JP2008077430A - Mobile body counting device and mobile body counting method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a new and improved mobile body counting device, capable of automatically changing setting of a detection area for detecting a mobile body, and a mobile body counting method therefor. <P>SOLUTION: The mobile body counting device 100 comprises an image acquisition part 110 acquiring an image at predetermined intervals; an image recognition part 130 recognizing a form similar to a preset form contained in the image; a position acquisition part 150 acquiring the coordinate position of the similar form contained in the image; a detection area generation part 160 generating a detection area based on the coordinate position; a mobile body detection part 170 detecting mobile bodies contained in the image; and a mobile body counting part 180 counting the number of mobile bodies contained in the detection area. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a moving body counting apparatus and a moving body counting method.

  Surveillance cameras are installed in various locations, such as offices and downtown areas in the city, and automated teller corners of financial institutions. Conventionally, surveillance cameras have been used for the purpose of issuing warnings when abnormal situations occur in the above-mentioned locations by monitoring images taken by surveillance personnel. It is used to provide various services to users by automatically detecting people, vehicles, etc. from the images and measuring the number of detections.

  An example of an object counting device using such a monitoring camera is disclosed in Patent Document 1. The technique disclosed in Patent Document 1 is a circular movement when a moving body passes through a fixed detection area in a band shape set in an image photographed by a camera installed from above to below a passage. Only an object is extracted, pattern matching processing is performed only on a moving object having the same size as a human head, it is determined whether or not it is a human, and the number of humans passing through the detection area is counted. It was.

JP-A-8-23882

  However, the method disclosed in Patent Document 1 has a problem that the detection area remains fixed unless the operator changes the setting. The number of people using the automated teller corner of a store or financial institution changes depending on the time of day or date, and the length and shape of the queue formed by people lining up at terminals such as cash registers and ATMs change accordingly. To do. However, in a state where the detection area is fixed, there is a problem that it is impossible to correctly count the moving objects passing through the detection area without being able to cope with changes in the length and shape of the queue.

  In order to solve the above problem, it is conceivable to manually reset the set detection area. In this case, the operator is burdened with resetting the detection area, or the detection area is set. There is a problem that a new problem arises that the number of operators must be secured and the cost increases.

  Therefore, the present invention has been made in view of the above problems, and an object of the present invention is new and improved, which can automatically change the setting of a detection area for detecting a moving object. Another object of the present invention is to provide a moving body counting apparatus and a moving body counting method.

  In order to solve the above problems, according to an aspect of the present invention, an image acquisition unit that acquires images at predetermined intervals, an image recognition unit that recognizes a shape similar to a preset shape included in the image, and A position acquisition unit that acquires a coordinate position of a similar shape included in the image, a detection area generation unit that generates a detection area based on the coordinate position, a mobile body detection unit that detects a mobile body included in the image, There is provided a moving body counting device including a moving body counting unit that counts the number of moving bodies included in the detection area.

  According to this configuration, the image acquisition unit acquires images at predetermined intervals, the image recognition unit recognizes a shape similar to a preset shape from the images acquired by the image acquisition unit, and the position acquisition unit recognizes the image. A coordinate position of a shape similar to the preset shape recognized by the unit, the detection area generation unit generates a detection area based on the coordinate position of the shape similar to the preset shape acquired by the position acquisition unit, The moving object detection unit detects a moving object included in the image acquired by the image acquisition unit, and the moving object counting unit includes the number of moving objects detected by the moving object detection unit included in the detection area generated by the detection area generation unit. Count. As a result, the detection area can be dynamically set by analyzing the acquired image.

  The moving body detection unit stores the current image storage unit that stores the current image acquired by the image acquisition unit, the background image storage unit that stores the image acquired by the image acquisition unit as a background image, and the current image storage unit. When the current image is compared with the background image stored in the background image storage unit and the two are different by a predetermined value or more, the background image in the background image storage unit is changed to the current image stored in the current image storage unit. A background image update unit to be updated, a difference image extraction unit that extracts a difference between the current image stored in the current image storage unit and the background image updated by the image update unit may be included.

  According to this configuration, the current image storage unit stores the current image acquired by the image acquisition unit, the background image storage unit stores the image acquired by the image acquisition unit as a background image, and the background image update unit The current image stored in the image storage unit is compared with the background image stored in the background image storage unit, and if both are different by a predetermined value or more, the background image in the background image storage unit is displayed in the current image storage unit. Updating to the stored current image, the difference image extraction unit extracts a difference between the current image stored in the current image storage unit and the background image updated by the image update unit. As a result, the moving object can be detected by extracting the difference between the acquired images.

  In order to solve the above-described problem, according to another aspect of the present invention, an image acquisition unit that acquires images at predetermined intervals, and attribute information of a reference object included in the image is acquired, and position information of the reference object An attribute information acquisition unit that acquires position information acquisition unit that acquires position of a reference object by converting position information into a coordinate position in an image, and a detection area generation unit that generates a detection area based on the position of the reference object There is provided a moving body counting apparatus comprising: a moving body detecting unit that detects a moving body included in an image; and a moving body counting unit that counts the number of moving bodies included in a detection area. .

  According to such a configuration, the image acquisition unit acquires images at predetermined intervals, the attribute information acquisition unit acquires the attribute information of the reference object included in the image, acquires the position information of the reference object, and the position acquisition unit Converts the position information of the reference object acquired by the attribute information acquisition unit to the coordinate position in the image to acquire the position of the reference object, and the detection area generation unit detects based on the position of the reference object acquired by the position acquisition unit An area is generated, the moving body detection unit detects a moving body included in the image acquired by the image acquisition unit, and the moving body counting unit is detected by the moving body detection unit included in the detection area generated by the detection area generation unit. Count the number of moving objects. As a result, it is possible to dynamically set the detection area by analyzing the acquired image and to count the moving bodies included in the detection area.

  The moving object detection unit includes a current image storage unit that stores the current image acquired by the image acquisition unit, a background image storage unit that stores a background image included in the current image stored in the current image storage unit, and a current image The current image stored in the storage unit is compared with the background image stored in the background image storage unit, and if both differ by a predetermined value or more, the background image in the background image storage unit is stored in the current image storage unit A background image update unit that updates the current image, a difference image extraction unit that extracts a difference between the current image stored in the current image storage unit and the background image updated by the image update unit. Good.

  According to this configuration, the current image storage unit stores the current image acquired by the image acquisition unit, the background image storage unit stores the image acquired by the image acquisition unit as a background image, and the background image update unit The current image stored in the image storage unit is compared with the background image stored in the background image storage unit, and if both are different by a predetermined value or more, the background image in the background image storage unit is displayed in the current image storage unit. Updating to the stored current image, the difference image extraction unit extracts a difference between the current image stored in the current image storage unit and the background image updated by the image update unit. As a result, the moving object can be detected by extracting the difference between the acquired images.

  The position information of the reference object is recorded in the electronic tag, and the attribute information acquisition unit may acquire the position information of the reference object from the reference object wirelessly. According to such a configuration, the attribute information acquisition unit wirelessly acquires the position information of the reference object recorded on the electronic tag. As a result, the detection area can be set by wirelessly reading the position information from the electronic tag.

  The attribute information is an ID number, and the detection area may be formed based on the order of the ID numbers. As a result, the detection area can be set dynamically only by changing the arrangement of the reference object.

  In order to solve the above problems, according to another aspect of the present invention, an image acquisition step of acquiring images at predetermined intervals, an image recognition step of recognizing a shape similar to a preset shape from the image, A position acquisition step of acquiring a coordinate position of a similar shape included in the image, a detection area generation step of generating a detection area based on the coordinate position, a moving body detection step of detecting a moving body included in the image, and detection And a moving body counting step for counting the number of moving bodies included in the area.

  According to this method, the image acquisition step acquires images at predetermined intervals, the image recognition step recognizes a shape similar to a preset shape from the images acquired in the image acquisition step, and the position acquisition step performs image recognition. The coordinate position of the shape similar to the preset shape recognized in the step is acquired, and the detection area generation step generates a detection area based on the coordinate position of the shape similar to the preset shape acquired in the position acquisition step, The moving body detection step detects moving bodies included in the image acquired by the image acquisition unit, and the moving body counting step includes the number of moving bodies detected in the moving body detection step included in the detection area generated in the detection area generation step. Count. As a result, the detection area can be dynamically set by analyzing the acquired image.

  In order to solve the above-described problem, according to another aspect of the present invention, an image acquisition step of acquiring images at predetermined intervals, and attribute information of a reference object included in the image are acquired, and position information of the reference object An attribute information acquisition step for acquiring the position, a position acquisition step for converting the position information into a coordinate position in the image to acquire the position of the reference object, a detection area generation step for generating a detection area based on the position of the reference object, There is provided a moving body counting method comprising a moving body detecting step for detecting a moving body included in an image and a moving body counting step for counting the number of moving bodies included in a detection area.

  According to such a method, the image acquisition step acquires images at predetermined intervals, and the attribute information acquisition step acquires attribute information of the reference object included in the image acquired in the image acquisition step, and obtains the position information of the reference object. The position acquisition step acquires the reference object position by converting the position information of the reference object acquired in the attribute information acquisition step into the coordinate position in the image, and the detection area generation step acquires the reference object acquired in the position acquisition step. A detection area is generated based on the position of the moving object, the moving body detection step detects a moving body included in the image acquired in the image acquisition step, and the moving body counting step is included in the detection area generated in the detection area generation step. The number of moving bodies detected in the moving body detection step is counted. As a result, it is possible to dynamically set the detection area by analyzing the acquired image and to count the moving objects that pass through the detection area.

  As described above, according to the present invention, there is provided a new and improved moving object counting apparatus and moving object counting method capable of automatically changing the setting of a detection area for detecting a moving object. is there.

  Exemplary embodiments of the present invention will be described below in detail with reference to the accompanying drawings. In addition, in this specification and drawing, about the component which has the substantially same function structure, duplication description is abbreviate | omitted by attaching | subjecting the same code | symbol.

(First embodiment)
First, the moving body counting device and the moving body counting method according to the first embodiment of the present invention will be described.

  FIG. 1 is an explanatory diagram for explaining a moving object counter according to a first embodiment of the present invention. Hereinafter, the moving body counter according to the first embodiment of the present invention will be described with reference to FIG.

  As shown in FIG. 1, the moving body counting apparatus 100 according to the first embodiment of the present invention includes an image acquisition unit 110, an image preprocessing unit 120, an image recognition unit 130, a shape recording unit 140, The position acquisition unit 150, the detection area generation unit 160, the moving body detection unit 170, the object number counting unit 180, and the detection result display unit 190 are configured.

  The image acquisition unit 110 acquires an image of a subject. As the image acquisition unit 110, for example, an optical device such as a camera including an image sensor such as a charge coupled devices (CCD) element or a complementary metal oxide semiconductor (CMOS) element can be used.

  The image preprocessing unit 120 performs an appropriate conversion process on the image acquired by the image acquisition unit 110 so that the image acquired by the image acquisition unit 110 can be recognized by the image recognition unit 130. The conversion processing here is a conventionally known image processing method such as reduction or removal of image noise or illumination unevenness, normalization of image shading, image binarization processing and thinning processing. Can be used.

  The image recognition unit 130 performs a recognition process on whether or not a shape similar to the shape recorded in advance in the shape recording unit 140 is included in the image subjected to the conversion process by the image preprocessing unit 120. Hereinafter, this pre-recorded shape is also referred to as “standard pattern”. Here, the “standard pattern” refers to a pattern having standard shape characteristics of an object used for partitioning a queue such as partition poles, ropes between partition poles, triangular cones, and reference markers.

  The shape recording unit 140 is for recording a standard pattern used for recognition processing in the image recognition unit 130. The image recognition unit 130 reads a standard pattern recorded in the shape recording unit 140 and includes a shape similar to a shape registered in advance for the image subjected to the conversion process by the image preprocessing unit 120. Recognition processing is performed.

  The position acquisition unit 150 acquires a coordinate position in the image having a shape similar to the standard pattern obtained as a result of the recognition processing performed by the image recognition unit 130.

  The detection area generation unit 160 sets a detection area for detecting a moving body in the image based on the coordinate position in the image having a shape similar to the standard pattern acquired by the position acquisition unit 150. . A method for setting the detection area in the image will be described later.

  The moving body detection unit 170 detects a moving body part from the subject image acquired by the image acquisition unit 110. A method for detecting the moving part will be described later.

  The object number counting unit 180 acquires the area of the detection area generated by the detection area generation unit 160 and measures the number of moving body parts existing in the detection area.

  The detection result display unit 190 displays the number of moving body parts present in the detection area, which is measured by the object number counting unit 180.

  The configuration of the moving object counting device according to the first embodiment of the present invention has been described above with reference to FIG. Next, the configuration of the moving object detection unit according to the first embodiment of the present invention will be described.

  FIG. 2 is an explanatory diagram illustrating the configuration of the moving object detection unit according to the first embodiment of the present invention. Hereinafter, the configuration of the moving object detection unit according to the first embodiment of the present invention will be described with reference to FIG.

  As shown in FIG. 2, the moving object detection unit 170 according to the first embodiment of the present invention includes a current image storage unit 171, a background image storage unit 172, a background image update unit 173, and a difference image extraction unit. 174 and a switch 175.

  The current image storage unit 171 stores the image acquired by the image acquisition unit 110 as a current image. The background image storage unit 172 stores the image acquired by the image acquisition unit 110 as a background image when the switch 175 is closed when the moving body counting apparatus 100 is activated. The background image stored in the background image storage unit 172 is preferably an image acquired in a situation where there is no moving object to be detected.

  The background image update unit 173 updates the background image stored in the background image storage unit 172. Background image update processing performed by the background image update unit 173 will be described later.

  The difference image extraction unit 174 acquires the difference between the current image stored in the current image storage unit 171 and the background image stored in the background image storage unit 172, and changes between the current image and the background image. The moving area to be detected, that is, the detection target is extracted.

  The switch 175 is closed when the moving body counting apparatus 100 is started in order to store the image acquired by the image acquisition section 110 when the moving body counting apparatus 100 is started in the background image storage section 172 as a background image.

  The configuration of the moving object detection unit according to the first embodiment of the present invention has been described above using FIG. Next, a moving body counting method using the moving body counting apparatus according to the first embodiment of the present invention will be described.

  FIG. 3 is a flowchart for explaining the moving object counting method according to the first embodiment of the present invention. Hereinafter, the moving body counting method according to the first embodiment of the present invention will be described with reference to FIG.

  First, the moving body counting apparatus 100 is activated, and the image acquisition unit 110 acquires an image (step S110). The acquired image is subjected to conversion processing in the image preprocessing unit 120 so as to be suitable for image recognition processing in the image recognition unit 130 (step S120). The image conversion processing in the image preprocessing unit 120 is conventionally known, for example, reduction or removal of image noise or illumination unevenness, image density normalization, image binarization processing, and thinning processing. An image processing method can be used.

  When the image conversion processing in the image preprocessing unit 120 is completed, the image recognition unit 130 performs image recognition processing using the image subjected to the conversion processing in the image preprocessing unit 120 (step S130). The image recognition processing in the image recognition unit 130 is to extract a shape similar to the standard pattern recorded in the shape recording unit 140 from the image subjected to the conversion processing in the image preprocessing unit 120. The image recognition processing in the image recognition unit 130 is performed by, for example, a template matching method (reference: “New Edition Image Processing Handbook” published by the University of Tokyo, pp 1669 to 1682). Hereinafter, the shape similar to the standard pattern is also referred to as “similar pattern”. When the similar pattern is extracted from the image subjected to the conversion process, the position acquisition unit 150 acquires the coordinate position of the similar pattern in the image (step S140).

  When the position acquisition unit 150 acquires the coordinate position of the similar pattern in the image, the detection area generation unit 160 sets the coordinate area of the detection area in the image based on the coordinate position (step S150). FIG. 4 is an explanatory diagram for explaining a detection area set in an image. FIG. 4A illustrates an example in which a rectangular parallelepiped detection area 303 is set based on the similar pattern 302 extracted by the four partition poles 305 when the image acquisition unit 110 captures the image 300. Show. In the present embodiment, the shape of the detection area is a rectangular parallelepiped, but the shape of the detection area is not limited to this in the present invention.

  Here, if it is desired to change the range of the detection area 303, it is only necessary to add the partition pole 305 to the visual field range of the image acquisition unit 110. FIG. 4B shows an example in which the detection area 303 is set based on the similar patterns 302 extracted by the six partition poles 305 when the image acquisition unit 110 captures the image 300. Yes. As described above, the detection area can be detected without performing manual resetting simply by adding an object that can be extracted as a similar pattern to the visual field range of the image acquisition unit 110 or removing it from the visual field range of the image acquisition unit 110. The range can be changed automatically.

  When the detection area generation unit 160 sets a detection area in the image, the moving body detection unit 170 detects a moving body existing in the detection area.

  First, the image acquired by the image acquisition unit 110 is sent to the moving body detection unit 170. In the moving body detection unit 170, an image change area is detected by the background subtraction method using the image data acquired by the image acquisition unit 110 (step S160). The background difference method is a method of extracting a change area by taking a difference between a background image prepared in advance and a current image. By using the background subtraction method, for example, an image of a situation where no moving object exists is recorded as a background image, and when an image including a moving object to be detected is captured, the difference between the two is output. Thus, it is possible to classify the stationary background portion and the moving object to be detected, and as a result, it is possible to extract only the moving object.

  A moving object detection method based on the background subtraction method will be described in detail with reference to FIG. In the moving object detection unit 170 shown in FIG. 2, when the moving object counting device 100 is first activated, the switch 175 is closed, and the background image storage unit 172 displays an image when the moving object counting device 100 is activated as a background image. Is remembered as At this time, the background image stored in the background image storage unit 172 is preferably an image in a state where there is no moving object to be detected.

  When the image is stored in the background image storage unit 172, the image stored in the background image storage unit 172 acquired by the image acquisition unit 110 in the current image storage unit 171 after a predetermined time has elapsed. Different current images are stored. Then, the difference image extraction unit 174 extracts a difference between the current image stored in the current image storage unit 171 and the background image updated by the background image update unit 173, and an area changing between the two, that is, detection The moving object that is the target of.

  Here, the background image update processing in the background image update unit 173 will be described in detail.

  The moving body counting apparatus 100 according to the present embodiment stores the background image once in the background image storage unit 172 when the moving body counting apparatus 100 is activated. However, if a change in the arrangement of the lighting or the installation occurs in the background portion of the current image acquired by the image acquisition unit 110, a difference from the background image stored at the time of startup occurs, and a moving object is erroneously detected. There is a fear. Therefore, in the moving object counting apparatus 100 according to the present embodiment, by providing the background image update unit 173, even if a change in the arrangement of the illumination or the installation occurs, the erroneous detection of the moving object is prevented by updating the background image. can do.

  The background image update unit 173 acquires images from both the current image storage unit 171 and the background image storage unit 172, respectively. Then, the pixel values of the current image stored in the current image storage unit 171 and the background image stored in the background image storage unit 172 are compared at the same coordinate position. If the difference between the pixel values is smaller than a preset value as a result of the comparison, the background image is updated by replacing the pixel value of the coordinate in the background image with the pixel value of the coordinate in the current image. . Then, the updated background image is sent to the difference image extraction unit 174 and also sent to the background image storage unit 172, and the background image stored in the background image storage unit 172 is replaced with the updated image.

  The background image that follows the current image can be stored by the above-described background image update process even when the environmental conditions such as the illumination and the arrangement of the installation object have changed. Further, even when the background image suddenly changes, such as when the position of an arrangement included in the image changes, the background image can be updated with a certain time delay.

  The background image update processing in the background image update unit 173 has been described above.

  When the moving object detection unit 170 completes the detection of the moving object existing in the detection area, the object number counting unit 180 applies the pixel block of each pixel block of the moving object part detected by the moving object detection unit 170 to the pixel block. Based on the size, a block estimated as a person area (hereinafter, a block estimated as a person area is also referred to as “area block”) is detected. Then, the object number counting unit 180 acquires the detection area generated by the detection area generation unit 160, counts the number of area blocks included in the acquired detection area, and extracts the coordinates where the area blocks exist (step). S170).

  When the number and coordinates of the area blocks are extracted by the object number counting section 180, the number and coordinates of the extracted area blocks are displayed on the detection result display section 190 together with the image acquired by the image acquisition section 110 (step S180). Note that the present invention is not limited to this, and when displaying on the detection result display unit 190, only the number and coordinates of the extracted area blocks may be displayed on the detection result display unit 190.

  When the number and coordinates of the extracted area blocks are displayed on the detection result display unit 190, it is determined whether or not the next image is to be acquired as necessary (step S190). Returning to S110, moving image processing can be realized by acquiring an image from the image acquiring unit 110 after a predetermined time has elapsed.

  As described above, according to the first embodiment of the present invention, a detection area in an image is dynamically set by recognizing an object used to form a queue such as a partition pole. Can do. Therefore, the detection area in the image is automatically changed by simply changing the shape of the queue by installing or removing objects such as partition poles according to the cashier in the store or the ATM corner. Will be reset. Therefore, according to the first embodiment of the present invention, there is the convenience that it is not necessary to manually reset the detection area with the installation of a partition pole or the like.

(Second Embodiment)
In the first embodiment of the present invention, when a detection area is set, a moving object count that can dynamically set the detection area by extracting a reference object such as a partition pole by image recognition processing. The apparatus and the moving object counting method have been described. In the second embodiment of the present invention, instead of image recognition processing, the detection area is dynamically set by detecting the position of a reference object using an electronic tag such as RFID (Radio Frequency IDentification system). A movable body counting apparatus and a movable body counting method that can be used will be described.

  FIG. 5 is an explanatory diagram for explaining a moving object counting device according to a second embodiment of the present invention. Hereinafter, the moving body counter according to the second embodiment of the present invention will be described with reference to FIG.

  As shown in FIG. 5, the moving body counting apparatus 200 according to the second embodiment of the present invention includes an image acquisition unit 110, an attribute information acquisition unit 220, a position acquisition unit 250, and a detection area generation unit 260. The mobile object detection unit 170, the object number counting unit 180, and the detection result display unit 190 are configured.

  Among these, the image acquisition unit 110, the moving body detection unit 170, the object number counting unit 180, and the detection result display unit 190 can be configured by the same units as those of the moving body counting device 100 according to the first embodiment of the present invention. Detailed explanation is omitted. Here, the attribute information acquisition unit 220, the position acquisition unit 250, and the detection area generation unit 260, which are different in configuration and function of the moving body counting apparatus 100 according to the first embodiment, will be described.

  The attribute information acquisition unit 220 acquires attribute information from a reference object used for partitioning a queue such as a partition pole. In the second embodiment of the present invention, the reference object 400 includes the attribute storage unit 410, and the attribute information acquisition unit 220 wirelessly reads the information stored in the attribute storage unit 410, so that the reference object 400 attribute information is acquired.

  The position acquisition unit 250 specifies the position of each reference object 400 based on the attribute information of the reference object 400 acquired by the attribute information acquisition unit 220, and the image acquisition unit 110 acquires the position of the specified reference object 400. The coordinates are converted into coordinate positions on the image. A method for converting the position of the reference object 400 into a coordinate position on the image will be described later.

  Based on the coordinate position of the reference object 400 acquired by the position acquisition unit 250 and the attribute information of the reference object 400 set in advance, the detection area generation unit 260 includes a detection area area for detecting a moving object in the image. It is to set. The detection area setting method can be set, for example, by assigning numbers in advance to the attribute storage unit 410 of the reference object 400 and tracing the reference object 400 on the image in the order of the assigned numbers.

  The configuration of the moving object counter according to the second embodiment of the present invention has been described above with reference to FIG. Next, the moving body counting method according to the second embodiment of the present invention will be described.

  FIG. 6 is an explanatory diagram illustrating a moving object counting method according to the second embodiment of the present invention. Hereinafter, the moving body counting method according to the second embodiment of the present invention will be described with reference to FIG.

  First, the attribute information acquisition unit 220 detects the attribute storage unit 410 included in the reference object 400 (step S210). The attribute information acquisition unit 220 is installed around a place where a queue is formed, such as before a cash register in a store such as a supermarket or an ATM corner of a bank. In the present embodiment, the attribute information acquisition unit 220 is installed on the ceiling of a place where a queue is formed, for example.

  FIG. 7 is an explanatory diagram illustrating an example of installation of the attribute information acquisition unit 220 and the reference object 400 according to the second embodiment of the present invention. In the example illustrated in FIG. 7, a plurality of attribute information acquisition units 220 are installed on the ceiling, the attribute storage unit 410 included in the reference object 400 is searched via wireless, and the attribute information stored in the attribute storage unit 410 is acquired. To do. The attribute information stored in the attribute storage unit 410 may be an ID number, for example.

  When the attribute information acquisition unit 220 detects the attribute storage unit 410 included in the reference object 400, the image acquisition unit 110 subsequently acquires an image including the reference object 400 (step S220). The acquired image is sent to the position acquisition unit 250.

  When an image is acquired by the image acquisition unit 110 and the acquired image is sent to the position acquisition unit 250, the position acquisition unit 250 uses a reference based on the attribute information of the attribute storage unit 410 acquired by the attribute information acquisition unit 220. The position of the object 400 is specified, and the position of the specified reference object 400 is converted into coordinates in the image acquired by the image acquisition unit 110 (step S230).

  The coordinate conversion process of the reference object 400 in the position acquisition unit 250 associates each positional relationship of the attribute information acquisition unit 220 with the coordinates on the image acquired by the image acquired by the image acquisition unit 110 in advance, This can be implemented by detecting the reference object having which attribute information at which position on the image using the above-described correspondence.

  When the position of the reference object 400 is converted into coordinates in the image acquired by the image acquisition unit 110, the detection area generation unit 260 generates a detection area on the image based on the converted coordinates (step S240). For example, a different ID number may be assigned to each of the reference objects 400, and the detection area may be generated based on the ascending order of the assigned ID numbers.

  FIG. 8 is an explanatory diagram for explaining a detection area generation method according to the second embodiment of the present invention. FIG. 8 shows a detection area generation method when nine reference objects 400 are installed and ID numbers 1 to 9 are assigned to the reference objects 400, respectively. In the second embodiment of the present invention, the image acquisition unit 110 determines that the detection area is set in advance by connecting the reference objects 400 in the order of ID numbers 1 to 7 and 8 to 9. A detection area can be set on the acquired image.

  In this way, by assigning attribute information to a reference object serving as a reference for setting a detection area, the range of the detection area can be dynamically changed simply by changing the arrangement of the reference object.

  When a detection area is set on the image acquired by the image acquisition unit 110, the same flow as the moving object counting method according to the first embodiment is executed thereafter. Specifically, when the detection area is set, the moving object detection unit 170 detects an image change area by the background difference method (step S250), and the object number counting unit 180 detects the detection area generated by the detection area generation unit 260. Obtaining and counting the number of area blocks included in the obtained detection area, extracting the coordinates where the area blocks exist (step S260), and obtaining the number and coordinates of the extracted area blocks by the image acquisition unit 110 The image is displayed on the detection result display unit 190 together with the image (step S270). In the present invention, similarly to the first embodiment, when displaying on the detection result display unit 190, only the number and coordinates of the extracted area blocks may be displayed on the detection result display unit 190.

  Then, when the number and coordinates of the extracted area blocks are displayed on the detection result display unit 190, it is determined whether or not to acquire the next image as necessary (step S280). Returning to step S <b> 210, moving image processing can be realized by acquiring an image from the image acquisition unit 110.

  As described above, according to the second embodiment of the present invention, a detection area in an image is dynamically set by acquiring attribute information from an object used to form a queue such as a partition pole. can do. Therefore, the detection area in the image is automatically changed by simply changing the shape of the queue by installing or removing a reference object such as a partition pole according to the store's cash register or the congestion situation in the ATM corner. Will be reset to Therefore, according to the second embodiment of the present invention, there is the convenience that it is not necessary to manually reset the detection area with the installation of a partition pole or the like.

  As mentioned above, although preferred embodiment of this invention was described referring an accompanying drawing, it cannot be overemphasized that this invention is not limited to the example which concerns. It will be apparent to those skilled in the art that various changes and modifications can be made within the scope of the claims, and these are naturally within the technical scope of the present invention. Understood.

  For example, in the above embodiment, the partition pole is taken as an example of the reference object, and the range of the detection area is changed depending on the presence or absence of the partition pole. However, the present invention is not limited to this, for example, a rope stretched between the partition poles. It is also possible to detect information such as the presence or absence of information and change the range of the detection area based on the information.

  Further, the frequency of detection area change according to the present invention may be stored as data. By accumulating the frequency of detection area changes, for example, by day of the week or by time of day, it is possible to obtain data that will serve as the basis for considering optimization of queue arrangement according to the day of the week or time of day, It can be used to form a queue position that does not cause inconvenience to customers coming to the store.

It is explanatory drawing explaining the moving body counting device concerning 1st Embodiment. It is explanatory drawing explaining the structure of the moving body detection part concerning the 1st Embodiment of this invention. It is a flowchart explaining the moving body counting method concerning the 1st Embodiment of this invention. It is explanatory drawing explaining the detection area set in the image. It is explanatory drawing explaining the moving body counting device concerning the 2nd Embodiment of this invention. It is explanatory drawing explaining the moving body counting method concerning the 2nd Embodiment of this invention. It is explanatory drawing which shows an example of installation of the attribute information acquisition part and reference | standard object concerning the 2nd Embodiment of this invention. It is explanatory drawing explaining the production | generation method of the detection area concerning the 2nd Embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100,200 Mobile body counting apparatus 110 Image acquisition part 120 Image pre-processing part 130 Image recognition part 140 Shape recording part 150,250 Position acquisition part 160,260 Detection area generation part 170 Mobile body detection part 171 Current image storage part 172 Background image Storage unit 173 Background image update unit 174 Difference image extraction unit 175 Switch 180 Object number counting unit 190 Detection result display unit 220 Attribute information acquisition unit 300 Image 302 Similar pattern 303 Detection area 400 Reference object 410 Attribute storage unit

Claims (8)

An image acquisition unit for acquiring images at predetermined intervals;
An image recognition unit for recognizing a shape similar to a preset shape included in the image;
A position acquisition unit for acquiring a coordinate position of the similar shape included in the image;
A detection area generator for generating a detection area based on the coordinate position;
A moving body detecting unit for detecting a moving body included in the image;
A moving body counter that counts the number of the moving bodies included in the detection area;
A moving body counting apparatus comprising: The moving body detection unit
A current image storage unit for storing a current image acquired by the image acquisition unit;
A background image storage unit that stores an image acquired by the image acquisition unit as a background image;
The current image stored in the current image storage unit and the background image stored in the background image storage unit are compared, and when both are different by a predetermined value or more, the background image of the background image storage unit is A background image update unit for updating to the current image stored in the current image storage unit;
A difference image extraction unit that extracts a difference between the current image stored in the current image storage unit and the background image updated by the image update unit;
The mobile body counting apparatus according to claim 1, wherein An image acquisition unit for acquiring images at predetermined intervals;
An attribute information acquisition unit that acquires attribute information of a reference object included in the image and acquires position information of the reference object;
A position acquisition unit that converts the position information into a coordinate position in the image and acquires the position of the reference object;
A detection area generator for generating a detection area based on the position of the reference object;
A moving body detecting unit for detecting a moving body included in the image;
A moving body counter that counts the number of the moving bodies included in the detection area;
A moving body counting apparatus comprising: The moving body detection unit
A current image storage unit for storing a current image acquired by the image acquisition unit;
A background image storage unit that stores an image acquired by the image acquisition unit as a background image;
The current image stored in the current image storage unit and the background image stored in the background image storage unit are compared, and when both are different by a predetermined value or more, the background image of the background image storage unit is A background image update unit for updating to the current image stored in the current image storage unit;
A difference image extraction unit that extracts a difference between the current image stored in the current image storage unit and the background image updated by the image update unit;
The moving body counting device according to claim 3, wherein   The position information of the reference object is recorded in an electronic tag, and the attribute information acquisition unit acquires the position information of the reference object from the reference object by wireless. Mobile counting device.   The mobile object counting apparatus according to claim 3, wherein the attribute information is an ID number, and the detection area is formed based on an order of the ID numbers. An image acquisition step of acquiring images at predetermined intervals;
An image recognition step for recognizing a shape similar to a preset shape included in the image;
A position acquisition step of acquiring a coordinate position of the similar shape included in the image;
A detection area generation step of generating a detection area based on the coordinate position;
A moving body detecting step for detecting a moving body included in the image;
A moving body counting step of counting the number of the moving bodies included in the detection area;
The moving body counting method characterized by including. An image acquisition step of acquiring images at predetermined intervals;
An attribute information acquisition step of acquiring attribute information of a reference object included in the image and acquiring position information of the reference object;
A position acquisition step of converting the position information into a coordinate position in the image to acquire the position of the reference object;
A detection area generating step for generating a detection area based on the position of the reference object;
A moving body detecting step for detecting a moving body included in the image;
A moving body counting step of counting the number of the moving bodies included in the detection area;
The moving body counting method characterized by including.

Images