the same. In this case, it is not determined to be an intermediate state, and each image region 60 is determined to be either similar or different (in this case, it may be expressed as dissimilar).

Whether each image area 60 of the reference image 44 and the stable image 46 is in a similar, different, or intermediate state (hereinafter, these are collectively referred to as “similarity”) is a plurality of images included in the image area 60. It may be determined from the average luminance value or the dispersed luminance value of the pixels. In this case, when the difference between the average luminance value or the dispersed luminance value of both images is less than a predetermined value, it is determined that the image areas 60 are similar, and the difference between the average luminance value or the dispersed luminance value of both images is predetermined. When it is equal to or more than the value, it is determined that the image area 60 is different.

Further, edges may be extracted from the reference image 44 and the stable image 46, respectively, and the similarity of the image area 60 may be determined by comparing the positions of the edges included in the image area 60.

Then, the reliability setting means 54 performs a process of increasing the reliability of the image area 60 determined to be similar and decreasing the reliability of the image area 60 determined to be different. The reliability of the image area 60 determined to be in the intermediate state, which does not correspond to the similarity or the difference, is not changed. The amount of increase in reliability when determined to be similar and the amount of decrease in reliability when determined to be different may be predetermined. The amount of rise and the amount of fall may be the same or different from each other. If the reliability of the image area 60 is already the highest value (256 in this embodiment), and if it is determined that the reference image 44 and the stable image 46 are similar in the image area 60, the reliability is increased. The degree shall not increase any more. Further, when the reliability of the image area 60 is already the minimum value and it is determined that the reference image 44 and the stable image 46 are different in the image area 60, the reliability does not decrease any more. It shall be. It should be noted that the reliability of the image area 60 from which the difference is extracted is not changed by making a difference between frames of the captured image 40 taken a short time before the stable image 46 determined to be similar to the reference image 44. It may be. Further, the reliability may be set for each shooting mode. That is, the reliability set in comparison with the stable image 46 taken in the daytime mode and the reliability set in comparison with the stable image 46 taken in the nighttime mode may be set respectively.

The reliability setting means 54 intermittently repeats the similarity determination between the reference image 44 and the stable image 46 in each image area 60, and the increase / decrease in the reliability of each image area 60. As described above, since the stable image extracting means 52 extracts a new stable image 46 each time the reliability is updated, the reference image 44 is different every time in the reliability updating process by the reliability setting means 54. It will be compared with image 46. As described above, the reliability setting means 54 determines each image according to a plurality of similarity between the reference image 44 and the plurality of stable images 46 obtained by sequentially comparing the reference image 44 and the different stable images 46. The reliability of the area 60 is cumulatively updated.

In the present embodiment, as described above, the stable image extraction means 52 extracts the stable image 46 based on the plurality of captured images 40, but the stable image extraction means 52 is trusted by the reliability setting means 54. One photographed image 40 acquired at the timing of the update process may be used as the stable image 46. In this case, the stable image extracting means 52 determines whether the captured image 40 acquired at the timing of the reliability update processing can be the stable image 46 by background subtraction processing and inter-frame difference processing (that is, the background image 42 and the background image 42). (Determining whether or not there is little change with other frame images), if it is determined that the stable image 46 can be obtained, the captured image 40 is designated as the stable image 46. When it is determined that the captured image 40 cannot be the stable image 46, the stable image extracting means 52 attempts to determine whether or not the captured image 40 captured next can be the stable image 46.

In the present embodiment, the reliability setting means 54 updates the reliability of each image area 60 every 10 minutes, but the update timing of the reliability is not limited to this. For example, the reliability may be updated at the timing when the difference becomes less than a predetermined value by the inter-frame difference processing between the newly captured captured image 40 and the captured image 40 captured immediately before the captured image 40. Alternatively, the reliability may be updated when the environment of the monitoring space changes. The change in the environment of the monitoring space can be determined based on, for example, a comparison of the average brightness value and the exposure amount between the newly captured captured image 40 and the captured image 40 captured immediately before the captured image 40. The environmental change of the monitoring space may be determined based on the detection result of the illuminance sensor 34. Further, in a time zone in which the environment change of the monitoring space is unlikely to occur (for example, midnight), the update interval may be longer than in other time zones. The reliability may not be updated periodically, and the reliability may be set or updated before the determination by the monitoring status determination means 56 described later is performed. That is, when the determination by the monitoring status determination means 56 is performed, the reliability may be set by using the stable image 46 stored in the storage unit 38 before the determination.

By comparing each stable image 46 having a different monitoring space environment with the reference image 44, the image area 60 of the reference image 44, which is less susceptible to environmental changes in the monitoring space, and the monitoring space The image area 60 that is susceptible to changes in the environment can be specified. That is, the reliability is an index showing the degree of resistance to changes in the environment of the monitoring space. The environmental change of the monitoring space here includes the movement of small items (such as stationery and small cardboard) in the monitoring space that do not affect the monitoring status even if they move, but light and shadow. Etc. (sunlight insertion, lighting extinguishing, shadow generation, etc.) may be included.

By the above-mentioned processing, the image in which the reliability is set in each image area 60 with respect to the reference image 44 is the reliability image 48, and the reliability image 48 is stored in the storage unit 38. FIG. 5 shows an example of the reliability image 48. In the example of FIG. 5, the shaded image region 60 is a region having relatively low reliability. As described above, the reliability can be expressed in a plurality of stages, but in FIG. 5, for the sake of simplicity, the plurality of image areas 60 are divided into two areas, a high reliability area and a low reliability area, and the reliability is high. The high image area 60 is represented without shading, and the low reliability image area 60 is represented with shading. Comparing the reference image 44 shown in FIG. 3 with the stable image 46 shown in FIG. 4, the image area 60 in which the shadow of the pillar is cast by the sunlight, before and the movement of the object in front of the pillar are compared. The image area 60 including the rear position, the image area 60 including the window where the sun appears, and the image area 60 including the top of the desk where the pen has disappeared are the image areas 60 with low reliability.

Desirably, the reliability setting means 54 extracts edges from the reference image 44 and the stable image 46, respectively, in addition to the above-mentioned similarity determination in the reliability update process, and for each image area 60, from the reference image 44. The reliability is set according to at least one of the feature amount of the extracted edge and the similarity between the reference image 44 and the stable image 46.

In the present embodiment, the update reliability of the image area 60 determined to be similar between the reference image 44 and the stable image 46 is calculated by the following equation 1.
Update reliability = current reliability + (edge feature amount of reference image 44 included in the image area 60 × coefficient A × similarity between the reference image 44 and the stable image 46 in the image area 60) ... 1)
The edge feature amount of the reference image 44 included in the image area 60 is the total value of the edge feature amounts of the plurality of edge pixels (pixels corresponding to the edges) included in the image area 60. That is, the clearer the edges included in the image area 60, or the more pixels as edges (for example, the longer the edge length), the larger the edge feature amount.

According to Equation 1, the larger the edge feature amount of the reference image 44 included in the image area 60, the higher the reliability when the reference image 44 and the stable image 46 are similar. This is because the pixels corresponding to the contour of the object tend to have a large edge feature amount, and in the image region 60 where the edge feature amount of the reference image 44 is large, the reference image 44 and the stable image 46 are similar to each other. This is because it is clearer that there was no movement of the object between the two images in the image area 60, and it can be said that such an image area 60 has high reliability.

Further, the greater the similarity between the reference image 44 and the stable image 46 in the image area 60, the greater the reliability when the reference image 44 and the stable image 46 are similar. The similarity here may be calculated by the above-mentioned normalization correlation, or is calculated based only on the edges extracted from the reference image 44 and the stable image 46 (for example, the degree of coincidence of the edge positions in both images). You may. In the above (Equation 1), the update reliability is set using the edge feature amount, the coefficient A, and the similarity, but the reliability is set using any one or two. May be good.

On the other hand, in the present embodiment, the update reliability of the image area 60 determined to be dissimilar (difference) between the reference image 44 and the stable image 46 is calculated by the following equation 2.
Update reliability = current reliability-coefficient B ... (Equation 2)
As shown in Equation 2, in the present embodiment, the reliability of the image region 60 determined to be dissimilar between the reference image 44 and the stable image 46 is reduced by a certain value (coefficient B). Even when it is determined that the reference image 44 and the stable image 46 are dissimilar, the edge feature amount of the reference image 44 included in the image area 60 and the degree of similarity between the reference image 44 and the stable image 46 are determined. The reliability may be lowered accordingly. For example, the larger the edge feature amount, the lower the reliability, and the smaller the similarity, the lower the reliability.

Incidentally, the coefficient A included in the equation 1 and the coefficient B included in the equation 2 may be predetermined. The coefficient A and the coefficient B may have the same value or may be different values from each other.

Further, the reliability setting means 54 may reduce the amount of decrease in reliability when the reference image 44 and the stable image 46 are dissimilar to each image area 60 as the current reliability is higher. Good. In other words, the lower the current reliability, the greater the amount of decrease in reliability when the reference image 44 and the stable image 46 are dissimilar.

More important is the edge extracted from the reference image 44, and the reliability setting means 54 sets the reliability to the lowest value (the present embodiment) for the image area 60 in which the edge does not exist in the reference image 44 among the image areas 60. Then, it may be fixed to 0) so that the reliability of the image area 60 is not updated. In particular, when the image area 60 is one pixel, the reliability setting means 54 may set the reliability only for the edge pixels extracted from the reference image 44. When the edge pixels are connected, the reliability may be set by setting the connected edge pixels as one image area 60.

As described above, the reliability setting means 54 cumulatively updates the reliability of each image area 60, but the update history of the reliability distinguished for each image area 60 is stored in the storage unit 38. You may. From the reliability update history, it is possible to grasp the time change of the reliability of each image area 60. Here, the reliability setting means 54 does not repeat the increase and decrease of the reliability (that is, the increase and decrease of the reliability are not repeated, in other words, the reliability is generally increased or decreased) with respect to the image area 60 which repeatedly increases and decreases with time of the reliability. The amount of decrease in reliability when the reference image 44 and the stable image 46 are dissimilar to each other may be made larger than the image region 60 (which tends to occur). For example, the image area 60 in which the branches or leaves of the plant are shaken, or the image area 60 in which shadows are repeatedly generated and disappeared by turning off and on the lights, repeats increase and decrease with time change of reliability. become. Such an image region 60 may be determined to be similar in comparison between the reference image 44 and the specific stable image 46, and the reliability may be temporarily increased, but a plurality of stable images 46 are compared. If you look at it, there is a change in the image, that is, the image area 60 whose reliability should be lowered. Therefore, for the image region 60 that repeatedly increases and decreases with time of reliability, the amount of decrease in reliability when the reference image 44 and the stable image 46 are dissimilar is increased in order to further reduce the reliability. Good.

The monitoring status determining means 56 obtains an evaluation value indicating the degree to which the target image, which is the captured image 40 to be determined (for example, the newly acquired captured image 40), and the reliability image 48 are not similar to each other, and obtains the evaluation value. Based on, it is determined whether or not the monitoring space is in a situation where the presence of a moving object can be monitored.

Specifically, first, the monitoring status determining means 56 calculates the degree of difference between the target image and the reliability image 48 (the content of the image is the same as the reference image 44) for each image area 60. The degree of difference is calculated from the number of edge disappearances described later.

Next, the monitoring status determining means 56 sets the weight of each image area 60 based on the reliability set for each image area 60 by the reliability setting means 54. Specifically, the higher the reliability, the larger the weight, and the lower the reliability, the smaller the weight. As a result, the weight of the low-reliability image area, which has a lower reliability than the high-reliability image area, is set smaller than the weight of the high-reliability image area, which has a relatively high reliability.

Then, the product of the degree of difference and the weight is calculated for each image area 60, and the evaluation value is calculated by summing the plurality of products obtained from the plurality of image areas 60. When the calculated evaluation value is equal to or higher than a predetermined value, the monitoring status determining means 56 determines that the monitoring space cannot monitor the existence of the moving object (that is, it is difficult to monitor), and the calculated evaluation value. If is less than a predetermined value, it is determined that the monitoring space is in a situation where the presence of a moving object can be monitored. When the monitoring status determining means 56 determines that the monitoring space is not in a situation where the presence of a moving object can be monitored, the monitoring status determining means 56 notifies the security center device 18.

As described above, the monitoring status determining means 56 determines whether or not the monitoring space is in a monitorable situation based on the evaluation value indicating the degree of difference between the target image and the reliability image 48 (reference image 44). However, in the process of calculating the evaluation value, since the weight of the image region 60 having low reliability is small, the similarity between the target image and the reliability image 48 is not considered so much (ultimately ignored). ). That is, the evaluation value is calculated without considering the region susceptible to environmental changes in the monitoring space. Therefore, even if there is a dissimilar part between the target image and the reliability image 48 due to the environmental change in the monitoring space even though the monitoring space is originally in a monitorable situation. The monitoring status determining means 56 can correctly determine that the monitoring space is in a monitorable situation. It should be noted that the evaluation is performed using only the judgment result of the high-reliability image area, ignoring the judgment result of the low-reliability image area without calculating the product of the difference and the weight (without setting the weight). The value may be calculated. In this case as well, it can be said that the evaluation value is calculated by making the weight of the high-reliability image area larger than the weight of the low-reliability image area.

For example, when determining the target image 40a as shown in FIG. 6, the shadow of the pillar due to the sunlight is generated in front of the pillar as compared with the reference image 44 (reliability image 48) shown in FIG. An object is gone and a box is placed in place of the pen on the desk. The monitoring space shown by the target image 40a is essentially a situation in which a moving body can be monitored. However, if a conventional method, that is, a determination method based on the degree of difference between the target image 40a and the reliability image 48 is used without using the reliability, the monitoring space can be monitored because there is a difference between the two images. There was a risk that it would be judged that the situation was not the same.

According to the present embodiment, as shown in FIG. 7, the shaded image area is a low-reliability image area, and the reliability is low, that is, the weight is small. Therefore, the low-reliability image region is not considered so much (ultimately ignored), and an evaluation value indicating the degree of difference between the target image 40a and the reliability image 48 is calculated. Therefore, the monitoring status determining means 56 can correctly determine that the monitoring space can be monitored even when the environment of the monitoring space changes so as not to cause a problem in the monitoring status.

On the other hand, FIG. 8 shows a target image 40b taken with a shield installed between the photographing unit 32 and the monitoring space. The monitoring space shown by the target image 40b is a situation in which it is really impossible to monitor a moving object. In the present embodiment, an evaluation value indicating the degree of difference between the target image 40a and the reliability image 48 is calculated in full consideration of the high-reliability image region without shading. Therefore, when the target image 40b and the reference image 44 of FIG. 3 are compared, there is an image region 60 (for example, the image region 60a of FIG. 9) that has a difference even in the high-reliability image region. Based on the difference in the image area 60, the monitoring status determining means 56 can correctly determine that the monitoring space is in a situation where it is impossible to monitor a moving object based on the target image 40b. In the target image 40b, the monitoring space is an example in which it is impossible to monitor a moving object due to a shield. However, according to the present embodiment, for example, the image sensor 22 is coated with a paint such as a spray. It is possible to detect even when the monitoring space becomes unmonitorable due to factors other than the obstruction.

Desirably, the monitoring status determining means 56 extracts edges from the target image and the reliability image 48, respectively, and compares the position of the edge extracted from the target image with the position of the edge extracted from the reliability image 48. Based on this, the above evaluation value may be calculated. Note that the target image 40a of FIG. 7 is an example of determining that the monitoring space can be monitored even when the environment changes due to sunlight or the movement of small objects. According to this, it can be detected that the situation can be monitored even if there is a change in the environment due to other factors. For example, even when smoke such as cigarettes hits the image sensor 22 and the captured image 40 is temporarily cloudy, the edge is extracted from the captured image because it is cloudy due to the smoke. Therefore, by comparing the edge extracted from the target image and the edge extracted from the reliability image 48 in the high-reliability image area, the similarity of the high-reliability image area becomes high, so that the monitoring space can be monitored. Can be determined to be. If the captured image 40 becomes cloudy for a predetermined time due to tar or oil from cigarette smoke or the like adhering to the image sensor 22, objects such as tar adhere to the image sensor 22 and the edges cannot be extracted, so that monitoring is not possible. It can be determined that the situation is different.

In the present embodiment, the monitoring status determining means 56 first classifies each image area 60 into a plurality of reliability ranks according to the reliability set by the reliability setting means 54. For example, it is classified into a low-reliability image region (reliability 0 to 110), a medium-reliability image region (reliability 111-180), and a high-reliability image region (reliability 181-255). Further, in each image area 60, the number of edge disappearances (degree of difference), which is the number of pixels that were edge pixels in the reliability image 48 but are not edge pixels in the corresponding pixels of the target image, is obtained.

On top of that, for each image area 60, the pixel number of the edge pixel included in the reliability image 48 is the edge pixel in the reliability image 48, but the corresponding pixel of the target image is no longer an edge pixel. Find the discrepancy rate (evaluation value), which is the ratio of numbers. Specifically, in the present embodiment, the discrepancy rate is calculated by the following equation 3.
Mismatch rate = (number of edge disappearances in low-reliability image area x 1 + number of edge disappearances in medium-reliability image area x 2 + number of edge disappearances in high-reliability image area x 3) / edge pixels extracted from reliability image 48 Total number x 6 ... (Equation 3)

Also in Equation 3, the number of edge disappearances in the low-reliability image region is multiplied by a small coefficient (1), and the image region 60 having a higher reliability is multiplied by a larger coefficient. It can be said that the calculation is made in consideration of the weight according to the reliability of each image area 60.

The monitoring status determining means 56 uses the calculated mismatch rate as an evaluation value, determines that the monitoring space cannot be monitored when the mismatch rate is equal to or higher than a predetermined value, and determines that the mismatch rate is less than the predetermined value. Determines that the monitoring space is in a monitorable situation. In the above method, an edge newly appearing in the target image at a position that did not exist in the reliability image 48 is not used for the determination. Although the discrepancy rate increases due to the movement / appearance of small items, the newly appearing edge may also be used to calculate the discrepancy rate as the number of edge disappearances.

For example, when there is a change in the environment of the monitoring space due to light, shadow, etc., the brightness of the entire or most of the target image changes uniformly with respect to the reliability image 48 (reference image 44). Edge extraction can be performed with high accuracy. Therefore, by calculating the evaluation value using the edge, the accuracy of the processing of the monitoring status determination means 56 can be improved.

In the calculation of the discrepancy rate, in the present embodiment, the number of edge disappearances in the low-reliability image region and the medium-reliability image region was considered, but the discrepancy rate is based only on the number of edge disappearances in the high-reliability image region. May be calculated. In that case, the discrepancy rate is calculated by, for example, the following equation 4.
Mismatch rate = number of edge disappearances in high-reliability image area / total number of edge pixels extracted from reliability image 48 ... (Equation 4)
In Equation 4, it can be considered that the weights of the low-reliability image region and the medium-reliability image region are set to 0 (minimum) and the weights of the high-reliability image region are maximized in Equation 3. Further, it is not necessary to calculate the discrepancy rate from the product of the number of edge disappearances and the weight. For example, the weight of the number of edge disappearances in the high-reliability image area is made larger than the number of edge disappearances in the low-reliability image area (ignoring the low-reliability image area), and the evaluation value is calculated from the number of edge disappearances. May be good.

Further, the monitoring status determining means 56 extracts an edge from the reliability image 48, and in a region where the edge does not exist in the reliability image 48 or a region where the number of pixels (edge pixels) corresponding to the edge is less than a predetermined value. A certain homogeneous region is detected, and the evaluation value is calculated by emphasizing the comparison result of the edge positions of the reliability image 48 and the target image in the peripheral vicinity of the homogeneous region rather than the region other than the peripheral peripheral region of the homogeneous region. You may try to do it. Incidentally, the homogeneous region, the peripheral neighborhood region thereof, and the region other than the peripheral neighborhood region are concepts different from the above-mentioned image region 60. In this case, for example, one image area 60 is divided into a peripheral neighborhood area and two other divided image areas, and the reliability setting means 54 corresponds to the peripheral neighborhood area with respect to the weight set in the image area 60. After increasing the weight of the divided image area and decreasing the weight of the divided image area corresponding to the area other than the surrounding neighborhood area, the evaluation value is calculated by the above processing. In other words, the evaluation value emphasizes the edges of objects that are not easily affected by environmental changes in the surveillance space such as floors, walls, pillars, and desks, and the edges of objects that are likely to be unable to be photographed when obstacles are installed. Make a calculation. As a result, even if an obstacle having a color similar to that of the floor or wall is installed so as to overlap the floor or wall, it can be determined that the situation is difficult to monitor with high accuracy.

Further, the monitoring status determining means 56 may determine whether or not the monitoring space can be monitored based on the reliability itself set in the reliability image 48. Specifically, in the reliability image 48, when the average reliability of the plurality of image areas 60 is less than a predetermined value, it may be determined that the monitoring space cannot monitor the presence of the moving object.

Alternatively, the monitoring status determining means 56 may determine whether or not the monitoring space is in a monitorable situation based on the reliability update history stored in the storage unit 38 by the reliability setting means 54. .. Specifically, when the average reliability of the plurality of image regions 60 tends to decrease, it may be determined that the monitoring space cannot monitor the presence of the moving object. Here, the fact that the average reliability is on a downward trend is not only when the average reliability is continuously lowered in a predetermined number of updates, but also when the average reliability is slightly increased in the middle of the update. , It is a concept including the case where the average of the predetermined amount reliability is lowered in the update of a predetermined number of times.

In the present embodiment, the degree of difference is calculated from the number of edge disappearances (whether the target image and the reliability image match or do not match for each image area), but the difference is not limited to this. For example, the target image and the reliability image 48 may be compared to extract a difference (difference such as an average luminance value), and the degree of difference may be obtained according to the extracted difference value. For example, when there is no difference, the degree of difference may be determined to be 0, and the larger the difference, the higher the degree of difference may be requested.

In the present embodiment, the monitoring status determining means 56 intermittently (for example, at predetermined intervals) determines whether or not the monitoring space is in a monitorable state, but the timing of the determination is not limited to this. For example, the determination may be performed each time the captured image 40 is acquired. Further, for example, at the timing when the security mode of the security device 14 is changed, it may be determined whether or not the monitoring space is in a state where the presence of a moving object can be monitored. Further, at the timing when the shooting mode of the shooting unit 32 is changed between the daytime mode and the nighttime mode, it may be determined whether or not the monitoring space is in a state where the presence of the moving object can be monitored. Further, as a result of the reliability update processing of the reliability setting means 54, the monitoring space moves at the timing when the reliability in the high reliability image area (image area 60 whose reliability is equal to or higher than a predetermined value) drops by a predetermined amount. It may be determined whether or not the presence of the body can be monitored.

The monitoring status determination means 56 may determine whether or not to determine the status of the monitoring space according to the number of times the reliability update process is performed by the reliability setting means 54. For example, when the reliability update process by the reliability setting means 54 is performed a predetermined number of times (for example, 10 times) or more, the monitoring status determination means 56 determines the status of the monitoring space, but the reliability update process is performed. If the number of times is less than the predetermined number of times, the status of the monitoring space may not be determined. Further, when the average reliability of the plurality of image regions 60 of the reliability image 48 is equal to or higher than a predetermined value, or when a predetermined number of image regions 60 having a reliability equal to or higher than a predetermined value exist, the status of the monitoring space is determined. In other cases, the status of the monitoring space may not be determined.

When the monitoring status determining means 56 determines that the monitoring space is in a situation where the presence of the moving object can be monitored, the moving object detecting means 58 moves from the target image by comparing the target image with the background image 42. Detect the body. Since the prior art may be used for the detection process of the moving body by comparing the target image and the background image 42, detailed description thereof will be omitted here. When a moving object is detected from the target image, the communication unit 30 transmits an intrusion alarm signal to the security device 14.

Hereinafter, the flow of the reliability setting process by the reliability setting means 54 will be described with reference to the flowchart shown in FIG.

In step S10, the reliability setting means 54 sets the initial reliability for the entire image area 60 of the reference image 44. In the present embodiment, as described above, the reliability setting means 54 sets the initial reliability of the entire image area 60 to the median value.

In step S12, the reliability setting means 54 selects one image area 60 to be set (updated) for reliability.

In step S14, the reliability setting means 54 compares the reference image 44 and the stable image 46 with respect to the image area 60 selected in step S12, and determines the similarity in step S16.

If it is determined in step S16 that they are similar, in step S18, the reliability setting means 54 increases the reliability of the image area 60. On the other hand, if it is determined in step S16 that they are different, in step S20, the reliability setting means 54 reduces the reliability of the image area 60. The increasing method and the decreasing method are as described above.

In step S22, the reliability setting means 54 determines whether or not the processing of steps S14 to S20 for all the image areas 60 included in the reference image 44 is completed. If there is an unprocessed image area 60, the process returns to step S12, and the reliability setting means 54 selects the unprocessed image area 60 in step S12 again, and repeats the processing of step S14 and subsequent steps.

When the processing for all the image areas 60 included in the reference image 44 is completed, the process proceeds to step S24, and in step S24, the reliability setting means 54 updates the reliability for each image area 60. The image 48 is stored in the storage unit 38.

Hereinafter, the flow of the monitoring status determination process by the monitoring status determination means 56 will be described with reference to the flowchart shown in FIG.

In step S30, the monitoring status determining means 56 determines whether or not the security mode of the security device 14 has been changed to the security set mode for monitoring the monitoring space. It waits until the security mode is changed to the security set mode, and if the security mode is changed to the security set mode, the process proceeds to step S32.

In step S32, the monitoring status determining means 56 determines whether or not the reliability of the reliability image 48 has been updated a predetermined number of times or more by the reliability setting means 54. When the number of times the reliability of the reliability image 48 is updated is less than a predetermined number of times, the monitoring status determination means 56 ends the monitoring status determination process without performing the process of step S34 or lower. When the number of times the reliability of the reliability image 48 is updated is equal to or more than a predetermined number, the process proceeds to step S34.

In step S34, the monitoring status determining means 56 calculates an evaluation value indicating the degree of difference between the target image and the reliability image 48 while considering the reliability of each image area 60 set by the reliability setting means 54. .. Specifically, as described above, the monitoring status determining means 56 sets the weight of each image area 60 to be larger as the reliability is higher, and the weight is set to be smaller as the reliability is lower, and then for each image area 60. The product of the degree of difference and the weight is calculated, and the evaluation value is calculated by summing the plurality of products obtained from the plurality of image areas 60.

In step S36, the monitoring status determining means 56 determines whether or not the evaluation value calculated in step S34 is less than a predetermined value. When the evaluation value is equal to or higher than the predetermined value, in step S38, the monitoring status determining means 56 determines that the monitoring space is not in a situation where the presence of the moving object can be monitored. On the other hand, when the evaluation value is less than a predetermined value, in step S40, the monitoring status determining means 56 determines that the monitoring space is in a situation where the presence of the moving body can be monitored.

Although the embodiments according to the present invention have been described above, the present invention is not limited to the above embodiments, and various modifications can be made without departing from the spirit of the present invention.

10 Security system, 12 Monitored property, 14 Security device, 16 Communication network, 18 Security center device, 20 User device, 22 Image sensor, 24 Recording device, 30 Communication unit, 32 Imaging unit, 34 Illumination sensor, 36 Lighting unit , 38 Storage unit, 40 Captured image, 40a, 40b Target image, 42 Background image, 44 Reference image, 46 Stable image, 48 Reliability image, 50 Signal processing unit, 52 Stable image extraction means, 54 Reliability setting means, 56 Monitoring status determination means, 58 moving object detecting means, 60, 60a image area.

Claims (9)

An image processing device that monitors whether or not a moving object exists in the monitoring space based on the captured image of the monitoring space.
Of the captured images, the reference image is captured according to the similarity between the reference image captured when the moving body is not present in the monitoring space and the comparative image based on the captured image different from the reference image. A reliability setting means for setting the reliability for each image area, and
The degree of difference between the reference image and the target image, which is the captured image to be determined, for each image area is obtained, and the monitoring space moves based on the degree of difference for each image area and the weight according to the reliability. A monitoring status determination means for determining whether or not the presence of the body can be monitored, and
With
The weight of the high-reliability image area is made larger than the set low-reliability image area.
An image processing device characterized by this. The monitoring status determining means obtains the degree of difference based on the comparison of the edges extracted from the reference image and the target image, respectively.
The image processing apparatus according to claim 1. In addition to the determination of similarity, the reliability setting means sets the reliability according to the feature amount of the edge extracted from each of the image regions of the reference image.
The image processing apparatus according to claim 2. The monitoring status determining means calculates an evaluation value indicating the degree of disagreement between the reference image and the target image from the weight and the degree of difference for each image region, and based on the evaluation value, the monitoring space creates an evaluation value. Determining if the presence of the moving object can be monitored,
The image processing apparatus according to any one of claims 1 to 3. The reliability setting means cumulatively updates the reliability for each image area according to the similarity obtained by sequentially comparing the reference image and the plurality of comparison images, and the image area. Of these, the higher the current reliability, the smaller the amount of decrease in the reliability when the reference image and the comparison image are dissimilar.
The image processing apparatus according to any one of claims 1 to 4. The monitoring status determining means is around a homogeneous region in which the number of pixels corresponding to the edge does not exist in the reference image or the number of pixels corresponding to the edge is less than a predetermined number among the edges extracted from the reference image with respect to the weight. It is determined whether or not the monitoring space is in a state where the presence of the moving body can be monitored based on the weight having a larger weight in the surrounding neighborhood region than the region other than the neighborhood region and the degree of difference.
The image processing apparatus according to claim 2 or 3. When the average of the reliabilitys of the plurality of image regions is less than a predetermined value, the monitoring status determining means determines that the monitoring space cannot monitor the presence of the moving object.
The image processing apparatus according to any one of claims 1 to 6, wherein the image processing apparatus is characterized. The reliability setting means cumulatively updates the reliability for each image area according to the similarity obtained by sequentially comparing the reference image and the plurality of comparison images.
The monitoring status determining means is in a situation where the monitoring space cannot monitor the presence of the moving body when the average of the reliabilitys of the plurality of image regions tends to decrease in the cumulative update of the reliability. Judge,
The image processing apparatus according to any one of claims 1 to 7. Computer,
An image processing device that monitors whether or not a moving object exists in the monitoring space based on the captured image of the monitoring space.
Of the captured images, the reference image is captured according to the similarity between the reference image captured when the moving body is not present in the monitoring space and the comparative image based on the captured image different from the reference image. A reliability setting means for setting the reliability for each image area, and
The degree of difference between the reference image and the target image, which is the captured image to be determined, for each image area is obtained, and the monitoring space moves based on the degree of difference for each image area and the weight according to the reliability. A monitoring status determination means for determining whether or not the presence of the body can be monitored, and
To function as an image processing device equipped with
The weight of the high-reliability image area is made larger than that of the set low-reliability image area.
An image processing program characterized by this.