Monitoring support systems, methods, and programs

Abstract

To support a surveillant in efficiently detecting a suspicious person or the like by appropriately displaying images to be focused on from surveillance images.SOLUTION: A camera unit (100) includes: a detection unit (151) that detects a specific action of a person in layer images by a detection method corresponding to respective layers, for a plurality of layer images corresponding to a plurality of layers that are captured by a camera at a predetermined time point and that differ in distance from the camera; an identification unit (152) that, when the specific action is detected, identifies an image group captured by the camera during a period from a first time before the predetermined time point to the predetermined time point; and a setting unit (153) that sets, for the identified image group, a priority for transmission that is higher than a normal priority of images.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present disclosure relates to a monitoring assistance system, method and program. [Background Art]

[0002] Generally, a surveillance camera samples some frame images from captured video data and transmits the sampled frame images to a monitor device. The monitor device displays the received frame images on the screen at intervals longer than the frame rate at which the images were actually captured. Therefore, when a monitor performs surveillance on the monitor screen, the video captured by the surveillance camera appears to the monitor to be played back frame-by-frame. Accordingly, a delay may occur in the detection of a suspicious person or the like by the monitor.

[0003] Patent Document 1 discloses a technique for changing the imaging area of a neighboring camera when the number of persons in a captured video is equal to or greater than a predetermined value. [Prior Art Documents] [Patent Documents]

[0004] [Patent Document 1] Japanese Unexamined Patent Application Publication No. 2005-117542 [Summary of the Invention] [Problem to be Solved by the Invention]

[0005] From the above, there has been a problem that it is difficult for a monitor to detect a suspicious person or the like from video displayed on a monitor device for video captured by surveillance cameras installed in stores and on streets. Note that the technique according to the aforementioned Patent Document 1 does not detect suspicious movement.

[0006] The present disclosure has been made in view of the above-described problem, and it is an object of the present disclosure to appropriately identify video that should be focused on from surveillance video By displaying this information, it helps monitors efficiently detect suspicious individuals. The purpose is to provide support systems, methods, and programs. [Means for solving the problem]

[0007] A first aspect of this disclosure is a photograph taken by a camera at a predetermined time, and the distance from the camera For multiple layered images corresponding to multiple layers with different properties, a detection method is applied to each layer. A detection unit detects a specific action of a person in the layered image, and the detection of the specific action If this occurs, the image captured by the camera between the predetermined time and the first time before A unit for identifying a group, and a priority for sending images with higher priority than images with normal priority. The present invention provides a monitoring support system comprising a setting unit for setting a specified group of images.

[0008] A second aspect of this disclosure is a computer that is photographed by a camera at a predetermined time and prior to For multiple layer images corresponding to multiple layers at different distances from the camera, each layer The steps include detecting a specific action of a person in the layer image using a detection method corresponding to the preceding If a specific action is detected, the camera will be activated between the predetermined time and the first time before that time. The steps involve identifying the group of images that were captured and sending them with priority over images of normal priority. This invention provides a monitoring support method that includes the step of setting a priority for the identified image group for the purpose of monitoring the specified image group. To provide.

[0009] A third aspect of this disclosure is a photograph taken by a camera at a predetermined time, and the distance from the camera For multiple layered images corresponding to multiple layers with different properties, a detection method is applied to each layer. detecting a specific action of a person in the layer image by means of when the specific action is detected, an image captured by the camera between the predetermined time point and a first time before the predetermined time point identifying an image group; setting a priority for causing the image group to be transmitted with higher priority than normal priority images setting the priority to the identified image group; and provides a monitoring support program that causes a computer to execute the steps .

Effects of the Invention

[0010] According to the present disclosure, by appropriately displaying video that should be noted from surveillance video, a monitor can detect suspicious individuals and the like efficiently, and a monitoring support system, method and program for supporting such detection can be provided.

Brief Description of Drawings

[0011] [Figure 1] It is a block diagram showing the overall configuration including the monitoring support system according to the first embodiment. [Figure 2] It is a block diagram showing the configuration of the camera unit according to the first embodiment. [Figure 3] It is a block diagram showing the configuration of the switch unit according to the first embodiment. [Figure 4] It is a block diagram showing the configuration of the monitor unit according to the first embodiment. [Figure 5] It is a flowchart showing the flow of monitoring support processing in the camera unit according to the first embodiment. [Figure 6] It is a flowchart showing the flow of monitoring support processing in the camera unit according to the first embodiment. [Figure 7] It is a diagram for explaining the concept of layer images and pattern matching according to the distance from the camera according to the first embodiment. [Figure 8]It is a diagram illustrating an example where a frame image transmitted at normal priority according to the first embodiment is displayed on a monitor unit. [Figure 9] It is a diagram illustrating an example where a frame image transmitted when an upper region matches a pattern (at a higher priority than normal) according to the first embodiment is displayed on a monitor unit. [Figure 10] It is a diagram illustrating an example where a frame image transmitted when a lower region matches a pattern (at a higher priority than normal) according to the first embodiment is displayed on a monitor unit. [Figure 11] It is a diagram illustrating an example where a frame image transmitted when a person moves from an upper region to a lower region (at a higher priority than normal) according to the first embodiment is displayed on a monitor unit. [Figure 12] It is a block diagram showing the hardware configuration of a monitoring support apparatus according to another embodiment. DETAILED DESCRIPTION OF EMBODIMENTS

[0012] Hereinafter, specific embodiments of the present disclosure will be described in detail with reference to the drawings . In each drawing, the same elements are denoted by the same reference numerals, and for clarity of description, necessary redundant descriptions will be omitted as appropriate.

[0013] <First Embodiment> FIG. 1 is a block diagram showing the overall configuration including the monitoring support system 1000 according to the first embodiment . The monitoring support system 1000 includes camera units 100-1 to 100-n (where n is a natural number of 1 or greater) and a switch unit 200. Monitor units 300 -1 to 300-m (where m is a natural number of 1 or greater) appropriately display video data captured by the monitoring support system 1000 on their screens. A monitor then monitors the screen of the monitor unit 300-1 or other units to detect suspicious persons or the like within the screen. Therefore, the monitoring support system 100 0 allows surveillance personnel to efficiently identify suspicious individuals by appropriately displaying noteworthy footage from the surveillance images. This is an information system to assist in target detection. Here, camera unit 100 Each of the -1 to 100-n is connected to the switch unit 200. Each of the 300-1 to 300-m units is connected to the switch unit 200.

[0014] Camera unit 100-1 etc. are surveillance support systems including surveillance cameras installed in stores and on the streets. This is an example of a configuration. Each of the camera units 100-1 to 100-n has an equivalent configuration. In cases where there is no particular need to distinguish between them, the following explanation will simply refer to them as camera units. It shall be written as 100. Also, each of the monitor units 300-1 to 300-m In the following explanation, if they have equivalent configurations and there is no particular need to distinguish between them, then they are simply referred to as: It shall be referred to as Monitor Unit 300.

[0015] The camera unit 100 takes pictures at predetermined intervals and sets priority for the captured images and switches The image is sent to the camera unit 200. In particular, the camera unit 100 transmits the captured image to the camera at a distance from the camera. The image is acquired as multiple layered images corresponding to the distance, and specific actions of the person are detected from each layered image. In this case, the priority of the frame image corresponding to the layer image is set higher than usual.

[0016] Figure 2 is a block diagram showing the configuration of the camera unit 100 according to this embodiment 1. Furthermore, the camera unit 100 can be considered an example of a surveillance support system or surveillance support device. The unit 100 consists of a camera 110, a depth sensor 120, a timer 130, and an image buffer 1 Camera 1 comprises a 40, detection unit 151, identification unit 152, setting unit 153, and transmission unit 154. 10 takes pictures at predetermined intervals, for example, at a frame rate, and acquires the captured frame images. The depth sensor 120 measures the distance between the camera 110 and the subject as depth. Therefore, the camera 110 and the depth sensor 120 together are TOF (Time Of Flight). It may also be a camera. At a minimum, camera 110 captures frames at a predetermined time. Along with the image, based on the depth measured by the depth sensor 120, the camera 110 and the subject Obtain multiple different layered images depending on the distance. Here, the layered images are obtained at the same point in time. This is an image of a specific region within a captured frame image. However, multiple images taken at the same time are not included. The layered images may have overlapping regions within the frame images. Also, camera 110 This process saves the acquired frame image and the corresponding multiple layer images to the image buffer 140. ru.

[0017] The image buffer 140 contains frame images 141, 142, taken at multiple different time points. ...is a memory area that holds ... For example, frame image 141 is layer image 1411 , including 1412.... Or, frame image 141 is layer image 1411, 141 It is saved in association with 2... Images from frame image 142 onwards have a similar structure. The image buffer 140 is, for example, a semiconductor memory.

[0018] Timer 130 periodically outputs sampling pulses to detection unit 151. Here, The interval between pumping pulses should be sufficiently longer than the shooting interval of camera 110. For example, if camera 110 is shooting at 60fps (frames per second), the sample The ring pulse shall be output at intervals of 3 frames or more. Furthermore, Timer 130 shall... For example, analog circuits or digital circuits.

[0019] The detection unit 151, upon receiving a sampling pulse from the timer 130, For each layer, image recognition and other processing are performed on the layer image. Specifically, the detection unit 151 This uses a layer-specific detection method to detect the specific actions of people within the layer image of the corresponding layer. The detection unit 151 performs detection of whether a person in the layer image may be a suspicious person. It determines whether or not there is a possibility, and if it determines there is a possibility, it detects a specific action. Alternatively, detection unit 1 51 determines whether a person in the layer image is performing any suspicious actions or behaviors, and determines whether the suspicious actions If it is determined that movement or other actions are occurring, a specific action will be detected.

[0020] Furthermore, the detection unit 151 identifies people by pattern matching between images as a detection method. It is good to detect the actions of the object. For example, the detection unit 151 detects the layer image taken at the first time. The image is compared with a layer image of the same layer taken at a second time point, which is later than the first time point. Then, the difference in data size and the change in pixels are calculated. The detection unit 151 then... If the difference amount, etc., exceeds a predetermined value, the person is considered to match a specific behavioral pattern. It may be detected that a specific action is being performed. A specific action is, for example, a person moving. This includes things like the movement of specific body parts of a person. The detection unit 151 detects, for example, the first and second If the positions of people in the layered images taken at the same time are separated by a predetermined distance or more, then the first The person moved between the first and second time points, and this can be detected as a specific action. Furthermore, the detection unit 151 compares the layer images of the first and second time points to identify specific parts of a person in the image. If the distance and angle change at a predetermined rate, the identification of the person from the first time point to the second time point can be determined. A body part is being moved and may be detected as a specific action. Here, a specific body part is, for example, This could include, but is not limited to, the arms, fingertips, or the direction of the face.

[0021] Furthermore, the detection unit 151 detects when the layer is less than a predetermined distance from the camera 110. By using pattern matching of specific body parts of a person within a layered image as the detection method, Specific actions may be detected. Furthermore, the detection unit 151 detects when the layer is at a predetermined distance from the camera 110. If the distance exceeds a certain level, the detection method will be pattern matching of the overall position of the person within the layer image. By doing so, specific actions may be detected.

[0022] Furthermore, the detection unit 151 detects a second time interval in each layer according to the distance from the camera 110. And, corresponding to the layers, the pixels in a predetermined number of layer images taken at different time points. The average value of the change is calculated. Then, the detection unit 151 detects if the average value is greater than or equal to a predetermined value. For layer images taken at a predetermined time corresponding to Ya, the detection method corresponding to the layer is used. It is good to detect specific behaviors. Furthermore, the first and second times mentioned above are: The interval is greater than or equal to the shooting interval by camera 110, and within the interval of the sampling pulse, and is predetermined It is a fixed time. Also, the first time and the second time may be different times. Example For example, the second time is the time when the frame image containing one of the layer images was taken, and the other This is the interval between the time the frame image containing the layer image was taken and the time the second layer image was taken. The time may differ for each layer.

[0023] If a specific action is detected, the identification unit 152 will perform the following actions between a predetermined time and the first time before that time. The group of images captured by camera 110 is identified. Here, for example, the capture times are at predetermined intervals. If the shooting interval is time t1, t2, t3, t4..., then the sampling pulse is The timing is assumed to be times t1, t4, etc. In this case, from a predetermined time, the first The time period before refers to the time when the detection unit 151 last received a sampling pulse from the timer 130. The timing of the digit, for example, the most recent time received after time t1, for example, time t4 This refers to the period up to the specified time. In other words, the period from a predetermined time to the first time before is defined as time t1, t2, or t3 This will be the time from one of these to time t4. Also, for example, if the shooting interval of camera 110 is 6 If the frame rate is 0fps, the time interval between a given point in time and the first time interval before that point in time is, for example, 1 / 20th of a second. This is not limited to this.

[0024] The setting unit 153 specifies a priority for sending images with higher priority than normal images. Set to the selected image group. Note that the image with normal priority is the one switched from camera unit 100. The priority level for sending data to the monitor unit 300 via the control unit 200 is set to normal level. This refers to an image that is a sample of some of the frame images from captured video data. Frame images sampled at regular intervals correspond to images with normal priority. Prioritizing sending images over other priority images is based on whether a specific action was detected. Compared to the frame images, the group of frame images with a higher transmission priority, i.e., the video data The priority is set accordingly. Here, priority can be expressed as a numerical value or level value, etc. For example, Previously, the degree of fit of pattern matching or the importance of suspicious behavior may also be used.

[0025] The transmitting unit 154 transmits a set of images with assigned priority to the display device. Specifically, The transmitting unit 154 switches the monitor unit 300 as the destination, regardless of the priority level. Each image packet is sent to the control unit 200. The transmission unit 154 sends packets with higher priority. It is acceptable to send the packet with priority. In that case, if no lower-priority packets have been sent. That's good too.

[0026] Furthermore, the detection unit 151, the identification unit 152, the setting unit 153, and the transmission unit 154 are functional blocks. Yes, depending on the hardware, software, or combination of hardware and software. It will be realized.

[0027] Figure 3 is a block diagram showing the configuration of the switch unit 200 according to this embodiment 1. The switch unit 200 includes a buffer 210, a receiver 220, a scheduler 230, and It includes a transmitting unit 240. The buffer 210 receives multiple packets received by the switch unit 200. This is a memory area that holds ket 211, 212, ... Buffer 210 is, for example, These are semiconductor memory, etc. Packet 211 has the source address, destination address, In addition to the packet size, it includes priority 2111. Also, packet 211 is payload It contains compressed data corresponding to the frame image. Similarly, packet 212 takes precedence over the header. It includes a degree 2121 and the payload contains compressed data corresponding to the frame image.

[0028] The receiving unit 220 receives signals from each of the multiple camera units 100-1 to 100-n. The camera unit receives packets via the corresponding network interface. The transmitted packets are stored in the buffer 210. The receiving unit 220, for example, each camera unit This includes circuits and other components, such as a communication interface module with the device.

[0029] The scheduler 230 sends each packet stored in the buffer 210 according to its priority. Select it as the target. Specifically, scheduler 230 prioritizes packets with higher priority. The system then selects the target for transmission. The scheduler 230 is, for example, a hardware circuit. Alternatively, it could be a processor that executes the scheduler program.

[0030] The transmitting unit 240 receives packets selected by the scheduler 230 from the monitoring unit 30. It sends to 0. The transmitting unit 240 is, for example, a communication interface with the monitor unit 300. This refers to circuits, etc., that include modules.

[0031] Figure 4 is a block diagram showing the configuration of the monitor unit 300 according to this embodiment 1. The monitor unit 300 comprises a receiver 310, a decoder 320, and a monitor 330. The signal unit 310 receives packets from the switch unit 200 via the network interface. The receiver 310 receives the signal and outputs it to the decoder 320. The receiver 310 is, for example, a switch unit. This includes circuits such as a communication interface module for the 200.

[0032] The decoder 320 receives the compressed data contained in the payload of the packet received from the receiver 310. The data is decoded and restored to image data. Decoder 320 performs, for example, decoding. It is an analog circuit or a digital circuit, etc. Then, the monitor 330 displays the restored image data The data is displayed on the screen. The monitor 330 is, for example, a liquid crystal display device.

[0033] Figures 5 and 6 show the flow of monitoring support processing in the camera unit according to this embodiment 1. This is the flowchart shown. First, the camera 110 is measured by the depth sensor 120. Based on depth, multiple layered images captured at a predetermined point in time are acquired. Then, Camera 1 10 saves the acquired multiple layer images to the image buffer 140 (S101).

[0034] Figure 7 shows the layer image and pattern matching according to the distance to the camera according to this embodiment 1. This is a diagram to explain the concept of layering. In this example, layer image 41 is below the frame image. Layer image 42 shows the upper region of the same frame image, while layer image 42 shows the upper region of the same frame image.

[0035] Layer image 41 is image data from Layer 1, which is relatively close to the camera 110. Layer 1 indicates a distance less than a predetermined distance from camera 110. Layer image 41 contains rays. It contains pixel information indicating that a person exists at position P2 within Y1. Even if a person located at position P2-1 moves their arm from position P2-1 to P2-4, this person will not move to the same position. It is assumed that it remains in ear 1. Therefore, in layer image 41, position P within layer 1 is shown. Pixel information is included to indicate the presence of arms in 2-1 and P2-4.

[0036] Layer image 42 is image data from Layer 2, which is relatively far from the camera 110. Layer 2 indicates a distance greater than a predetermined distance from camera 110. Layer image 42 contains rays. This includes pixel information indicating the presence of a person at location P1-1 or P1-4 within Y2. Furthermore, even if a person located at position P1-1 moves to position P1-4, if this person remains the same It is assumed that it remains in Layer 2. Therefore, Layer Image 42 contains the position within Layer 2. The pixels at locations P1-1 and P1-4 contain pixel information indicating the presence of a person.

[0037] Returning to Figure 5, we continue the explanation. After step S101, for example, timer 130, sample Assume that a pulse is output to the detection unit 151. At this time, the detection unit 151 detects multiple pulses. The unselected layer is selected (S102). Then, the detection unit 151 selects Read the layer images corresponding to the layer from the most recent image up to a predetermined time prior from the image buffer 140. (S103). Furthermore, the detection unit 151 uses the most recent layer image and the layer image immediately preceding it, that is, A total of 2 or more layers, and at least one untransmitted layer image, shall be read. Meanwhile, detection unit 1 51 excludes the layer images that have already been sent. Alternatively, camera unit 100 selects the most recent Image data captured prior to the frame image sent may be deleted.

[0038] Then, the detection unit 151 calculates the amount of pixel change between consecutive layer images (S10 4). Here, consecutive layer images are layer images taken at adjacent times. For example, The detection unit 151 encodes the layered image and calculates the bitrate for each picture type. Picture type refers to the type of encoding, such as I-picture, B-picture, and P-picture. Then, the detection unit 151 calculates the bitrate for each picture type for consecutive layer images. The difference is calculated. Alternatively, the detection unit 151 calculates the pixel amount of each layer image and then... The detection unit 151 calculates the difference in the number of pixels between consecutive layer images. Alternatively, the detection unit 151 calculates the difference in the number of pixels between each layer Calculate the data size of the difference frames of the image, and then calculate the difference in data size after encoding. Furthermore, the amount of pixel change refers to the change in pixel brightness, etc., between layered images, exceeding a predetermined value. This is the number of pixels.

[0039] Then, the detection unit 151 calculates the average value of the pixel change amount (S105). For example, If three or more layer images are read in step S103, then adjacent sets of layer images The result is 2 or more. Then, in step S104, the detection unit 151 detects a pixel change of 2 or more. The quantity and difference value are calculated. Therefore, the detection unit 151 calculates the pic calculated in step S104. The average value of cell change amounts, etc., is calculated. Note that two layer images are read in step S103. If this occurs, since there is only one pair of adjacent layer images, step S105 is omitted, and The calculated value of S104 can be interpreted as the average value.

[0040] Subsequently, the detection unit 151 determines whether the average value is greater than the threshold corresponding to the layer. (S106). Here, the threshold corresponding to the layer is, for example, in the case of layer 1, a specific part This is the value of the pixel change that may indicate suspicious behavior, and in the case of Layer 2, the movement of the person The movement value represents the amount of pixel change that may indicate suspicious behavior. The thresholds should be adjustable and changeable for each layer.

[0041] If in step S106 it is determined that the average value is greater than the threshold corresponding to the layer, the detection unit 151 selects the detection method corresponding to the selected layer (S107). For example, the detection method The formula, in the case of Layer 1, is a pattern matching method for detecting the movement of a person's arm. In the case of Layer 2, a pattern matching method may be used to detect the movement of a person.

[0042] The detection unit 151 then detects the most recently captured layer image and the layer image captured immediately before it. The image is pattern-matched using the selected detection method (S108). For example, layer In case 1, the detection unit 151 detects the region of a person's arm from each layer image by image recognition, The differences in position, length, angle, etc. of the arm region between the two layer images are calculated. In this case, the detection unit 151 detects the area of ​​the person's overall position from each layer image by image recognition. Then, the difference in the detected area between the two layered images is calculated.

[0043] Then, the detection unit 151 determines whether or not it matches the pattern (S109). Specifically, the detection unit 151 determines whether the difference calculated in step S108 is greater than or equal to a predetermined value. If the difference is greater than or equal to a predetermined value, the detection unit 151 determines that it matches the pattern. For example, in the case of Layer 1, if the difference in the position, length, angle, etc. of the arm region exceeds a predetermined value, the arm The movements are suspicious and match the pattern. Also, in the case of Layer 2, the whole person If the difference in position, i.e., the distance traveled, exceeds a predetermined value, it is considered a sudden movement and is therefore suspicious. Let's assume it matches the pattern. Note that this is not the only example of a pattern match.

[0044] If it is determined in step S109 that there is a match in the pattern, the detection unit 151 determines the shooting time For the layer image corresponding to the selected layer, a higher priority than usual is determined. (S110). Furthermore, the frame image 51 is stored in memory, etc., with its layer image and priority associated with it. You may record it.

[0045] After step S110, or after determining in step S109 that there is no pattern match In this case, the detection unit 151 determines whether all layers have been selected or not (S111). If step S106 determines that the average value is below the threshold corresponding to the layer, then step S111 is also performed. Execute.

[0046] In step S111, not all layers were selected; in other words, there are unselected layers. If so, return to step S102 and repeat the subsequent processing. In step S111, all layers If selected, the specific unit 152 gives a higher priority than usual to the processed layer image. It is determined whether the degree has been determined (S112). Specifically, the identification unit 152 determines the memory, etc. Refer to this and determine whether the priority associated with the layer image is higher than usual.

[0047] If it is determined in step S112 that a higher priority than usual has been determined, the specific unit 152 The frame image set from a predetermined time to the first time before is identified (S113). This occurs after step S101, when the detection unit 151 receives a sampling pulse, that is, This is the point at which the frame image or layer image to be processed is captured. Then, the specific unit 15 2 selects from the image buffer 140 the frame image at a predetermined time and the first time from the predetermined time. Read one or more frame images taken up to the point in time as a set of frame images. Therefore, a group of frame images with an interval shorter than the sampling interval is identified.

[0048] Then, the setting unit 153 sets step S110 for each packet of the identified frame image group. The priority determined is set (S114). That is, the setting unit 153 sets the identified frame The compressed data obtained by compressing each image in the image group is set as the payload of the transmission packet, and the header A higher priority than usual is set for this. Furthermore, the setting unit 153 divides one image into multiple packets. You can.

[0049] If it is determined in step S112 that a higher priority than usual has not been determined, then the specific unit 1 52, the identification unit 152 identifies the frame image at a predetermined time (S115). Identification unit 15 Step 2 reads the frame image from the image buffer 140 at a predetermined point in time. Then, it sets... Unit 153 sets the normal priority for each packet of the identified frame image (S116 ). In other words, the setting unit 153 sends compressed data, which is the image of the identified frame image compressed. Set the payload of the trust packet and set the normal priority in the header. Note: Configuration section 15 Option 3 allows for splitting a single image into multiple packets.

[0050] After step S114 or S116, the transmitting unit 154 sends each packet to the switch unit The data is sent to 200 (S117). At this time, the transmitting unit 154 sends each packet in the order in which it was captured. You can send it as an afterthought.

[0051] Note that steps S104 to S106 may be omitted. In that case, step S After step 103, execute steps S107 onwards. Alternatively, execute steps S107 to S109. This process may be omitted. In that case, in step S106, the average value is the threshold corresponding to the layer. If it is determined that the value is greater than the specified value, the steps from S110 onward are executed.

[0052] Subsequently, the switch unit 200 controls multiple camera units 100-1 to 100-n Packets received from each are stored in buffer 210, and according to the priority of each packet... The packets are then sent to monitor units 300-1 through 300-m in that order.

[0053] The monitor unit 300 displays the image data corresponding to the received packet each time. Therefore, the video will be displayed according to the reception interval. If the packet is sent to the monitor unit 300 according to the normal priority, the monitor unit The T300 will reproduce the frame image at the sampling interval. Therefore, the monitor If the interval between packet receptions from the switch unit 200 is long, the Knit 300 monitor For them, the video will be displayed like a stop-motion animation.

[0054] Meanwhile, the camera unit 100 detects frames in which a specific action has been detected by the detection unit 151. Set a high priority for the image and a predetermined number of preceding frame images, and then switch unit It is sent to the 200. Therefore, the switch unit 200 will send packets with normal priority. The packet group with the highest priority is sent to the monitor unit 300. Therefore, The monitor unit 300 uses a shorter reception interval than usual for high-priority packet groups. The received and decoded frame image is displayed on the screen. In other words, monitor unit 300 For footage in which suspicious behavior is detected, the system will display it at a higher frame rate than usual. This allows for a more detailed and smoother observation of the movements of a person who may be a suspicious individual. The video footage allows for visual confirmation, enabling efficient detection of suspicious individuals.

[0055] Figure 8 shows the frame image transmitted with normal priority according to this embodiment 1 on the monitor unit. This figure shows an example displayed in T300. Here, frame images 51, 53 and 54 Each of these consists of the upper region 511 and the lower region 512, the upper region 531 and the lower region 532, It includes the upper region 541 and the lower region 542. Also, the upper region 511 and the lower region 512 are There may be some overlapping areas. The same applies to other areas thereafter. Also, turtle Ra110 shall take pictures at intervals of time t1, t2, t3, t4... The sampling interval of the camera unit 100 is set so that time t4 follows time t1. The same applies to other explanations thereafter.

[0056] At time t1, the monitor unit 300 displays frame image 51. In the upper region 511 of image 51, a person is located at position P1-1, far from camera 110, and below In area 512, a person is located at position P2, close to camera 110, and the person's arm is positioned It is assumed to be located at P2-1.

[0057] Subsequently, the frame image taken at time t4, after the sampling interval has elapsed from time t1. Regarding 54, the camera unit 100 performs the processes shown in Figures 5 and 6 above. At that time, the camera unit 100 captures the frame image 54 and the frame taken at the most recent time t3. The layer images are compared with image 53 to determine whether or not the patterns match. In this example, the camera unit 100 is positioned at position P1-b in the upper region 541 and upper region 53 Since the pixel change amount from position P1-a is below the threshold, the upper region, which is layer 2, It is determined that the amount of movement of the person is small. Also, the camera unit 100 is in the lower region 542. Pixel change between position P2 and position P2-1 and position P2 and position P2-1 in the lower region 532 Since the quantity is below the threshold, it is determined that the arm movement of the person in the lower region, which is Layer 1, is minimal. Therefore, the camera unit 100 sets the normal priority to the frame image 54 and switches The data is sent to the switch unit 200. Conversely, the camera unit 100 takes a picture at time t3. The frame image 53 is not transmitted. The switch unit 200 then transmits the frame image The packet of image 54 is sent to the monitor unit 300. Therefore, the monitor unit 300 At time t4, frame image 54 is displayed. In other words, at time t1, frame image 51 is displayed. After that, frame image 51 was not updated at times t2 and t3, and at time t4 the frame image 54 is displayed. Therefore, the observer can view the frame images from 51 to 54 in a stop-motion manner. It appears to have been updated. However, the person in upper region 541 has moved from position P1-1 to position P1-b Since it was merely a movement, there is little need to issue a warning as suspicious behavior, thus it has little impact on surveillance. There are few.

[0058] Figure 9 shows the case where the upper region according to this embodiment 1 matches the pattern (higher than usual). This figure shows an example of a frame image previously transmitted being displayed on the monitor unit 300. The frame image 61 displayed at time t1 is the same as the frame image 51 described above. do.

[0059] Subsequently, at time t4, after the sampling interval has elapsed from time t1, the camera unit 100 is the position P1-4 in the upper region 641 of frame image 64 and the position in the upper region 631. Because the pixel change between P1 and P3 is greater than the threshold, it is determined that the person in the upper region has moved a large amount. Therefore, if the pattern matches in Layer 2, the camera unit 100 will be at time t4. Then, identify frame images 62 to 64 up to time t2, which is the first time period prior, and in each packet A higher priority than usual is set, and each packet is sent to the switch unit 200. The camera unit 100 transmits frame images 62 and 6, which are not subject to transmission under normal priority. 3 is also transmitted. Then, the switch unit 200 transmits frame images 62, 63 and 64. The packet is sent to the monitor unit 300 with priority. Therefore, the monitor unit 300 At time t1, frame image 51 is displayed, then at time t2, frame image 62 is displayed, and at time t3, Frame image 63 displays frame image 54 at time t4. Therefore, the observer will see frame From image 61 to 64, the person in the upper area is located at positions P1-1, P1-2, P1-3, P1 The movement of -4 can be seen with smooth motion. Therefore, the person in the upper area is suspicious. It allows for efficient determination of whether or not someone is a person.

[0060] Figure 10 shows the case where the lower region according to this embodiment 1 matches the pattern (higher than usual). This diagram shows an example where a frame image sent to the (priority) unit is displayed on the monitor unit 300. The frame image 71 displayed at time t1 is the same as the frame image 51 described above. Let's assume that.

[0061] Subsequently, at time t4, after the sampling interval has elapsed from time t1, the camera unit 100 is the position P2-4 in the lower region 742 of frame image 74 and the position in the lower region 732 Because the pixel change between P2 and P3 is greater than the threshold, the amount of arm movement of the person in the lower region is high. This is the determination. Therefore, the pattern matches in Layer 1, and the camera unit 100 determines that time t Identify frame images 72 to 74 from 4 to time t2, which is the first time period prior, and for each packet... Set a higher priority than usual for each packet and send it to the switch unit 200. Furthermore, the camera unit 100 transmits frame images 72 and which are not subject to transmission under normal priority. It also transmits frame images 72, 73 and 7 Packet 4 is given priority and sent to monitor unit 300. Therefore, monitor unit 30 0 displays frame image 71 at time t1, then frame image 72 at time t2, and time t Frame image 73 is displayed at time 3, and frame image 74 is displayed at time t4. Therefore, the observer, From frame images 71 to 74, the arms of the person in the lower region are at positions P2-1, P2-2, P2- 3. The movement from P2 to P4 can be seen as a smooth motion. Therefore, in the lower region It is possible to efficiently determine whether the person is suspicious. In this example, the observer is at the time Between t2 and t3, the person in the upper region moves from position P1-1 to P1-3. This can also be seen.

[0062] Figure 11 shows the case when a person moves from the upper area to the lower area according to this embodiment 1 (normally This shows an example where a frame image sent to a higher priority unit is displayed on the monitor unit 300. This is a diagram.

[0063] At time t1, the monitor unit 300 displays frame image 81. In the upper region 811 of image 81, a person is located at position P3-1, which is far from camera 110, and below Assume that no people exist in area 812.

[0064] Subsequently, at time t4, after the sampling interval has elapsed from time t1, the camera unit 100 is the position P3-4 in the lower region 842 of frame image 84 and the position in the lower region 832 Because the pixel change amount between P3 and P3 is greater than the threshold, it is determined that the amount of movement of the person in the lower region is large. Therefore, if the pattern matches in Layer 1, the camera unit 100 will be at time t4. Then, identify frame images 82 to 84 up to time t2, which is the first time period prior, and in each packet A higher priority than usual is set, and each packet is sent to the switch unit 200. The camera unit 100 transmits frame images 82 and 8, which are not subject to transmission under normal priority. 3 is also transmitted. Then, the switch unit 200 transmits frame images 82, 83 and 84. The packet is sent to the monitor unit 300 with priority. Therefore, the monitor unit 300 At time t1, frame image 81 is displayed, then at time t2, frame image 82, and at time t3... Frame image 83, frame image 84 is displayed at time t4. Therefore, the observer, From image 81 to 84, the person who was in the upper area is located at positions P3-1, P3-2, P3-3, P The movement from 3 to 4 can be seen with smooth motion. Therefore, the person on the screen is not a suspicious person. It allows for efficient determination of whether or not that is the case.

[0065] Here, in order to narrow down the target video for surveillance and to detect suspicious individuals, It is also possible to train an image recognition algorithm on the behavior of the judge. However, In some cases, this required costs and time. In contrast, in this embodiment, advanced image recognition Without needing to train a logic system, it can identify noteworthy footage from surveillance images and prioritize communication. It is raised and sent to the display device. Therefore, if the possibility of suspicious activity is low, the communication load and While keeping costs down, if there is a high possibility of suspicious behavior, the video will be displayed at an appropriate frame rate. It is possible to display an image.

[0066] <Other Embodiments> Furthermore, the aforementioned camera unit 100 is realized by an information processing device, which is a monitoring support device. This may also be done. Figure 12 shows the hardware of the monitoring support device 100a according to another embodiment. This is a block diagram showing the configuration. The monitoring support device 100a includes a camera 110 and a depth sensor 12 0, Timer 130, Image Buffer 140, Control Unit 150, Storage Unit 160, Memory 170 and It also includes a communication unit 180, a camera 110, a depth sensor 120, a timer 130, and an image battery. Since F140 is the same as in Figure 2 above, its explanation will be omitted.

[0067] The memory unit 160 is an example of a storage device such as a hard disk or flash memory. Unit 160 stores the monitoring support program 161. The monitoring support program 161 is as described above. This is a computer program that implements monitoring support processing and the like according to the embodiment described above.

[0068] Memory 170 is a volatile storage device such as RAM (Random Access Memory), and the control unit This is a memory area for temporarily holding information during operation of 150. The communication unit 180 monitors The internal configuration of the support device 100a and the communication interface circuit with the switch unit 200 be.

[0069] The control unit 150 is a processor, or control device, that controls each component of the monitoring support device 100a. The control unit 150 transfers the monitoring support program 161 from the storage unit 160 to the memory 170. The data is loaded, and the monitoring support program 161 is executed. As a result, the control unit 150 performs the above-mentioned The detection unit 151, identification unit 152, setting unit 153, and transmission unit 154 are realized.

[0070] Alternatively, each component of the monitoring support device 100a may be implemented with dedicated hardware. It may also be done. Furthermore, some or all of the components of each device may be general-purpose or dedicated circuits. (Circuitry), processors, etc., or combinations thereof may be implemented. These are It may consist of a single chip, or multiple chips connected via a bus. It may be configured as follows. Some or all of the components of each device may be programmed with the circuits etc. described above. It may also be achieved by combining with M. Furthermore, as a processor, CPU (Central Pr Processing Unit), GPU (Graphics Processing Unit), FPGA (Field-Program Mabl) It is possible to use a Gate Array, a quantum processor (quantum computer control chip), etc. ru.

[0071] The present invention has been described above in reference to the above embodiments, but the present invention is based on the configuration of the above embodiments. Not limited to, but also including, a person skilled in the art within the scope of the claims of the present patent application. This naturally includes all possible variations, modifications, and combinations.

[0072] Although the above embodiment was described as a hardware configuration, it is not limited to this. This disclosure does not mean that any processing will cause the CPU to execute a computer program. It could also be achieved through [this method].

[0073] In the example described above, the program, when loaded into a computer, is described in the embodiment. A set of instructions (or software) to cause a computer to perform one or more of the specified functions. Includes (Accord). The program is stored on a non-temporary computer-readable medium or in physical storage. It may be stored in a medium. Not limited to, but for example, a computer-readable medium or a physical medium Storage media include random-access memory (RAM), read-only memory (ROM), and flash memory. Solid-state drives (SSDs) or other memory technologies, CD-ROMs, digital versatile drives SC (DVD), Blu-ray (registered trademark) discs or other optical disc storage, magnetic cassette Includes magnetic tapes, magnetic disk storage, or other magnetic storage devices. The program may be transmitted on a temporary computer-readable medium or communication medium. As an example, rather than a fixed one, temporary computer-readable media or communication media are electrical, optical, Includes acoustic or other forms of propagating signals. [Explanation of Symbols]

[0074] 1000 Monitoring Support System 100 Camera Units 110 Camera 120 Depth Sensor 130 timer 140 Image Buffer 141 frame image 1411 Layer Image 1412-layer image 142 frame image 151 Detection unit 152 Specific part 153 Settings Section 154 Transmitter 200 Switch Units 210 buffers 211 packets 2111 Priority 212 packets 2121 Priority 220 Receiver 230 Scheduler 240 Transmitter 300 Monitor Units 310 Receiver 320 Decoders 330 monitors 100a Monitoring support device 150 Control Unit 160 Storage section 161 Monitoring Support Program 170 memory 180 Communications Department 41, 42 Layer images 100-1~100-n Camera Unit 300-1~300-m Monitor Unit

Claims

1. The system includes a detection unit that detects specific actions of a person in a layer image using a detection method appropriate to each layer, for multiple layer images corresponding to multiple layers that are captured by a camera at a predetermined time and are at different distances from the camera. The detection unit is Within a predetermined time period corresponding to the distance from the camera, the average value of the amount of change in the target pixel corresponding to each layer is calculated for a predetermined number of layer images corresponding to each layer and captured at different time points. For the layer image captured at the predetermined time corresponding to the layer whose average value is greater than or equal to a predetermined value, the detection method is used to detect the specific action. Monitoring support system.

2. Computers The steps include: detecting a specific action of a person in a layer image using a detection method corresponding to each layer, with respect to multiple layer images corresponding to multiple layers that are captured by a camera at a predetermined time and are at different distances from the camera; The steps include: calculating the average value of the amount of change in a target pixel corresponding to a layer in a predetermined number of layer images corresponding to the layer and captured at different time points, within a predetermined time period corresponding to the distance from the camera; The steps include detecting the specific action using the detection method with respect to the layer image taken at a predetermined time corresponding to the layer whose average value is greater than or equal to a predetermined value, A monitoring support method that performs this task.

Images