JP2022184576A - Image processing device, information processing method, and program - Google Patents

Abstract

To reduce processing load when partial images are used to measure a flow of measurement objects.SOLUTION: An image processing device counts the number of objects that pass through a detection line in a moving image, and sets an upper limit of the number of a plurality of partial images set in the vicinity of the detection line and to be referred to for counting the number of objects. The image processing device sets the detection line on the moving image according to the upper limit, and extracts the plurality of partial images set in the vicinity of the detection line from respective images in the moving image. The partial images are used to count the number of the objects that pass through the detection line.SELECTED DRAWING: Figure 6

Description

The present invention relates to technology for detecting an object from an image.

In recent years, there have been proposals for devices that analyze the flow of objects to be measured, such as the flow of people, or more specifically, the amount and direction of the flow of people, in a shooting area from moving images captured by a camera or the like. ing. Patent Literature 1 discloses a congestion estimation device that divides an image into a plurality of patches and determines whether a person is moving or staying in each patch.

JP 2009-110152 A

As the number of partial images (patches) to be analyzed increases, the analysis load increases corresponding to the increased number of patches, consuming computational resources. In the method described in Patent Document 1, since patches are generated from the entire image, there are cases where computational resources are consumed more than necessary in analyzing the flow to be measured.

SUMMARY OF THE INVENTION An object of the present invention is to reduce the processing load when measuring the flow of a measurement target using partial images.

In order to achieve the object of the present invention, for example, an image processing apparatus according to one embodiment has the following configuration. That is, an image processing device for measuring the number of objects passing through a detection line in a moving image, wherein the number of a plurality of partial images that are set near the detection line and referred to for measuring the number of the objects a first setting means for setting an upper limit; a second setting means for setting the detection line in the moving image in accordance with the upper limit; and a plurality of partial images set near the detection line. An extraction means for extracting from each of a plurality of images in a moving image, and a measurement means for measuring the number of the objects passing through the detection line using the partial images.

The processing load can be reduced when measuring the flow to be measured using the partial images.

2 is a diagram showing the hardware configuration of an image processing apparatus according to the first embodiment; FIG. 1 is a diagram showing a functional configuration diagram of an image processing apparatus according to Embodiment 1; FIG. 4 is a flowchart showing information processing performed by the image processing apparatus according to the first embodiment; 4A and 4B are diagrams showing an image group used by the image processing apparatus according to the first embodiment; FIG. 4 is a view showing a patch number setting GUI displayed by the image processing apparatus according to the first embodiment; FIG. FIG. 5 is a diagram showing a setting display of detection lines by the image processing apparatus according to the first embodiment; 4A and 4B are diagrams showing the display of patches with respect to detection lines of the image processing apparatus according to the first embodiment; FIG. FIG. 4 is a diagram showing analysis processing for each patch by the image processing apparatus according to the first embodiment; FIG. 7 is a diagram showing an example of redundant patch deletion processing by the image processing apparatus according to the first embodiment; FIG. 5 is a diagram showing a GUI when there are multiple measurement targets according to the first embodiment; FIG. 4 is a diagram showing the upper limit of the number of patches based on computational resources according to the first embodiment;

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. In addition, the following embodiments do not limit the invention according to the scope of claims. Although multiple features are described in the embodiments, not all of these multiple features are essential to the invention, and multiple features may be combined arbitrarily. Furthermore, in the accompanying drawings, the same or similar configurations are denoted by the same reference numerals, and redundant description is omitted.

[Embodiment 1]
An image processing apparatus according to this embodiment sets a detection line in a moving image and measures the flow of a measurement target passing through the detection line. For this reason, the image processing apparatus 100 according to the present embodiment sets an upper limit for the number of partial images to be referred to for measuring the flow of the measurement target, which is set in the vicinity of the detection line. Set the detection line accordingly. Next, the image processing apparatus 100 extracts a plurality of partial images set near the detection line from each of the plurality of images in the moving image, and uses the extracted partial images to measure the flow of the measurement target passing through the detection line. do.

The image processing apparatus 100 used in the passing people counting system according to the present embodiment will be described below. Here, the passing people counting system is a system that detects measurement targets in an image by image analysis and counts the number of measurement targets that have passed through a detection line provided in the image. Although the object to be measured is described as a human body, it is not particularly limited as long as it can be detected, for example, an animal or a vehicle, and detection processing may be performed simultaneously on a plurality of types of objects to be measured.

Here, the detection line refers to a line segment or line-shaped line provided in an image for counting the number of passages of a measurement target (hereinafter simply referred to as “number”) used in the above-described passing people counting system. shall be the area of As the detection line and the partial areas for analysis set in the vicinity of the detection line, those used in a general traffic flow measurement system can be arbitrarily used, and a detailed description of known techniques will be omitted. Here, partial images in partial regions are extracted from a plurality of images included in the moving image, and analysis processing is performed on the extracted partial images to measure the flow to be measured. In the following description, "analysis" refers to the process of counting the number of people described above.

This image processing device is used, for example, by a traffic counting system. A traffic volume measurement system is a system that detects objects to be measured in an image by image analysis and counts the number of objects to be measured that have passed through a detection line provided in the image. Although the object to be measured is described as being a human body or a vehicle, it is not particularly limited as long as it can be detected, such as an animal. Here, by dividing the area for image analysis into one or more partial areas (patches) and performing processing on each of them, analysis processing can be performed in parallel to realize high-speed processing.

FIG. 1 is a block diagram showing an example of the hardware configuration of an image processing apparatus 100. As shown in FIG. The image processing apparatus 100 includes a processor 101 , memory 102 , network interface (I/F) 103 , display device 104 and input device 105 . The processor 101 is, for example, a CPU, and performs processing by each functional unit of the image processing apparatus 100 shown in FIG. The memory 102 is, for example, ROM and RAM, and stores data or programs used by the processor 101 . The display device 104 is a liquid crystal display or the like, and displays the results of processing by the processor 101 and the like to the user. The input device 105 is a mouse and keyboard, a touch panel, buttons, or the like, and acquires user input. An I/F 103 is an interface for connecting the image processing apparatus 100 to a network. In this embodiment, the functions of the functional units of the image processing apparatus 100 are realized by the processor 101 executing processing based on the programs stored in the memory 102 .

FIG. 2 is a block diagram showing an example of the functional configuration of the image processing apparatus 100 according to this embodiment. The image processing apparatus 100 has an image acquisition unit 201 , a display unit 202 , a calculation unit 203 , a setting unit 204 and an analysis unit 205 .

The image acquisition unit 201 acquires one or more images to be processed. The image acquisition unit 201 may acquire an image stored in the memory 102, may acquire an image via the I/F 103, or may acquire an image using an imaging unit (not shown). The image acquisition unit 201 can acquire a set of images meeting specific conditions (for example, images captured by the same camera at predetermined time intervals) by using information about images such as image capture dates and times.

The setting unit 204 sets an upper limit for the number of partial images (patches) set in the vicinity of the detection line for measuring the flow to be measured, and sets the detection line based on the set upper limit. Here, the detection line is a line segment indicating the position where the flow (flow rate or direction of flow) to be measured is measured in the moving image. Here, the flow rate may be the total number of objects to be measured that pass through the set detection line, or the number per predetermined time. Hereinafter, a patch refers to a partial area (and a partial image extracted from that area) set in the vicinity of the detection line described above.

The setting unit 204 according to the present embodiment sets the upper limit of the number of patches based on user input acquired by the display unit 202, which will be described later. The upper limit of the number of patches is not limited to being set based on user input in this way, and may be determined based on computational resources allocated to analysis processing, for example. That is, the setting unit 204 can acquire the amount of computational resources in the processor 101 or memory 102 that can be used for analysis processing, and calculate the upper limit of the number of patches that can be set in an image based on the computational resources.

Next, the setting unit 204 limits the length of the detection line set based on the user input acquired by the display unit 202, which will be described later, and generates the detection line to be displayed on the image. The setting of the detection line will be described later together with the description of the display unit 202 .

The display unit 202 displays the image acquired by the image acquisition unit 201 . The display unit 202 then obtains user input to set the sensing line on the obtained image. For example, the display unit 202 can display a preview image for setting the detection line. Although detailed description will be given later with reference to FIGS. 4 to 8, the display unit 202 may superimpose and display specifiers used when setting detection lines on the image. In that case, the display unit 202 may acquire a user input for operating a specifier for setting a detection line by a click operation via a mouse cursor or by a touch operation on a touch panel. Here, it is assumed that the start point and the end point of the detection line are arranged by user operation, and the line segment from the start point to the end point is set as the detection line input by the user. This processing is not particularly limited as long as it acquires user input for setting the detection line. For example, the display unit 202 may acquire designation of the coordinates of the start point and end point of the detection line as the user input. Alternatively, the setting unit 204 may obtain designation of two points without distinction between the start point and the end point, then obtain the direction of the detection line, or may obtain inputs of the start point, direction, and length. The display unit 202 acquires a predetermined number of detection line settings equal to or greater than 1 (for example, 1, 2, or more). may be configured to set the detection line.

Next, the setting unit 204 limits the length of the detection line set based on the user input, and the display unit 202 displays the detection line with the limited length on the image. Processing for displaying the detection line with the limited length will be described later with reference to FIG. In addition, the display unit 202 may display, in the vicinity of the displayed detection line, a number of patches that include the detection line and are equal to or less than the set upper limit. The setting unit 204 can display the detection lines (and patches) set by the same process again when receiving the user's input to reset the detection lines. Here, the input of resetting the detection line by the user may be resetting of the position of the starting point or the end point of the detection line, or may be a change in direction. There may be.

The calculation unit 203 can calculate parameters used for analysis processing based on the setting of the detection line. For example, the calculation unit 203 may set a region to be analyzed in the image, and the positions and shapes of patches set by dividing the region, based on the setting of the detection line. The setting process performed by the calculation unit 203 will be described later with reference to FIG.

A setting unit 204 performs settings required for analysis processing. The setting unit 204 according to the present embodiment refers to, edits, and registers settings used for human body detection, tracking, and number counting processing. Further, the setting unit 204 may perform part or all of the processing performed by the display unit 202 and the calculation unit 203 described above. The analysis unit 205 measures the flow to be measured based on the moving image. For this purpose, the display unit 202 sets a detection line on the image, which indicates the position at which the flow of the measurement target is to be measured.

The analysis unit 205 performs analysis processing on the image acquired by the image acquisition unit 201 . The analysis unit 205 can measure, for example, the flow rate of the measurement target that crosses the detection line and moves from one region to the other region separated by the detection line, that is, the number of measurement targets that have passed.

Various methods are available for flow measurement. For example, there are a method of detecting and tracking a human body to be measured, and a method of estimating the position, moving direction, moving speed, etc. of the human body to be measured and directly obtaining the flow rate. Algorithms for realizing such a measurement method include, for example, a matching method, a method using optical flow, a method using machine learning, and a method using neural networks. A plurality of these methods can also be used in combination.

For flow measurement, a single partial image may be used, or a plurality of partial images may be used simultaneously. When using a plurality of partial images, partial images at the same time may be used, or partial images at different times may be used.

Specific processing methods by the analysis unit 205 include, for example, the following methods. First, _the analysis unit 205 estimates the position of the measurement target around the detection line at time t1 by inputting each partial image at time _t1 to the neural network. Similarly, the analysis unit 205 also estimates the position of the measurement target around the detection line at time _t2 by inputting each partial image at time _t2 to this neural network. This neural network can be trained to estimate the position of a measurement target (for example, the head of a human body) in the image from the image. Another method for improving the estimation accuracy is a neural network trained to estimate the density distribution of the measurement target in the image, and a neural network trained to estimate the position of the measurement target from the density distribution. may be used in combination with a network. With such a method, the analysis unit 205 can independently estimate the position of the measurement target in each different area using the partial images extracted from this area.

Next, the analysis unit 205 matches the estimated position of the measurement target at time _t1 with the estimated position of the measurement target at time _t2 , thereby obtaining the trajectory of the measurement target from time _t1 to time _t2 . to estimate As a matching method, a method of minimizing the cost according to the distance between measurement objects to be matched can be used, for example, the Hungarian matching method can be used. When the trajectory thus estimated intersects the detection line, it can be determined that one measurement target has passed the detection line. Such matching processing and trajectory estimation may be performed simultaneously based on the position of the measurement target detected from each partial image.

However, the flow measurement method is not limited to the above method. The analysis unit 205 may independently estimate the trajectory of the measurement target or measure the flow of the measurement target in each of the different regions using partial images extracted from this region. For example, the analysis unit 205 may estimate the trajectory of the measurement target for each partial image in addition to estimating the position of the measurement target for each partial image. Alternatively, partial images at the same position at time _t1 and time _t2 may be input to a neural network to estimate the position, movement direction, movement speed, etc. of the measurement target, thereby estimating the flow of the measurement target. .

4 to 8 are diagrams for explaining the analysis processing according to this embodiment. FIG. 4 shows an example of a preview image displayed by the display unit 202. As shown in FIG. Image group 400 is a plurality of images presented to the user to designate candidates. A human body 402 is shown in the image 401 and is detected by analysis processing. Also, the image group 400 may include an image that does not show the measurement target in the image 401 .

FIG. 5 is a diagram showing an example of a GUI display for the setting unit 204 to set the upper limit of patches. A screen 500 displays a button 501 for setting the upper limit of the number of patches from computational resources and a button 502 for manually setting the number of patches. The setting unit 204 can determine a setting method for the upper limit of the number of patches depending on whether the user presses the button 501 or the button 502 . A setting frame 503 for manually setting the upper limit of the number of patches is displayed on the screen 500, and the setting unit 204 sets the number input by the user in the setting frame 503 as the upper limit of the number of patches. do.

Here, the display unit 202 may present to the user the upper limit of the number of patches that can be set calculated by the setting unit 204 based on the computational resources. That is, regardless of whether or not it is decided to set the upper limit of the number of patches from the computational resources in FIG. 5, the upper limit of the computational resources may be presented to the user. According to such a configuration, when the user manually sets the detection line, it is possible to perform the setting process while confirming the number of patches that can be set in terms of computational resources.

FIG. 6 is a diagram for explaining an example of a process of limiting the length of detection lines according to the upper limit of the number of patches by the setting unit 204. In FIG. A preview screen 601 is displayed on the setting screen 600 , and a specifier 602 of the detection line, a start point 603 and an end point 604 that are end points of the specifier 602 , and a mover 605 are superimposed on the preview screen 601 . In FIG. 6, the user operates a mover 605 to move a start point 603 and an end point 604, and a designator 602 of a detection line is defined.

A setting screen 606 displays a detection line set based on a specifier 602 on a preview screen 601 . Here, if the detection line specified by the designator exceeds the length restricted based on the upper limit of the number of patches set by the setting unit 204, the display unit 202 displays the direction from the start point 603 to the end point 604. A line segment extending to a limited length is displayed as a detection line. Further, for example, the display unit 202 may set, as the detection line, a line segment obtained by aligning the specifier with the center and reducing the specifier to a length limited by the setting unit 204 .

A setting screen 607 is an example of a screen that visualizes and displays a portion that is not set as a detection line with respect to the designator 602 in addition to displaying detection lines similar to the setting screen 606 . Here, the part from the end point of the detection line to the moving element 505 is indicated by a dotted line. It is not included in the line, and only the actual combat part is set as the detection line.

Also, when the confirmation button 610 on the setting screens 600, 606, or 607 is pressed, the detection line used for the analysis process is determined from the detection line specified by the specifier 602 at that time and the upper limit of its length. set and displayed on the screen. These setting screens also display an image change button 609 for changing the displayed image. When the user designates a detection line by referring to a different image, the image to be displayed can be switched by pressing the image change button 609 . Here, the specifier before the image to be displayed is changed is deleted, and a new candidate is specified in the newly displayed image, but the specifier placed in the image remains even after the image is changed. , the detection line may be set based on all the specifiers arranged across the multiple images.

FIG. 7 is a diagram for explaining the process of setting patches in the vicinity of the set detection line. In the image 702 , the setting unit 204 sets a patch 705 near the detection line 701 having a start point 703 and an end point 704 . Also, the setting unit 204 sets the patch 706 so that the detection line is included in the patch 705 and the patch 706 . Thus, the setting unit 204 can set a plurality of patches so as to include detection lines.

The size of each patch is not limited, but the setting unit 204 may set the size of the patch according to the position of the patch. In the example of FIG. 7, the setting unit 204 sets a patch 705 centered on the start point 703, then sets a patch 706 adjacent to the patch 705 and including the end point 704, and completes the patch generation process. Here, the setting unit 204 may estimate the size of the measurement target at a predetermined position (here, center position) in each patch, and set the size of the patch based on the size. For example, the patch 705 may be set as a square area whose sides are a constant multiple (6.5 times, 8 times, etc.) of the shoulder width of the human body estimated from the position of the center of the patch 705 . Note that the shape of the patch is not limited to a square, and may be, for example, a rectangle, a quadrangle such as a trapezoid, or a circle.

Therefore, the calculation unit 203 may be capable of estimating the size of the measurement object corresponding to each position in the image. For example, the calculation unit 203 may have parameters in advance for calculating the size of the measurement object from each position, or may calculate based on the positions and sizes of at least three subjects in the image. good.

FIG. 8 is a diagram for explaining how the passage counting process is performed by the analysis unit 205. As shown in FIG. An image group 800 is an image acquired by the image acquisition unit 201, and an image 801 and an image 802 included in the image group 800 are temporally continuous images. Patches 803 and 804 are patches set in the moving image, and passage count analysis processing is performed in these patches.

Partial images 805 and 806 are partial images extracted from the image 801 using patches 803 and 804, respectively, and partial images 807 and 808 are converted to the same size. A partial image 809 and a partial image 810 are images showing that the analysis unit 205 has performed human body detection processing and human body discrimination processing on the partial images 807 and 808 and has discriminated the person appearing in each image. The asterisks 811 and 812 in the partial images 809 and 810 indicate positions where the human body is detected. Also, a partial image 813 and a partial image 814 are images generated from the image 802 by the same processing as the partial images 809 and 810 and indicating that the person appearing in each image has been determined. Stars 815 and 816 in image 813 and image 814 are positions where the human bodies corresponding to stars 811 and 812 are detected, respectively.

Here, the analysis unit 205 uses partial images extracted using the same patch (here, partial images 809 and 813 or 810 and 814) to perform processing for detecting passage of the detection line. A partial image 815 is an image showing how the partial images 809 and 813 are used to detect passage. First, the analysis unit 205 obtains a trajectory 817 of the human body based on the stars 811 and 815, and determines whether or not the trajectory 817 intersects the detection line. If the analysis unit 205 determines that there is an intersection here, it is assumed that passage has been detected, and if it is determined that there is no intersection, it is assumed that passage has not been detected. Here, the analysis unit 205 may record, as one of the results of the passage detection processing, the traveling direction determined from the trajectory or the direction of the detection line. Further, the analysis unit 205 may record the position of the intersection of the path and the passing line, or the patch where the passage is detected, as one of the results of the passage detection processing. A partial image 816 is an image showing how the partial images 810 and 814 are used to perform the passage detection process in the same manner as the partial image 815 .

Here, the analysis unit 205 performs each process on two images, the image 801 and the image 802, but the number of images to be processed is not particularly limited, and any number of images can be processed. It can be carried out. Further, the analysis unit 205 may group by the traveling direction of the human body or by the patches in which the passage was detected, and then aggregate the number of detected passages for each group. The passage detection processing by the analysis unit 205 is not limited to that shown in FIG. 8, and other known passage detection processing may be used.

Further, when there is a patch in which the passing amount of the measurement target passing through the detection line within a certain period of time is less than the threshold value, the analysis unit 205 may stop the analysis processing for that patch. For example, the analysis unit 205 calculates the total number of measurement targets within a predetermined period or the amount per hour in a certain patch, and if the value is equal to or less than a predetermined threshold, the analysis processing in the patch and delete the information for that patch. At this time, the display unit 202 may notify the user that the patch has been deleted (the process for that patch has been stopped). The threshold used here may be set in advance, may be calculated from the passing amount in each patch, or may be set by the user as a desired condition. For example, when there is a patch whose passing amount is less than 5% of all patches, the analysis unit 205 can stop the analysis processing for that patch. Further, for example, when there is a patch in which the number of passing persons per hour is one or less, the analysis unit 205 can stop the analysis processing for that patch.

FIG. 9 is a diagram showing an example of stopping the patch analysis process described above. A screen 900 is the screen before the patch 902 is deleted, and a screen 901 is the screen after the patch 902 is deleted. In the patch 902 including the end point 903 of the detection line, the number of passages detected in a certain period of time was small, so the information about the patch 902 was deleted by the analysis unit 205, and the location where the detection line was included in the patch 903 was also deleted. there is According to such a process, it is possible to stop the process for patches with a small passing amount to be measured and to save computational resources.

FIG. 3 is a flowchart showing an example of information processing performed by the image processing apparatus 100 according to this embodiment. In S301, the setting unit 204 acquires user input specifying the upper limit of the number of patches. In S302, the calculation unit 203 calculates the length of the detection line that can be set based on the upper limit of the number of patches. In S<b>303 , the display unit 202 acquires user input specifying a detection line. In S304, the setting unit 204 determines whether the detection line specified by the user input has a length equal to or less than that calculated in S303. If it is equal to or less than the length, the process proceeds to S305; otherwise, the process proceeds to S306.

In S305, the setting unit 204 determines the detection line specified by the user as the detection line to be used for analysis processing as it is, and advances the processing to S307. In S306, the setting unit 204 limits the length of the detection line to the length set in S302, and advances the process to S307.

In S307, the display unit 202 determines whether or not the user has performed detection line confirmation processing. If the confirmation process is being performed, the information on the detection line is stored and the process ends. If not, the process returns to S303.

According to such a configuration, when measuring the flow of the measurement target passing through the detection line, the upper limit of the patch set near the detection line for measurement is set, and the length of the detection line is adjusted according to the upper limit. can be restricted. Therefore, by setting the detection line so that the number of patches falls within the set upper limit, it is possible to perform passage detection processing in consideration of available computational resources.

Although the setting unit 204 has been described in various ways assuming that the human body is used as the object to be measured, for example, when different objects such as a human body and a vehicle are used as the object to be measured at the same time, the objects may be measured without distinguishing between them. detection and analysis processing may be performed separately. When performing analysis processing by distinguishing between different measurement objects, the processing by each functional unit shown in FIG. 2 shall perform the above-described processing for each of the different types of measurement objects.

FIG. 10 is a diagram showing an example of a GUI for separately performing various settings for the human body and the vehicle, displayed by the display unit 202 according to this embodiment. UI 1000 displays a box 1001 for designating the number of patches. Here, the number entered by the user in box 1001 is set as the upper limit of patches, but the user entry may be made in a different format. In the example of FIG. 10, the setting unit 204 sets the value input in the box 1001 as the upper limit of the total number of patches set for the human body and the vehicle.

The UI 1000 also displays a box 1002 for specifying the upper limit of the number of patches set for the human body and a box 1006 for setting the upper limit of the number of patches for the vehicle. The UI 1000 also includes boxes 1003 and 1004 for setting the upper limit of the number of patches for two detection lines for detecting passage of a human body, and boxes 1004 and 1004 for setting the upper limit of the number of patches for one detection line for detecting passage of a vehicle. A box 1007 for setting is displayed. Here, if there is a discrepancy in the set number of patches, for example, if the values input in boxes 1003 and 1004 exceed the value input in box 1002, display unit 202 can notify the user. can. The display unit 202 may display, for example, a message indicating an error as a warning, or may issue a warning with a notification sound. good too. Here, a message is displayed as a balloon 1005 indicating that there is a discrepancy between the numbers of patches that have been input. According to such a display, even when setting conditions for the number of patches for a plurality of measurement targets, it is possible to set detection lines so that the set upper limit number is not exceeded.

FIG. 11 is a diagram showing an example of a GUI that, in addition to the display of FIG. 10, calculates the upper limit of the number of patches from available computing resources and presents the upper limit to the user. In the UIs 1100 and 1101, the calculation unit 203 calculates that the upper limit of the number of patches based on the calculation resources is 10, and this is presented as a display 1102 to the user. Similar to FIG. 10, boxes for inputting the upper limit of patches for the human body and the vehicle are displayed. Here, the display unit 202 displays the configurable (remaining) number of items for which the user has not yet input the upper limit on the number of patches, based on the upper limit entered for other measurement targets and/or detection lines. An upper limit may be displayed. In the example of FIG. 11, a box 1103 for setting the upper limit for one of the two patches of the detection line for detecting the passage of a human body, and a box 1104 for setting the upper limit for the patch for detecting the passage of a vehicle. , the upper limit value is entered. On the other hand, the box for another patch of sensing lines for detecting the passing of a human body is filled in. Therefore, the display unit 202 displays on the UI 1100 a display 1105 indicating that the remaining number of available patches is three, based on the display 1102 and the input contents of the boxes 1103 and 1104 . According to such a display, it is possible to confirm the upper limit of the number of patches automatically calculated from available computing resources and set the upper limit of various patches. In addition, while setting the upper limit, it is possible to perform setting while confirming the upper limit values that can be set for other items.

(Other examples)
The present invention supplies a program that implements one or more functions of the above-described embodiments to a system or apparatus via a network or a storage medium, and one or more processors in the computer of the system or apparatus reads and executes the program. It can also be realized by processing to It can also be implemented by a circuit (for example, ASIC) that implements one or more functions.

The invention is not limited to the embodiments described above, and various modifications and variations are possible without departing from the spirit and scope of the invention. Accordingly, the claims are appended to make public the scope of the invention.

201: image acquisition unit, 202: display unit, 203: calculation unit, 204: setting unit, 205: analysis unit

Claims (19)

An image processing device that measures the number of objects passing through a detection line in a moving image,
a first setting means for setting an upper limit on the number of a plurality of partial images to be referred to for counting the number of objects set in the vicinity of the detection line;
a second setting means for setting the detection line in the moving image according to the upper limit;
extracting means for extracting a plurality of partial images set near the detection line from each of the plurality of images in the moving image;
measuring means for measuring the number of the objects passing through the detection line using the partial image;
An image processing device comprising: 2. The second setting means sets the detection line such that the number of partial images set by a predetermined method for the detection line is equal to or less than the upper limit. The image processing device according to . When a plurality of detection lines are set, the second setting means sets the total number of the partial images set by a predetermined method for each of the plurality of detection lines to be equal to or less than the upper limit. 3. The image processing apparatus according to claim 2, wherein the detection line is set. 4. The image processing apparatus according to claim 1, wherein said plurality of partial images are set so as to include said detection line. 5. The image processing apparatus according to claim 4, wherein the size of said partial image is determined according to the position of said partial image. Further comprising estimation means for estimating the size of the object corresponding to each position in the moving image,
5. The image processing according to claim 4, wherein the size of each of said partial images is set based on the size of said object corresponding to the position of said partial image estimated by said estimation means. Device. 5. The image processing apparatus according to claim 4, wherein the size of said partial image is changed according to the length of said detection line and said upper limit. 8. The image according to any one of claims 1 to 7, wherein said first setting means sets said upper limit based on computational resources allocated to the process of measuring the number of said objects. processing equipment. further comprising a first acquisition means for acquiring a first user input designating the upper limit;
8. The image processing apparatus according to claim 1, wherein said first setting means sets said upper limit based on said first user input. calculation means for calculating an upper limit of the number of the plurality of images that can be set in the moving image based on the calculation resources allocated to the process of measuring the number of the objects;
presentation means for presenting the upper limit calculated by the calculation means to the user;
10. The image processing apparatus according to claim 9, further comprising: further comprising a second acquisition means for acquiring a second user input designating the sensing line;
11. The image processing apparatus according to claim 1, wherein said second setting means sets said detection line based on said second user input. the second user input includes specifying start and end points of the sensing line;
12. The image processing apparatus according to claim 11, wherein said second setting means sets said detection line extending from said start point toward said end point and having a length corresponding to said upper limit. display means for displaying a second detection line designated by the second user input and the detection line set by the second setting means when the second user input is acquired; 13. The image processing device according to claim 11 or 12, further comprising: 14. The image processing apparatus according to claim 13, wherein said display means further displays said plurality of partial images set with respect to said detection line. The measurement means measures the number of objects passing through the detection line measured within a predetermined period of the plurality of partial images, if the partial image is equal to or smaller than a predetermined threshold. 15. The image processing apparatus according to any one of claims 1 to 14, characterized in that the measurement of the number of objects passing through the detection line using the partial images below is stopped. 16. The method of claim 15, wherein the number of objects is the total number of objects passing through the sensing line measured within the predetermined time period, or the number of objects passing through the sensing line per hour. image processing device. An image processing device that measures the number of objects passing through a detection line in a moving image,
a first setting means for setting an upper limit on the number of a plurality of partial images to be referred to for counting the number of objects set in the vicinity of the detection line;
an acquisition means for acquiring user input designating the detection line;
second setting means for setting the partial image in the vicinity of the detection line based on the user input,
a second setting means for limiting the number of partial images set in the vicinity of the detection line to the upper limit or less;
extracting means for extracting a plurality of partial images set near the detection line from each of the plurality of images in the moving image;
measuring means for measuring the number of the objects passing through the detection line using the partial image;
An image processing device comprising: An information processing method performed by an image processing device for measuring the number of objects passing through a detection line in a moving image,
setting an upper limit on the number of partial images to be referred to for counting the number of objects, which are set in the vicinity of the detection line;
A step of setting the detection line in the moving image,
limiting the length of the sensing line according to the upper limit;
a step of extracting a plurality of partial images set in the vicinity of the detection line from each of the plurality of images in the moving image;
counting the number of objects passing through the sensing line using the partial image;
An information processing method, comprising: A program for causing a computer to function as each means of the image processing apparatus according to any one of claims 1 to 17.

Images