JP2008217690A - Moving object detection method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a moving object detection method for detecting a moving object by simply determining a center of gravity of the moving object at high speed and tracking the determined center of gravity. <P>SOLUTION: An entrance/exit passage 12 where a passer-by 18 of an automatic ticket gate 10 through which the moving body moves is picked up by a moving body imaging camera 17, a specific figure is located at a part corresponding to the passer-by 18 in the picked up image data, determines a center of gravity of the passer-by 18 from two centers of gravity by scanning the located figure in the moving direction of the moving objects and in its orthogonal direction, and tracks its center of gravity. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a moving body detection method, and more particularly, to a moving body detection method for detecting a moving body by quickly and easily determining the center of gravity of a moving body and tracking the determined center of gravity.

  In order to understand the number of customers in stores and halls where there are many visitors, and to detect visitors in the passages and management areas of automatic ticket gates, in the people counting device, human detector, monitoring device etc. that process them Various methods for detecting the movement of a moving body are used. As an example of a conventional method of detecting a moving body, the passing number measuring device in the “passing number measuring device and entrance / exit number management system using the same” disclosed in Patent Document 1 is used as a passing number measuring device. By depicting the movement of a visitor on spatial coordinates with a plurality of imaging means provided, and capturing the movement of the person in three dimensions, the movement of the person overlapping in the depth direction can be accurately tracked, A person was detected. Furthermore, in the entrance / exit number management system using the passing number measuring device, the number of persons entering / leaving the entrance / exit is counted based on the movement of the person detected by the passing number measuring device.

In order to track the movement of the moving body in the passing person counting apparatus as described above, the processing step of combining the images obtained from the plurality of imaging means, the movement of the moving body on the spatial coordinates based on the combined images A processing step for rendering, a processing step for tracking the movement of the moving body in three dimensions on the space coordinates, and the like are required.
Japanese Patent Laid-Open No. 10-49718

  However, in the method for detecting a moving body as described above, many processing steps and processing time are required in the process of processing images input from a plurality of imaging means. Further, in order to be able to accurately detect a moving body that is overlapped in the depth direction, a high-performance moving body detection device that includes a plurality of imaging means and the like is required. In particular, if an attempt is made to detect the movement of a moving body that moves in a passage in a short time, such as an automatic ticket gate, with a low-performance device, it is difficult to process the image with high accuracy. There was a problem that the movement could not be detected or the processing took a long time.

  In view of the above-described problems, the object of the present invention is to quickly and easily determine the center of gravity of a moving object even in a low-performance moving object detection device such as an automatic ticket gate. The object of the present invention is to provide a moving object detection method for determining and detecting a moving object by tracking the determined center of gravity.

  The moving body detection method according to the present invention is configured as follows in order to achieve the above object.

  A first moving body detection method according to the present invention (corresponding to claim 1) includes moving body imaging means for capturing an image of a path along which the moving body moves, and the center of gravity of the moving body imaged by the moving body imaging means. In a moving body detection method for detecting a moving body by tracking a pixel, a pixel of an image of the captured moving body is divided, and a specific figure is arranged in a portion corresponding to the moving body of the divided pixels. The moving figure is scanned in the direction in which the moving body moves to determine the first line of the moving body and the second line of the moving body by scanning in the vertical direction thereof, and the determined first line The center of gravity of the moving object is determined from the second line.

  According to the above moving object detection method, a specific figure is placed in a portion corresponding to the moving object in the captured image and the moving object is depicted as an object, and the object is determined from the moving direction of the moving object and its vertical direction. Since the center of gravity of an object can be determined quickly and easily by combining two lines by scanning each other, a moving object can be detected without much effort and time by tracking the center of gravity even with a low-performance detector. It becomes possible to do.

  The second moving body detection method (corresponding to claim 2) is preferably characterized in that, in the first moving body detection method, a specific figure is not arranged in a portion not corresponding to a pixel passage. . According to the above moving body detection method, no specific figure is arranged in a portion that does not correspond to the path of the pixel, so that the center of gravity of the object can be determined easily at high speed.

  The third moving body detection method (corresponding to claim 3) is preferably the above-described second moving body detection method, in which, when determining the first line and the second line, the specific figure is When the number of pixels in which a specific figure is not arranged between the arranged pixels exceeds a predetermined number, it is also determined that a plurality of moving bodies exist in the passage. According to the above moving body detection method, it is possible to determine that there are a plurality of moving bodies in the passage while determining the center of gravity.

  In the fourth moving body detection method (corresponding to claim 4), in the second moving body detection method, preferably, when the first line and the second line are determined, the specific figure is When the number of arranged pixels exceeds a predetermined number, it is also determined that a plurality of moving bodies exceeding the predetermined number of passages are present in the passage. According to the above moving body detection method, it is possible to determine that there are moving bodies exceeding a predetermined number of paths in the path while determining the center of gravity.

  According to the present invention, a specific figure is placed in a portion corresponding to a moving body of an image obtained by capturing the moving body and the moving body is depicted as an object, and the object is represented by the moving direction of the moving body and its vertical direction. Since it is possible to determine the center of gravity of an object at high speed and easily by combining two lines by scanning each, especially when tracking the center of gravity with a low-performance detection device, many processing steps and processing time are required. It becomes possible to detect a moving body without necessity.

  DESCRIPTION OF EMBODIMENTS Preferred embodiments (examples) of the present invention will be described below with reference to the accompanying drawings.

  An embodiment of an automatic ticket gate to which the present invention is applied will be described with reference to FIGS. 1 and 2. FIG. 1 is a diagram showing an automatic ticket gate to which the present invention is applied, and FIG. 2 is a block diagram showing the configuration of the ticket gate body of the automatic ticket gate.

  The automatic ticket gate 10 includes a ticket gate body 11, an entrance / exit passage 12, a ticket ticket slot 13, a ticket ticket outlet 14, a guidance display 15, an opening / closing door 16, and a moving body imaging camera 17. The entrance / exit passage 12 is a passage for the passerby 18 to enter in the direction of the arrow A indicating the traveling direction. Also, the direction opposite to the arrow A is the direction of participation. The moving body imaging camera 17 is installed in the direction of the entrance / exit passage 12, and images the passerby 18 as a moving body. In this embodiment, the moving body imaging camera 17 is installed at the upper right on the entrance side of the entrance / exit passage 12. This is so that even when there are a plurality of passers-by 18 passing through the entrance / exit passage 12, the passers-by 18 can be photographed with the least overlap. Further, the installation position of the mobile camera 17 is not limited to this as long as the passerby 18 can shoot without overlapping.

  The ticket gate main body 11 is input from the control unit 19 that controls the entire automatic ticket gate 10, the image data input unit 20 that inputs image data captured by the moving body imaging camera 17, and the image data input unit 20. A mobile object generation unit 21 that generates a moving object of image data, an object gravity center determination unit 22 that determines the center of gravity of an object generated by the mobile object generation unit 21, and an object determined by the object gravity center determination unit 22 An object center of gravity tracking unit 23 that tracks the center of gravity; a moving body detection unit 24 that detects a moving body based on the center of gravity tracked by the object center of gravity tracking unit 23 and outputs a moving body detection result; a monitor 25 that displays image data; A key for initial setting when starting the operation of the automatic ticket gate 10 described later with reference to the monitor 25 Consisting of head 26.

  First, an initial setting method when starting the operation of the automatic ticket gate 10 will be described with reference to FIG. FIG. 3 is an image of image data processed at the time of initial setting of the automatic ticket gate main body 11.

  3A shows an image of the image data of the entrance / exit passage 12 of the automatic ticket gate 10 imaged by the moving body imaging camera 17, and FIG. 3B shows the image data in which a mask object to be described later is set. FIG. 3C shows an image of image data in which a moving body moving direction 29 indicating the direction in which the passerby 18 moves through the entrance / exit passage 12 is set.

  In the initial setting, first, the entire image data of the entrance / exit passage 12 as shown in FIG. 3A is input from the moving body imaging camera 17 to the image data input unit 20, and a passerby 18 moving through the entrance / exit passage 12. Is adjusted so that the imaging direction of the moving body imaging camera 17 is adjusted. Next, a plurality of objects 27 divided at an arbitrary interval as shown in FIG. 3B, the image data in which the mask object 28 is set in a portion not corresponding to the entrance / exit passage 12 of the ticket gate body 11 is stored. Generate. The portion where the mask object 28 is set is not detected by the passerby 18. The image data in which the mask object 28 is set serves as reference image data for detecting the passerby 18. A method for detecting passers-by will be described later. Subsequently, image data in which a moving body moving direction 29 indicating the direction in which the passerby 18 moves as shown in FIG. 3C is set is generated.

  In order to set the mask object 28 and the moving body moving direction 29, the keypad 26 is operated while referring to the image data displayed on the monitor 25 inside the ticket gate body 11 of the automatic ticket gate 10. Set. Alternatively, an input / output device equipped with a monitor 25 and a keypad 26 is installed in a place where a station staff member in charge of the ticket gate where the automatic ticket gate 10 is installed, and the mask object 28 and the moving object move through them. Direction 29 may be set. Further, the mask object 28 and the moving body moving direction 29 may be automatically set by the control unit 19 without using the monitor 25 and the keypad 26.

  Next, a method for generating a moving object from image data will be described with reference to FIG. FIG. 4 is an image of the image data processed when the mobile object of the automatic ticket gate main body is generated.

  4A shows an image of the image data of the entrance / exit passage 12 where no passerby 18 exists, and FIG. 4B shows the generation of the moving object at the time of FIG. 4A. It is image data that is sometimes processed, that is, the same image data as the reference image data described above. FIG. 4C shows an image of the image data of the entrance / exit passage 12 where a passerby 18a that has passed through the entrance / exit passageway 12 and a passerby 18b that starts to pass through the entrance / exit passageway 12 exist. FIG. 4D shows image data processed at the time of generating the moving object in FIG.

  In the image data of FIG. 4D, the graphic object 30 is set in the object 27 corresponding to the passers-by 18a and 18b in the portion corresponding to the entrance / exit passage 12 where the mask object is not set, and the moving object 31a and 31b are depicted. The mobile object 31a is an object corresponding to the passerby 18a, and the mobile object 32b is an object corresponding to the passerby 18b. In accordance with the accuracy of detecting the moving object, the graphic object 30 may be set to the object 27 corresponding to the passersby 18a and 18b by a predetermined ratio (90% as an example). In this embodiment, since the object 27 of the image data picked up from the mobile imaging camera 17 is divided into squares, the graphic object 30 uses a quadrangle of the same size, but the passersby 18a and 18b. If the graphic object 30 can be described by the graphic object 30, the graphic object 30 is not limited to a square, and the mobile objects 31a and 31b may be generated using a graphic object 30 such as a square or a circle having an appropriate size.

  Next, a method for determining the center of gravity of the moving object will be described with reference to FIGS. FIG. 5 is an image of image data processed at the time of determining the center of gravity of the mobile object of the automatic ticket gate, FIG. 6 is an image showing the distribution of graphic objects, and FIG. 7 is the mobile object of the automatic ticket gate. It is the 2nd image of the image data processed at the time of determination of the gravity center. FIG. 8 is a third image of the image data processed when determining the center of gravity of the mobile object of the automatic ticket gate main body, and FIG. 9 is a second image representing the distribution of the graphic objects.

  The image data shown in FIG. 5A shows a moving body moving direction 29 and a moving body moving vertical direction 32 indicating a direction perpendicular thereto. When an object is scanned in the moving body moving direction 29 as shown in FIG. 5B, an image of the distribution of the number of graphic objects in the line in the moving direction of the moving body is shown as shown in FIG. Can do. In FIG. 6A, the horizontal axis indicates the number of scanning lines, and the vertical axis indicates the number of graphic objects for each line. The figure shows the number of graphic objects existing on the scanned line by arranging the graphic objects 30 while scanning the object in the moving body moving direction 29. Subsequently, when the object is scanned in the moving body moving vertical direction 32 as shown in FIG. 5C, the distribution of the number of graphic objects in the line in the moving body moving vertical direction as shown in FIG. 6B. Can represent the image. Similar to FIG. 6A, the horizontal axis of FIG. 6B indicates the number of scanning lines, and the vertical axis indicates the number of graphic objects for each line.

  In the image of the distribution of the graphic objects in the moving direction of the moving body shown in FIG. 6A, the moving body objects 31a and 31b are shown. Then, based on the number of predetermined blank lines 35 set in advance, the first moving object 31a and the second moving object 31b are separated. Between the first moving object 31a and the second moving object 31b, there is a portion where no graphic object exists on the scanning line. This corresponds to an interval between passers-by passing through the entrance / exit passage 12 of the automatic ticket gate 10. A graphic object exists in the scan line corresponding to the passerby, but no graphic object exists in the scan line corresponding to the interval between the passers-by. What is necessary is just to set the number of the predetermined blank lines 35 according to the space | interval of passers-by. For example, since the distance between passers-by is relatively narrow in the entrance / exit passage 12 of the automatic ticket gate 10, the number of predetermined blank lines 35 may be set to be small accordingly. Then, the graphic objects 30 are counted in the order in which the graphic objects 30 constituting the moving object 31a are arranged, and the center (the total number of the graphic objects 30 constituting the moving object 31a / 2 [th]). The line in which the graphic object 30 exists is determined as the object barycentric line 33a in the moving direction of the moving object. The moving body is represented by a collection of graphic objects 30, and a scanning line with a relatively large number of graphic objects 30 is determined as an object barycentric line 33a in the moving direction of the moving object. A moving body passing therethrough can be detected at high speed and easily. In the same way, the line in which the graphic object 30 at the center (the total number of graphic objects 30 constituting the moving object 31b / 2 [th]) constituting the moving object 31b exists is moved in the moving direction of the moving object. The center of gravity line 33b is determined.

  Further, the moving object 31c is shown in the image of the distribution of the graphic objects in the moving vertical direction shown in FIG. 6B. In FIG. 6A described above, since the graphic object is scanned in the moving direction of the moving body, the two passers-by are shown as the moving body objects 31a and 31b. In (B), since the graphic object is scanned in the vertical direction of the moving body, two passersby are not separated into the two moving body objects 31a and 31b, but one movement in which the moving body objects 31a and 31b overlap each other. It is shown by a body object 31c. Both the mobile object 31a, 31b and the mobile object 31c are the same two mobile objects. Accordingly, the total number of the graphic objects 30 constituting the moving object 31a and the number of the graphic objects 30 constituting the moving object 31b is the same as the number of the graphic objects 30 constituting the moving object 31c. Become. In the same manner as described above, the moving object moves vertically on the line where the graphic object 30 in the center (the total number of graphic objects 30 constituting the moving object 31c / 2 [th]) constituting the moving object 31c exists. It determines with the object gravity center line 33c of a direction. The object center-of-gravity line 33c is common to the two passers-by moving objects 31a and 31b. When the object barycentric lines 33a and 33b are represented on the image data shown in FIG. 5B, the image data shown in FIG. 7A is obtained, and the object barycentric line 33c is displayed in the image shown in FIG. When represented on the data, the image data is as shown in FIG. FIG. 7A shows object center-of-gravity lines 33a and 33b in the moving direction of the moving object obtained by scanning the object in the moving direction of the moving object, and FIG. 7B shows the vertical movement of the moving object. A direction object barycentric line 33c is shown.

  Then, in the image data as shown in FIG. 7C in which the image data shown in FIG. 7A and the image data shown in FIG. 7B are combined, the object in the moving direction of the moving object is shown. The graphic object 30 in which the center of gravity and the center of gravity of the object in the vertical direction overlap is defined as the center of gravity 34a of the moving body 18a or the center of gravity 34b of the moving body 18b. When the actual image captured by the moving body imaging camera 17 and the obtained center of gravity are combined, image data as shown in FIG. 7D is obtained.

  8A and 8B is image data taken immediately after FIG. 7D, and the passerby 18a will soon pass through the entrance / exit passage 12. In addition, the passerby 18b moved to approximately the center position of the entrance / exit passage 12. When these image data are scanned in the moving direction of the moving body and the moving body in the vertical direction by the method described above as shown in FIG. 8C and FIG. 8D, FIG. , An image of the distribution of the number of graphic objects in the moving direction and vertical line of the moving object as shown in FIG.

  By the method described above, the center of gravity line 33b in the moving direction of the moving object 31b as shown in FIG. 8C and the moving vertical direction of the moving objects 31a and 31b as shown in FIG. Centroid line 33c. The image data shown in (C) of FIG. 8 and the image data shown in (D) of FIG. 8 are combined to determine the center of gravity 34c of the moving object as shown in the image data of (E) of FIG. .

  Since the number of graphic objects 30 constituting the moving object 31b shown in FIG. 9A does not satisfy a predetermined number, the moving object 31b is a noise object that is not detected as a moving object. Therefore, the center of gravity of the moving object 31b is not determined. As shown in FIG. 8F, the passerby 18a will soon pass through the entrance / exit passage 12, so that the center of gravity 34a of the moving object of the passerby 18a is lost. Further, the center of gravity 34b of the moving object of the passerby 18b has moved to the center of gravity 34c of the moving object.

  In addition, a process when the center of gravity of the moving object is not determined will be described with reference to FIG. FIG. 10 shows a fourth image of the image data processed when determining the center of gravity of the moving object of the automatic ticket gate main body.

  As shown in FIG. 10A, there are cases where there are more passers-by 18d-18f than normal people in the ticket gate passage 2, and image data in which passers-by 18d-18f overlap is input. When this is depicted by the graphic object 30, it becomes image data in which a moving object 31d as depicted in FIG. 10B is depicted. When this image data is scanned for each object in the moving direction of the moving body, an image of the distribution of the number of graphic objects in the line in the moving direction of the moving body can be represented as shown in FIG. If the number of graphic objects 30 constituting the object is equal to or greater than a preset threshold value, it is determined that the passersby overlap the entrance / exit passage 12 without determining the center of gravity of the passersby.

A method for tracking the center of gravity of the passerby 18 will be described with reference to FIG.
FIG. 11 is a diagram for explaining how the center of gravity of a passerby moves.

  11A to 11F show images obtained by synthesizing actual images taken by the moving body imaging camera 17 at regular intervals and the obtained center of gravity. FIG. 11 (F) shows an image in which the detected movement of the center of gravity of the moving body is continuously depicted.

  In FIG. 11A, there is a center of gravity 34a of the passerby 18a and a center of gravity 34b of the passerby 18b. In FIG. 11B, the center of gravity 34a of the passerby 18a disappeared, and the center of gravity 34b indicated by the broken line of the passerby 18b moved to the center of gravity 34c. As shown in FIGS. 11C and 11D, the center of gravity 34c moves in the order of the centers of gravity 34d and 34e. An object whose center of gravity moves is defined as a moving object. By tracking the moving object, it can be detected that the moving body moves. In FIG. 11E, the center of gravity 34e indicated by the broken line of the passerby 18b disappears, and the center of gravity 34f of the next passerby 18c is shown. Since the center of gravity 34e and the center of gravity 34f are the center of gravity of another passerby, the coordinates are separated from each other by a predetermined distance. Since this center of gravity is newly determined, an object having this center of gravity is defined as a new object. When the moving object of the passerby 18b is tracked, as shown in FIG. 11 (F), the center moves from the center of gravity 34b to the center of gravity 34e in order.

  The overall processing flow of the moving object detection method will be described with reference to FIG. FIG. 12 is a flowchart showing an overall processing flow of the moving object detection method.

  First, when starting the operation of the automatic ticket gate 10, the mask object 28 and the moving body moving direction 29 are set as initial settings (step S101). The moving body imaging camera 17 images the passerby 18 passing through the entrance / exit passage 12 and outputs image data (step S102). The image data input unit 20 of the automatic ticket gate main body inputs the image data output from the moving body imaging camera 17 (step S103). The moving object generation unit 21 sets the graphic object 30 to the passerby 18 of the image data (step S104), and the object gravity center determination unit 22 determines the gravity center of the graphic object 30 (step S105). The object centroid tracking unit 23 tracks the passerby 18 based on the determined centroid (step S106), and the moving body detection unit 24 outputs the moving body detection result (step S107). If the predetermined time set in advance has not been reached (NO in step S108), the predetermined time is counted (step S109). When the predetermined time is reached (YES in step S108), the process returns to step S102, and the moving body imaging camera 17 Images the passerby 18 passing through the entrance / exit passage 12, and outputs image data.

  With reference to FIG. 13 and FIG. 14, the flow of the process of turning into a passer-by object and the process of detecting the center of gravity of the object shown in steps S104 and S105 of FIG. 12 will be described. FIG. 13 is a first flowchart showing a flow of processing for converting a passer-by to an object and processing for detecting the center of gravity of the object. FIG. 14 shows a flow of processing for converting a passer-by to an object and processing for detecting the center of gravity of the object. It is a 2nd flowchart shown.

  First, the moving object generation unit 21 scans a plurality of objects 27 obtained by dividing image data at arbitrary intervals in the moving object moving direction 29 (step S201). If the object 27 is not the mask object 28 (NO in step S202) and there is a difference between the object 27 and the object 27 of the reference image data (YES in step S203), the graphic object 30 is set in the object 27. To do. If the object 27 is the mask object 28 (YES in step S202) or if there is no difference between the object 27 and the object 27 of the basic image data (NO in step S203), the graphic object 30 is added to the object 27. Without setting, the process skips to step S205. The ratio at which the graphic object 30 is set in the object 27 is calculated (step S205).

  Until all the objects 27 on the currently scanned line have been processed (NO in step S206), the process returns to step S202 and the process is repeated. If all the objects 27 on the line being scanned have been processed (YES in step S206), until all the objects 27 on the line in the moving body moving direction 29 have been processed (NO in step S207). The process returns to step S201 and is repeated.

  Then, the processing is completed for all the line objects 27 (YES in step S207). If the ratio of the graphic object 30 set to the object 27 calculated in step S205 exceeds a predetermined ratio, this process is performed. If the ratio does not exceed the predetermined ratio (NO in step S208), the difference determination threshold is lowered (step S209), the process returns to step S202, and the ratio is predetermined. The process is repeated until the ratio of is exceeded.

  If the graphic object 30 generated by scanning in the moving direction of the moving object has a predetermined number of blank lines 35 (YES in step S301), the object gravity center determination unit 22 separates the graphic object 30 (step S302). . Further, if the graphic object 30 does not have the predetermined number of blank lines 35 (NO in step S301) and the graphic object 30 exceeds the predetermined number (YES in step S303), a plurality of entrance / exit passages 12 are provided. It is determined that the passerby 18 is present (step S304), and this process ends without determining the center of gravity of the object. If the number of graphic objects 30 does not exceed the predetermined number (NO in step S303) and the number of graphic objects 30 exceeds the noise object determination threshold (YES in step S305), the object is designated as a noise object. If the number of graphic objects 30 does not exceed the noise object determination threshold value (NO in step S305), the scanning line where the center of the graphic object 30 is located is the center of gravity in the moving direction of the moving object. The line is determined (step S307).

  If the object is divided in step S302, since all the graphic objects 30 have not been processed (NO in step S308), the process returns to step S303, and all the graphic objects 30 are processed. If all the graphic objects 30 have been processed (YES in step S308), since the graphic object 30 generated by scanning the moving object in the vertical direction of movement has not been processed, the process returns to step S301 and the same as described above. Is performed (NO in step S309).

  When the processing of the graphic object 30 generated by scanning in the direction perpendicular to the moving body moving direction 29 is completed (YES in step S309), the center of gravity line of the graphic object 30 in the moving direction of the moving body and the movement of the moving body The center of gravity of the graphic object 30 is determined by combining the center of gravity line of the graphic object 30 in the vertical direction (step S310).

  With reference to FIG. 15, the flow of processing for tracking the center of gravity of a passerby will be described. FIG. 15 is a flowchart showing a flow of processing for tracking the center of gravity of a passerby.

  The moving body detection unit 24 compares the center of gravity of the passerby of the current image data with the center of gravity of the passerby of the previous image data (step S401), and the previous image data has the center of gravity of the passerby (step S402). YES), if the coordinate distance between the centroid of the passerby in the current image data and the centroid of the passerby in the previous image data is within a predetermined range, the object of the centroid in the current image data is moved. (Step S404). In addition, when the previous image data does not have the centroid of the passerby in step S402 (NO in step S402), or when the coordinate distance between the two centroids is not within the predetermined range in step S403, the current image data An object having a center of gravity starts to be tracked as a new object (step S405). By tracking the center of gravity of the object, the fact that a passerby as a moving object has been detected is output (step S406).

  The configurations described in the above embodiments are merely schematically shown to the extent that the present invention can be understood and implemented, and the numerical values are merely examples. Therefore, the present invention is not limited to the described embodiments, and can be variously modified without departing from the scope of the technical idea shown in the claims.

  For example, in the present embodiment, the mobile object is described as a passerby, but the mobile object is not limited to this. The mobile object is a product produced in a factory, and the product is tracked and the logistics management is performed. It can also be applied to product management systems. The present invention can be applied in various forms in each field of movement detection that requires rapid and accurate.

It is a figure which shows the automatic ticket gate to which this invention is applied. It is a block diagram which shows the structure of the ticket gate main body of an automatic ticket gate. It is an image of image data processed at the time of initial setting of the automatic ticket gate main body. It is an image of the image data processed when the mobile object of the automatic ticket gate main body is generated. It is an image of the image data processed when determining the center of gravity of the moving object of the automatic ticket gate main body. It is an image showing distribution of a graphic object. It is a 2nd image of the image data processed at the time of determination of the gravity center of the moving body object of an automatic ticket gate main body. It is a 3rd image of the image data processed at the time of determination of the gravity center of the moving body object of an automatic ticket gate main body. It is a 2nd image showing distribution of a graphic object. It is a 4th image of the image data processed at the time of determination of the gravity center of the moving body object of an automatic ticket gate main body. It is a figure for demonstrating a mode that the gravity center of a passerby moves. It is a flowchart which shows the flow of the whole process of a moving body detection method. It is a 1st flowchart which shows the flow of the process which detects a passerby as an object, and the process which detects the gravity center of an object. It is a 2nd flowchart which shows the flow of the process which detects a passerby as an object, and the process which detects the gravity center of an object. It is a flowchart which shows the flow of the process which tracks the gravity center of a passerby.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Automatic ticket gate 11 Ticket gate main body 12 Entrance / exit passage 13 Passenger ticket slot 14 Passenger ticket outlet 15 Information display 16 Opening / closing door 17 Moving body imaging camera 18 Passer 19 Control section 20 Image data input section 21 Moving body Object generation unit 22 Object centroid determination unit 23 Object centroid tracking unit 24 Moving body detection unit 25 Monitor 26 Keypad 27 Object 28 Mask object 29 Moving body moving direction 30 Graphic object 31a to 31c Moving body object 32 Moving body moving vertical direction 33a to 33c Object center-of-gravity line 34a-34f Center of gravity 35 Blank line A Arrow indicating the direction of travel

Claims (4)

In a moving body detection method for detecting the moving body by tracking the center of gravity of the moving body imaged by the moving body imaging means, comprising moving body imaging means for capturing an image of a path along which the moving body moves,
Dividing pixels of the image of the imaged moving body;
A specific figure is arranged in a portion corresponding to the moving body of the divided pixels,
The arranged figure is scanned in the direction in which the moving body moves to determine the first line of the moving body and the second line of the moving body by scanning in the vertical direction thereof, respectively.
Determining the center of gravity of the movable body from the determined first line and the second line;
A moving body detection method characterized by the above.   The moving body detection method according to claim 1, wherein the specific figure is not arranged in a portion of the pixel not corresponding to the passage.   When determining the first line and the second line, the number of the pixels in which the specific graphic is not disposed between the pixels in which the specific graphic is disposed exceeds a predetermined number. In this case, it is determined that there are a plurality of the moving bodies in the passage.   When determining the first line and the second line, if the number of the pixels in which the specific figure is arranged exceeds a predetermined number, a plurality of the passages exceeding a predetermined number of the passages 3. The moving object detection method according to claim 2, wherein it is determined that a moving object exists in the passage.

Images