JP2011248513A - Busyness detection system and busyness detection program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technology useful for business management and the like which digitizes and detects changes such as a movement or a standstill of an object in a filming range area by using a moving image filmed by a fixed camera.SOLUTION: A busyness detection system 10 comprises: a learning data creation unit 12 for creating learning data 51 including a pixel value histogram of each area in a frame based on the moving image (a plurality of frames) of filming range 45 filmed by a monitoring camera 30; a detection target data creating unit 13 for creating detection target data 52 including the pixel value histogram of each area in the frame; a distance calculation unit 14 for calculating a distance value for each area comparing the learning data 51 and the detection target data 52; and a busyness information calculating unit 15 for calculating a degree of busyness which shows a level of busyness by the movement of the object in the filming range area based on the distance value of each area, and detecting busy spots based on the degree of busyness.

Description

  The present invention relates to a technique for performing data information processing using a moving image captured by a fixed camera such as a surveillance camera, and in particular, such as movement or stillness of an object in an area of a moving image captured in a store or the like. The present invention relates to a technology that is effective when applied to a system and program for detecting busy points from fluctuations.

  Currently, surveillance cameras and surveillance camera systems (hereinafter sometimes referred to simply as “surveillance cameras”) have been installed for various purposes such as crime prevention in various facilities such as convenience store stores. Yes. Surveillance cameras continue to be highly functional and high performance, and cameras having various functions are provided to the market.

  As a function of the surveillance camera, in addition to the basic function of shooting and recording images inside and outside the facility (moving images), as an additional function, a moving object is selected based on image analysis processing on the captured moving images. There are a function to detect and a function to detect a specific object such as a person or a face based on the detected feature of the moving object.

  Japanese Patent Laying-Open No. 2010-15465 (Patent Document 1) describes a system that measures the number of visitors to a store or the like using image data of a surveillance camera.

  Japanese Patent Laid-Open No. 2007-74330 (Patent Document 2) analyzes whether or not a customer has a willingness to purchase, whether or not a suspicious person, etc. It describes the technology to recognize.

  Japanese Patent Laying-Open No. 2006-197226 (Patent Document 3) describes a surveillance camera system that can be used not only for crime prevention purposes but also for store management. The use of a fixed camera and a movable camera is described.

JP 2010-15465 A JP 2007-74330 A JP 2006-197226 A

  Conventional surveillance camera systems are often installed and used for crime prevention purposes. For security purposes, it is possible to roughly grasp a person in a store and is effective. As a purpose other than the crime prevention purpose, it may be used for the purpose of business management or the like in a facility such as a convenience store store (for example, Patent Documents 1 to 3), but the utilization is insufficient.

  For example, in a conventional surveillance camera system, the possibility of quantifying and detecting changes in movement or stillness of a person or the like in an imaging range area to be useful for business management has not been sufficiently studied and realized. For example, a system that detects a busy spot or the like due to a change in an object in a background area (for example, an arrangement configuration of products in a store) corresponding to a shooting range has not been realized.

  In addition, by enhancing the functions such as image analysis provided in conventional surveillance camera systems, for example, it is conceivable to detect the flow lines of people in the area with high accuracy and use it for business management. There is a disadvantage that the cost increases.

  In view of the above, an object of the present invention is to manage a moving image captured by a fixed camera such as a surveillance camera by quantifying and detecting movement or stationary variation of an object such as a person in a shooting range area. It is to provide a system that can be used for management and the like. In particular, it is to provide a technique that can detect a busy place in a background area corresponding to an imaging range with reduced time and cost. The above and other objects and novel features of the present invention will be apparent from the description of this specification and the accompanying drawings.

  To achieve the above object, a typical embodiment of the present invention has the following characteristics. This busy detection system is a system that is built in or connected to a fixed camera system including a fixed camera that captures a predetermined shooting range, and that detects busyness due to changes in objects in the shooting range area. Based on a moving image or a still image, a frame processing unit that acquires a plurality of continuous still image frames and divides each frame into a plurality of areas, and a reference first corresponding to the shooting range Based on a plurality of frames, a first data creation unit that performs processing for creating first data including information of each pixel in each area of the frame, and a detection target corresponding to the imaging range Based on each frame, a second data creation unit that performs processing for creating second data including information on each pixel in each area of the frame, and the first data Distance calculation that compares pixel information of each area of the second data with pixel information of each area and calculates a distance value indicating the similarity between the first and second data for each area And the second data, based on the distance value for each area, for each area, calculating a busyness indicating the degree of busyness due to object fluctuations in the area of the shooting range, and for each area A busy information calculation unit that detects a busy point in the imaging range based on the busyness. The present invention can also be applied to a program that causes a computer to function as the system as described above.

  According to a typical embodiment of the present invention, a moving image captured by a fixed camera such as a surveillance camera is used to quantify and detect fluctuations in movement and stillness of an object such as a person in an imaging range area. It is possible to provide a system that can be used for business management. In particular, it is possible to detect a busy place in the background area corresponding to the photographing range with a small required time and cost.

It is the figure which showed the outline | summary about the structural example of the surveillance camera system containing the busy detection system which is one embodiment of this invention. It is the flowchart which showed the outline | summary about the example of the flow of the busy detection process in one embodiment of this invention. It is the figure which showed the example of a process of the division | segmentation into the several area in the frame for learning data in one embodiment of this invention. It is the figure which showed the processing example of calculation of distance value etc. by the comparison of the pixel value histogram of the target area between learning data and detection target data in one embodiment of this invention. It is a figure showing an example of processing of calculation of busyness in detection object data, and detection of a busy part in one embodiment of the present invention.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the surveillance camera system including the busy detection system according to the present embodiment, due to fluctuations in movement and stillness of an object from the imaging range area in a plurality of frames (learning data and detection target data) of a moving image taken by the surveillance camera. Detects the degree and location of busyness by quantifying them. Busy refers to a state in which a person or the like fluctuates due to movement or stationary with respect to the areas and time zones of the plurality of frames, and the busyness is detected in this system.

[System configuration]
FIG. 1 shows an outline of a configuration example of a monitoring camera system (fixed camera system) 100 including a busy detection system 10 according to an embodiment of the present invention. The surveillance camera system 100 includes a busy detection system 10, a surveillance camera server 20, a surveillance camera (fixed camera) 30, a store 40, and the like.

  One or more surveillance cameras 30 are installed in the store 40. The surveillance camera 30 captures a predetermined imaging range 45. Each surveillance camera 30 is connected to the surveillance camera server 20. The monitoring camera server 20 may be configured inside the store 40 or may be configured outside the store 40 via a network or the like. The busy detection system 10 is connected to the monitoring camera server 20. The busy detection system 10 may be configured as a system in the store 40 or may be configured as a system on an external network.

  The store 40 is, for example, a convenience store store. The imaging range 45 of the monitoring camera 30 is for a predetermined fixed space area such as a passage in the store 40 or a product display shelf. In the present embodiment, the photographing range 45 includes a commodity display shelf in the store 40 (FIGS. 3 and 300). A moving image shot for the shooting range 45 by the monitoring camera 30 is recorded in the monitoring camera server 20 as moving image data 23.

  The monitoring camera server 20 includes, for example, a computer system and each unit of the monitoring camera control unit 21 and the image analysis unit 22 implemented by a software program, and each data or database of the moving image data 23 and the analysis information 24.

  Accordingly, the monitoring camera server 20 performs a function of recording a moving image taken by the monitoring camera 30 as moving image data 23 and an image analysis process by the image analysis unit 22 based on the recorded moving image data. Thus, for example, it has a known function that can detect a moving person in a moving image. Analysis information 24 by the image analysis unit 22 is recorded in the monitoring camera server 20. The processing contents of the image analysis unit 22 and the like can be variably set based on the setting information 25. In the present embodiment, it is not essential that the surveillance camera server 20 has a moving object detection function, a person detection function, or the like.

  The monitoring camera server 20 controls the monitoring camera 30 by the monitoring camera control unit 21. This control may include, for example, execution / stop of shooting by the monitoring camera 30, setting or changing the shooting range 45, changing shooting conditions such as exposure, and the like. In particular, it is possible to set a desired shooting range 45 by changing the orientation of the monitoring camera 30 or the like.

  The busy detection system 10 is configured by, for example, a computer system, and is implemented by a frame processing unit 11, a learning data creation unit 12, a detection target data creation unit 13, a distance calculation unit 14, and a busy information calculation unit 15 implemented by a software program. Each unit includes each data or database of frame data 50, learning data 51, detection target data 52, and busy information 53. Accordingly, the busy information 53 in the photographing range 45 is calculated and detected from the moving images taken by the surveillance camera 30 and the surveillance camera server 20.

  Although not shown, the busy detection system 10 includes setting means for setting various information by operations of a store manager or the like, and can set a processing target frame, time, determination threshold, and the like. .

  The frame processing unit 11 receives the moving image data 23 of the surveillance camera server 20, creates a plurality of continuous still images (frames) from the moving image as frame data 50, and stores them. The frame processing unit 11 may perform area division processing described later.

  The learning data creation unit 12 performs processing for creating learning data (reference data) 51 based on the frame data 50. The learning data (reference data) 51 is data such as a frame that is a comparison source with the detection target data 52, and a static background region in a state where there is no person (a state with little fluctuation) in the photographing range 45, for example. This is data based on the contents of shooting, sampling, and learning.

  The detection target data creation unit 13 performs processing for creating the detection target data 52 based on the frame data 50. The detection target data 52 is data such as a frame at an arbitrary time as a target for detecting the busy information 53.

  The distance calculation unit 14 and the busy information calculation unit 15 perform processing for calculating and detecting the busy information 53 by comparison using the learning data 51 and the detection target data 52 or the like. Detailed processing contents of each part will be described later (FIG. 2 and the like).

  The busy information 53 is information obtained as a result of the processing of the busy detection system 10. The busy information 53 includes information on the detected busy points in the area of the shooting range 45 of the monitoring camera 30. The busy information 53 may be in a format that can be displayed and confirmed on a screen of a computer terminal, for example. Thereby, for example, the manager of the store 40 can determine which part of the imaging range 45 is busy. For example, it can be known that a specific product portion on the product display shelf associated with the shooting range 45 is busy, and can be used for business management such as purchase of products.

  In the present embodiment, an example in which the busy detection system 10 is implemented as a device / system independent of the monitoring camera server 20 is shown. However, the programs and data constituting each part of the busy detection system 10 are monitored. It may be incorporated in the camera server 20 so as to be integrated with the monitoring camera server 20.

[Processing content]
FIG. 2 shows an outline of an example of a flow of processing when calculating and detecting a busy place or the like based on a captured image of the monitoring camera 30 in this system (10, 100). For example, based on an operation on a computer terminal screen by a store manager or the like, the busy detection system 10 is subjected to (A) learning data creation processing (S10, S20) and (B) busy detection processing (S10, S30 to S50). Are instructed and executed. If the process (A) has already been completed and the learning data 51 exists, the execution of the process (A) can be omitted.

  (1) The frame processing unit 11 of the busy detection system 10 inputs and acquires the moving image data 23 corresponding to the shooting range 45 shot by the monitoring camera 30 from the monitoring camera server 20, and this moving image (for example, MPEG or the like) A frame process (S10) is performed to divide the difference data) into a plurality of continuous individual frames (still images). A plurality of frames are held as frame data 50. Each frame data includes data information such as a gradation value (color) for each pixel.

  A frame 300 in FIG. 3 shows an example of a frame corresponding to a predetermined shooting range 45, and includes a product display shelf and its passage as the background area 310. For ease of understanding, a frame when the product display shelf is viewed from the front is illustrated, but depending on the position and orientation of the monitoring camera 30, for example, the frame is obliquely above the passage.

  (2) The learning data creation unit 12 uses the frame data 50 to perform processing (S20) for creating the learning data 51 for the shooting range 45 (same as the detection range 45 of the detection target) by the monitoring camera 30. . This process (S20) is described in detail below (S21, S22).

  (S21) The learning data creation unit 12 (or the frame processing unit 11) divides each frame region of a plurality of still images to be acquired from the learning data 51 into a plurality of areas (vertical and horizontal), respectively. (Divided area) R is formed. This area R unit is used in the subsequent processing.

  FIG. 3 shows an example of division into a plurality of areas R in the frames (300, 301) for acquiring the learning data 51. In this example, the frame 301 is divided into 4 × 6 in length × width and divided into 24 rectangular areas R {R1-1 to R4-6}.

  The method of division into the area R is variable according to the setting or the like. For example, the detection sensitivity of the busy information 53 can be changed by adjusting the size of the area R and the number of divisions. Further, for example, area division may be set in association with a configuration such as product arrangement in the store 40 in the shooting range 45.

  (S22) Next, the learning data creation unit 12 sets a plurality of frames 301 for the learning data 51 as subsequent processing targets. Here, as a method of determining a plurality of frames used for the learning data 51, a continuous plurality of frames with a small variation (a continuous plurality of frames not including a portion with a large variation) is set. In simple terms, the static background region 310 in a state in which there is no person (a state in which fluctuations are small) may be imaged / sampled as the imaging range 45.

  It should be noted that the learning data 51 is subjected to a process of excluding a portion (area R, frame, etc.) with large fluctuations based on an appropriate photographed image (frame data 50) by automatic processing in the learning data creation unit 12. A plurality of frames may be configured. For example, based on the frame data 50, an area R in which a change in pixel value between frames (time axis) is larger than a certain level may be removed as inappropriate. Also, for example, for each color (gradation value) data of the histogram, statistical values such as average and standard deviation are acquired, and a part that clearly protrudes and changes too much is excluded, and another moving image is recaptured. May be. In order to make the comparison possible, learning data 51 having information on all areas R in the frame is created.

  Next, the learning data creation unit 12 creates pixel value histogram data to be the learning data 51 for each area R for a plurality (m) of frames 301 for the learning data 51. In the present embodiment, it is assumed that the learning data 51 is an average value over a plurality of (m) frames in a histogram (HA) of pixel values for each area R. The number (m) of frames to be sampled varies depending on the frame rate of the moving image data 23, but is about 100 to 200, for example.

  In FIG. 4, the left side (A) shows an example (401) of the pixel value histogram HA in one target area R example in the acquisition frame 301 of the learning data 51. In the pixel value histogram HA (401), the horizontal axis represents pixel values (gradation values, etc.), and the vertical axis represents frequency values.

  (3) The detection target data creation unit 13 performs a process of creating the detection target data 52 using the frame data 50 for the shooting range 45 by the monitoring camera 30 (S30). This process (S30) is described in detail below (S31, S32). As the timing of creation, after the creation of the learning data 51, for example, a store manager or the like designates start / end of photographing, and photographing / sampling is performed with an arbitrary number (n) of frames as detection targets.

  (S31) The detection target data creation unit 13 (or the frame processing unit 11) divides the target frame into a plurality of areas R as in S21. Then, each area R (all areas) of each frame is set as an area R to be processed later.

  (S32) Next, the detection target data creation unit 13 creates pixel value histogram data to be the detection target data 52 for each area R for each of a plurality of (n) frames (detection target frames). In the present embodiment, the detection target data 52 is obtained by calculating a histogram (HB) of pixel values for each area R for each frame.

  An example (402) of the pixel value histogram HB in one target area R example in the acquisition frame 302 of the detection target data 52 is shown on the right side (B) of FIG.

  As a modified example, instead of using all gradation (color) data as the histogram (HA, HB), a part of gradation (color) data may be set as a target.

  (4) The distance calculation unit 14 uses the learning data 51 (HA) in S20 and the detection target data 52 (HB) in S30 as inputs, and compares each frame of the detection target data 52 with the learning data 51. Thus, the process of calculating the distance value L, the first index value M, and the like (S40). This process (S40) is described in detail below (S41, S42).

  The distance value L is a value indicating the degree of similarity when the detection target data 52 in the image of the area R (moving image) is compared with the learning data 51. In the present embodiment, the distance value L is a pixel value histogram (HA, HB). Calculate by comparison.

  The first index value M is an index indicating the degree of variation such as movement and stillness of the object when the detection target data 52 in the image (moving image) of the area R is compared with the learning data 51 (background region 310, etc.). And is indexed by a binary or multi-valued method.

  (S41) The distance calculation unit 14 compares the learning data 51 (HA) and the detection target data 52 (HB), and calculates the distance value L. The distance value L is calculated by taking a difference of frequency values for each gradation value in the data of each histogram (HA, HB) and calculating the distance value L from the total amount or average value of the difference values. Decide.

  In FIG. 4, as an example of calculating the distance value L in S41, in the present embodiment, in comparison of the histograms (HA, HB), the frequency values of the two are obtained for each gradation value (horizontal axis) between the two. The difference is calculated, the difference values are accumulated in all gradations, and the accumulated value is set as a distance value L between the histograms HA and HB. The smaller L is, the higher the similarity is.

  (S42) The distance processing unit 14 calculates the first index value M from the distance value L in S41. In the present embodiment, a binary index value M (M = 0 or 1) is obtained by using a binary method and comparing the distance value L in S41 with a predetermined threshold value Lt. When the distance value L is greater than or equal to the threshold value Lt (L ≧ Lt), M = 1 (generally indicates variation), and when L is less than Lt (L <Lt), M = 0 (approximately Indicates no fluctuation).

  In the case of using a multi-value method, the distance value L is converted into a multi-value value M (for example, 0 ≦ M ≦ 1) by calculating a predetermined function or the like. Further, the distance value L may be used as it is in the subsequent processing by making the first index value M without conversion.

  When the learning data 51 (HA) takes a value such as an average / standard deviation, the following processing example is used. In the histogram HA, a range of [average value (a)] ± [standard deviation (σ)] × [specific multiple (b)] is used as a comparison source. The distance value L and the index value M are determined (determined) depending on whether or not the histogram HB to be detected is within the comparison source range (such as the extent). That is, the learning data 51 considers the variation in the amount of change in pixels. The above range can be set by an administrator or the like. For example, the range is a ± σ, a ± 2σ, or the like. As an example of the determination, the rate of the histogram HB that falls within the range is calculated. When the rate of the histogram that does not fall within the range exceeds a predetermined threshold, for example, 70%, M = 1, and when the rate is 70% or less, M = It is determined as 0. The threshold value (70%) can also be set by an administrator or the like.

  As the data information as a result of the processing of S40, the detection target data 52 has information (L, M) for each area R.

  When the distance value L is large (for example, M = 1), it indicates that the original image of the area R and the image to be detected are not similar, and for example, a person or the like is moving or stationary on the image of the background region 310. Can be estimated. For example, there may be a case where a person or the like is moving (passing) in front of a product on the product display shelf, a case where the person is stationary, a case where a product is being picked up, and the like.

  In the present system 10, in the process of S40, in the case of an image in the shooting range 45 of the fixed monitoring camera 30, particularly by using a binary method, the threshold Lt is set to be relatively high, so that the presence or absence of an object change There is an effect that the determination of (M = 0 or 1) is facilitated.

  (5) The busy information calculation unit 15 uses the data information (L, M) of the calculation result obtained by the distance calculation unit 14 in S40 to obtain the busy information 53 for the detection target data 52 (each frame of plural (n)). A calculation process (S50) is performed. This process (S50) is described in detail below (S51, S52). As the busy information 53, in this embodiment, the busy degree P and the busy part (the busy area R) are calculated.

  The busyness P is a value obtained by indexing the degree of busyness in the area R unit and the time zone T unit based on the first index value M relating to the movement or stillness of the moving image in the shooting range 45 due to the object. is there. The time zone T is a unit of a plurality of frames with predetermined divisions in a plurality (n) of frames to be detected. The busy spot (busy area R) is information obtained by detecting a busy spot in units of area R based on information on the busy degree P in the area R and the time zone T.

  (S51) For the detection target data 52, the busy information calculation unit 15 obtains a statistical value for each of the areas R and the time zone T where the first index value M is large (the distance value L is large). Then, the busy degree P is calculated. As a method of calculating the busyness P, for example, in a unit (R × T) of the area R and the time zone T, as a statistical value regarding the first index value M, M = 1 in the time zone T (multiple frames) The frequency of appearance of the area R (multivalue in the range of 0 to 1) and the like are calculated. Then, the statistical value is defined as busyness P (multiple values in the range of 0 to 1).

  In FIG. 5, in the frame 500 of the detection target data 52 (corresponding to a plurality (n) frames, corresponding to the time zone T), the result of the first index value M (M = 0 or 1) is obtained for each area R. An example of the corresponding category is shown. Reference numeral 501 denotes an example of the processing target area R ({R2-3, R2-4, R2-6, R3-3}) in which M = 1 in a certain frame 500. R0 is an area R with M = 0, and R1 is an area R with M = 1. In S51, the busyness P is calculated by taking a statistical value (R1 appearance frequency, etc.) regarding the first index value M in the unit of area R × time zone T. For example, when an area R1 with M = 1 frequently appears (that is, when P is large), it can be estimated that a person or the like frequently moves or stops in the area R × time zone T.

  (S52) The busy information calculation unit 15 detects a busy point (for each area R and for each time zone T) based on the statistical value (busy degree P) in S51. For example, when the busyness P (0 to 1) of a certain area R × time period T is larger than the predetermined threshold value Pt (P ≧ Pt), the location is detected as a busy location (busy area R). For example, information such as the detected busy area R and time zone T is described in the busy information 53 and output / stored.

  In FIG. 5, 510 indicates an example of a busy location (busy area R) as an example of the busy information 53 of the detection result. This is a case where a specific busy area R {R2-4, R3-3} in the example of the background region 310 of FIG. 3 is detected. As a result, the store manager or the like knows that a specific part of the shooting range 45 is busy. Depending on the setting of the shooting range 45 (background area 310), the busy part is associated with a passage, a shelf, a product, or the like.

  In the present system 10, the process of S <b> 50 has an effect that a busy spot can be efficiently detected even when the resolution (pixels) in the monitoring camera 30 (the shooting range 45) is rough.

[Effects]
As described above, according to the present embodiment, by using a moving image captured by the monitoring camera 30, the movement or stationary variation of an object such as a person in the region of the imaging range 45 is defined as the busyness P or the like. It can be digitized and used for business management. In particular, it is possible to detect a busy place in the background area 310 or the like corresponding to the imaging range 45 with less time and cost required. For example, the busy area R can be detected in a form associated with the arrangement configuration of products or the like in the store 40. Therefore, for example, it can be estimated that the reason for busyness is due to the purchase or interest of the product, which can be used for store management. The busy information 53 can be used for business management of the store 40. For example, it can be used as store management know-how. Alternatively, it may be used in comparison with product inventory / sales data by the POS system.

  Note that the system according to the present embodiment does not detect a specific person or calculate a person's flow line with high accuracy, but is busy due to a change in an object in an area R or time zone T unit. Can be detected efficiently, and can be realized with low load processing and low cost.

  As mentioned above, the invention made by the present inventor has been specifically described based on the embodiment. However, the present invention is not limited to the embodiment, and various modifications can be made without departing from the scope of the invention. Needless to say.

  INDUSTRIAL APPLICABILITY The present invention can be used for a surveillance camera system that performs data information processing using a moving image photographed by a stationary camera such as a surveillance camera, and particularly for a stationary camera system for the purpose of store management and the like.

DESCRIPTION OF SYMBOLS 10 ... Busy detection system, 11 ... Frame processing part, 12 ... Learning data creation part, 13 ... Detection object data creation part, 14 ... Distance calculation part, 15 ... Busy information calculation part,
20 ... surveillance camera server, 21 ... surveillance camera control unit, 22 ... image analysis unit, 23 ... moving image data, 24 ... analysis information, 25 ... setting information,
30 ... surveillance camera,
40 ... store, 45 ... shooting range,
50: Frame data, 51: Learning data, 52: Detection target data, 53: Busy information.

Claims (8)

Built-in or connected to a fixed camera system that includes a fixed camera that captures a predetermined shooting range, and is a system that detects busyness due to changes in objects in the area of the shooting range,
A frame processing unit that acquires a plurality of continuous frames of still images based on a moving image or a still image captured by the fixed camera and performs a process of dividing each frame into a plurality of areas;
A first data creation unit that performs processing for creating first data including information on each pixel in each area of the frame based on the first plurality of frames corresponding to the shooting range;
A second data creation unit that performs processing for creating second data including information on each pixel in each area of the frame based on each frame to be detected corresponding to the imaging range;
The distance indicating the similarity between the first and second data for each area by comparing the pixel information of each area of the second data with the pixel information of each area of the first data A distance calculator for calculating the value;
For the second data, based on the distance value for each area, for each area, a busyness indicating the degree of busyness due to object fluctuations in the area of the shooting range is calculated, and the busyness for each area is calculated. A busy information calculation unit for detecting a busy point in the area of the photographing range. In the busy detection system according to claim 1,
The first data creation unit performs processing for creating the first data including a statistical value of a histogram of pixel values for each area in the first plurality of frames,
The second data creation unit performs a process of creating the second data including a histogram of pixel values for each area in each frame,
The distance calculation unit compares the histogram of the pixel value of each area of the second data with the statistical value of the histogram of the pixel value of each area of the first data, and takes the difference between the two Or calculating a distance value for each of the areas by determining a degree of the latter falling within a range based on the former statistical value. In the busy detection system according to claim 1,
The distance calculation unit calculates a first index value relating to the variation of the moving object as a binary value for each area by comparing with a predetermined threshold based on the distance value for each area,
The busy information calculation unit calculates the busyness for each area based on a first index value for each area. In the busy detection system according to claim 1,
The distance calculation unit calculates a first index value related to the variation of the moving object as a multi-value for each area based on the distance value for each area,
The busy information calculation unit calculates the busyness for each area based on a first index value for each area. In the busy detection system according to claim 1,
The busy information calculation unit calculates the busyness by taking a statistical value for each area and for each time period using the distance value for each area or the first index value calculated from the distance value. Then, based on the busyness for each area and for each time zone, an area and a time zone that are busy locations in the area of the imaging range are detected. In the busy detection system according to claim 5,
The busy information calculation unit uses the distance value for each area or the first index value calculated from the distance value, and the statistical value is an area where the distance value or the first index value is large in the time zone. The busyness detection system characterized by calculating the said busyness degree by calculating the frequency of appearance of. In the busy detection system according to claim 1,
The busy location is a specific location in the store by setting the configuration of the spatial arrangement in the store in the shooting range and the configuration of a plurality of areas by dividing the frame. A busy detection system characterized by being detected as an area associated with the. A program that is built in or connected to a fixed camera system including a fixed camera that captures a predetermined shooting range, and that causes the computer to function as a system that detects busyness due to changes in objects in the shooting range area,
Based on a moving image or still image captured by the fixed camera, obtaining a plurality of continuous still image frames, and performing a process of dividing each frame into a plurality of areas;
Performing a process of creating first data including information of each pixel in each area of the frame based on a first plurality of frames corresponding to the imaging range as a reference;
Performing a process of creating second data including information of each pixel in each area of the frame based on each frame to be detected corresponding to the shooting range;
The pixel value information of each area of the second data is compared with the pixel value information of each area of the first data, and the similarity between the first and second data is calculated for each area. Calculating a distance value to indicate;
For the second data, based on the distance value for each area, for each area, a busyness indicating the degree of busyness due to object fluctuations in the area of the shooting range is calculated, and the busyness for each area is calculated. And a step of detecting a busy point in the area of the imaging range.