JP2004007174A - Image processing system, image processing method and image processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To properly specify the existing region of an object to be monitored in accordance with situations. <P>SOLUTION: This image processing system is provided with a first detecting means 11 for detecting an object region by calculating inter-frame difference from image data obtained by an image acquiring means 10 which fetches image data to be monitored from a camera, a second detecting means 12 for detecting the object region by calculating a background difference from the image data, and a selecting means 14 for selecting either the detected region only from the first detecting means 11 or the detected regions from the first and second detecting means 11 and 12 to be used on the basis of the evaluation result of an evaluating means 13 which evaluates the congesting conditions of objects to be monitored based on results obtained from the first detecting means 11 and the second detecting means 12, and evaluates the quality of the image data based on a video signal. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an image processing system, an image processing method, and an image processing program suitable for use in a road monitoring system, an intrusion monitoring system, and the like.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, in the field of road monitoring and intrusion monitoring, an image obtained by capturing an image of an object to be monitored by a camera is subjected to image processing to detect a traffic jam, an accident, or an intruder.
[0003]
2. Description of the Related Art Conventionally, as a method of detecting an object (including a person in the present specification) by performing image processing, there are known methods of a background difference process and an inter-frame difference process.
[0004]
The above-described background subtraction processing has an advantage that the contour of the monitoring target can be accurately detected when the background image can be created normally. However, for example, when the road is congested (including traffic congestion) and the road surface is covered by the vehicle, there is a problem that the background image cannot be created normally and the vehicle cannot be detected well.
[0005]
On the other hand, with the inter-frame difference, an object can be appropriately detected without being affected by a change in the environment, if the object is strong against a change in luminance or noise and has a relatively large movement. However, there is a problem that a stopped object cannot be detected.
[0006]
Therefore, an image processing system using a combination of the above two methods has been studied. Examples of such a system include JP-A-2000-175174, JP-A-7-334683, JP-A-6-325157 and JP-A-6-325157. No. 6,266,841 and the like.
[0007]
By the way, in an external environment in which the brightness changes with time, the surrounding environment changes with time, so an image processing method corresponding to the temporal change in brightness is required. In an environment such as a tunnel with sparse lighting and a spatial distribution of brightness, an image processing method corresponding to the spatial brightness distribution is required.
[0008]
[Problems to be solved by the invention]
The above-mentioned known system does not take such an environment into consideration and is not sufficient. The present invention has been made in view of the current situation, and has as its object to determine the advantages and disadvantages of the above two image processing methods and to appropriately select an image processing method according to various situations. The present invention provides an image processing system, an image processing method, and an image processing program that can appropriately specify a region where a monitoring target exists in accordance with a situation by performing the switching.
[0009]
[Means for Solving the Problems]
An image processing system according to the present invention includes: an image acquisition unit that captures monitoring target image data obtained by a camera; and a first detection that performs an inter-frame difference on the image data captured by the image acquisition unit to detect an object region. Means, a second detection means for performing object subtraction by performing background subtraction on the image data captured by the image acquisition means, and a result obtained by the first detection means and the second detection means. An evaluation unit that evaluates the congestion state of the monitoring target based on the video signal, and evaluates the quality of image data obtained from the camera based on a video signal, and only the first detection unit based on an evaluation result by the evaluation unit. A selection is made as to whether to use the detection area obtained or to use the detection area obtained by the first detection means and the second detection means. Characterized by comprising a selection means.
[0010]
The image processing system according to the present invention is characterized by further comprising an optimizing means for selecting and combining the detection areas selected by the selecting means and optimizing the area shape.
[0011]
In the image processing system according to the present invention, in the first detection unit, a plurality of detection areas are obtained by changing an interval between frames for obtaining a difference, and in the second detection unit, by changing a background image. The optimizing means is characterized by selecting and combining the plurality of detection areas to optimize the area shape.
[0012]
An image processing method according to the present invention includes an image acquisition step of capturing monitoring target image data obtained by a camera, and a first detection of performing an inter-frame difference on the image data captured by the image acquisition step to detect an object area. Step, a second detection step of performing object subtraction by performing background subtraction on the image data captured by the image acquisition step, and a result obtained by the first detection step and the second detection step. An evaluation step of evaluating the congestion state of the monitoring target based on the video signal, and evaluating the quality of image data obtained from the camera based on a video signal, and only the first detection step based on the evaluation result of the evaluation step. The obtained detection area is used or the first detection step and the second detection step are used. Characterized by comprising a selection step of performing one of selecting using the detection region obtained by.
[0013]
The image processing method according to the present invention is characterized by further comprising an optimizing step of selecting and combining the detection areas selected in the selecting step to optimize the area shape.
[0014]
In the image processing method according to the present invention, in the first detection step, a plurality of detection areas are obtained by changing a frame interval for obtaining a difference, and in the second detection step, changing a background image. In the optimizing step, the plurality of detection areas are selected and combined to optimize the area shape.
[0015]
An image processing program according to the present invention is an image processing program used in a computer that performs processing by a central processing unit based on a program stored in a main memory and obtains a result. An image acquisition step of capturing target image data, a first detection step of performing an inter-frame difference on the image data captured by the image acquisition step to detect an object area, and an image data captured by the image acquisition step. A second detection step of detecting an object area by performing background subtraction, and evaluating a congestion state of a monitoring target based on a result obtained by the first detection step and the second detection step; Evaluation step for evaluating the quality of image data obtained from Selecting whether to use a detection area obtained only by the first detection step or to use a detection area obtained by the first detection step and the second detection step based on the evaluation result of the evaluation step. And a selecting step of performing the following.
[0016]
The image processing program according to the present invention is further characterized by executing an optimizing step of optimizing an area shape by selecting and combining the detection areas selected in the selecting step.
[0017]
In the image processing program according to the present invention, in the first detection step, a plurality of detection areas are obtained by changing a frame interval for obtaining a difference, and in the second detection step, changing a background image. In the optimizing step, the plurality of detection areas are selected and combined to optimize the area shape.
[0018]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of an image processing system, an image processing method, and an image processing program according to the present invention will be described with reference to the accompanying drawings. FIG. 1 shows the configuration of the image processing apparatus. This image processing apparatus is realized by the image processing unit computer 2 of the monitoring system shown in FIG.
[0019]
As shown in FIG. 2, the monitoring system is connected via a network 1 to an image processing unit computer 2, an integrated processing unit computer 3, an event determination unit computer 4, and notification boards 5-1 to 5-n for performing display and the like. ing. The cameras 6-1 to 6-m (m is an integer) are connected to the image processing unit computer 2.
[0020]
In the above system, the operation is performed as follows. The cameras 6-1 to 6-m capture images of the monitoring target and send image data to the image processing unit computer 2. The image processing unit computer 2 specifies the area where the monitoring target exists, as described later, and sends processing result information to the integrated processing unit computer 3. Based on the information sent from the image processing unit computer 2, the integrated processing unit computer 3 determines, for example, a result of the image processing unit computer 2 processing the image data obtained by any of the cameras 6-1 to 6 -m. For example, the image data necessary for the next event determination is extracted or combined. The result is sent from the integrated processing unit computer 3 to the event determination unit computer 4. The event determination unit computer 4 determines the event that has occurred using the sent image data and the result processed by the integration processing unit computer 3. Here, the event is, for example, occurrence of an accident, occurrence of a falling object, traffic congestion, fire, and the like in a road monitoring system.
[0021]
The event determination unit computer 4 determines which information is to be sent to any of the notification boards 5-1 to 5-n based on the determination result, and notifies the necessary information via the network 1 to the required information. Send to the board. Thus, the notification boards 5-1 to 5-n provide notification of required information.
[0022]
In the image processing unit computer 2 described above, an image processing device is configured by each unit shown in FIG. This image processing apparatus includes an image acquisition unit 10, a first detection unit 11, a second detection unit 12, an evaluation unit 13, a selection unit 14, and an optimization unit 15. The image acquisition unit 10 captures monitoring target image data from the cameras 6-1 to 6-m, and converts the image data into computer-processable image data for output to the first detection unit 11 and the second detection unit 12. Converting. The first detection unit 11 performs an inter-frame difference on the image data captured by the image acquisition unit 10 to detect an object area. The second detection unit 12 performs background subtraction on the image data captured by the image acquisition unit 10 to detect an object area.
[0023]
Further, the evaluation unit 13 determines the congestion state of the monitoring target (congestion in the case of road monitoring, and a large number of persons in the case of intrusion monitoring) based on the results obtained by the first detection unit 11 and the second detection unit 12. And the quality of image data obtained from the cameras 6-1 to 6-m is evaluated based on the video signal. The selection unit 14 uses a detection area obtained only by the first detection unit 11 based on the evaluation result by the evaluation unit 13 or uses a detection area obtained by the first detection unit 11 and the second detection unit 12. To make a selection. Further, the optimizing unit 15 selects and combines the detection regions selected by the selecting unit 14 to optimize the region shape.
[0024]
The image processing unit computer 2 that constitutes the image processing apparatus described above is configured by, for example, a personal computer, a workstation, or another computer, and includes, for example, the components illustrated in FIG. In other words, the computer shown in FIG. 3 has a CPU 51 that controls the entire apparatus, and is connected to a main storage device 52 that stores information such as programs and data used by the CPU 51. Further, a keyboard control unit 54, a display control unit 55, a printer control unit 56, a communication interface 57, a mouse control unit 58, and a magnetic disk control unit 59 are connected to the CPU 51 via a system bus 53. The keyboard control unit 54 is connected with a keyboard input device 60 capable of inputting various information by keys, the display control unit 55 is connected with a CRT display device 61 for displaying information, and the printer control unit 56 is printed with information. A printer device 62 for output is connected, a communication interface 57 is connected to a communication processing unit 63 for performing communication via a line, a mouse control unit 58 is connected to a mouse 64 as a pointing device, The magnetic disk device 65, which is an auxiliary storage device, is connected to the disk control unit 59. The image processing unit computer 2 includes at least a CPU 51, a main storage device 52, a magnetic disk control unit 59, a magnetic disk device 65, a communication interface 57, and a communication processing unit 63. Further, a flexible disk drive, a magnetic card or IC card reader, an MO (magneto-optical disk) drive, and the like are provided as needed.
[0025]
Each unit shown in FIG. 1 is realized by the CPU 51 executing the image processing method according to the present invention by a program corresponding to the flowchart shown in FIG. Will be described. The program corresponding to the above-described flowchart is an image processing program according to the present invention, which is stored in, for example, the magnetic disk device 65 and is read out and executed by the main storage device 52 constituting the main memory. First, the operation is started, and image data is taken in from the cameras 6-1 to 6-m (S1: image acquisition means 10).
[0026]
Next, image data for each frame is sequentially obtained, and the presence area of the object is detected based on the inter-frame difference (S2). In addition, the background image data and the frame image data obtained sequentially are concurrently obtained. The difference is obtained, and the detection of the existing area of the object is executed (S3).
[0027]
While the frame difference is performed in step S2 and the background difference is performed in step S3, the evaluation of the congestion state of the monitoring target and the evaluation of the image data based on the video signals obtained from the cameras 6-1 to 6-m are performed. By performing the quality evaluation, the two types of differences are evaluated (S4). That is, in this situation, it is evaluated whether it is better to use the result based on the inter-frame difference alone or whether it is better to use the result based on the inter-frame difference and the background difference.
[0028]
Based on the evaluation in the above step S4, a selection is made between a result based on only the inter-frame difference or a result based on the inter-frame difference and the background difference (S5), and the detection area of the object related to the result based on the selected difference is optimized. The processing is performed (S6), and the detection area of the object obtained by the optimization is output as vehicle information (S7).
[0029]
The details of the processing of steps S2 to S7 above are as shown in the flowcharts of FIGS. The details will be described below. The inter-frame difference process (S2) and the background difference process (S3) perform a plurality of inter-frame difference processes (S2-1 to S2-n) that can obtain different results, as shown in the flowchart of FIG. , A plurality of background difference processes (S3-1 to S3-n) that can obtain different results. Specifically, in a plurality of inter-frame difference processes (S2-1 to S2-n), the intervals between frames for which a difference is determined in each process are different, or the thresholds for binarization are different, so that different processes are performed. Results. In the plurality of background difference processes (S3-1 to S3-n), different results are obtained by setting different background images or different threshold values for binarization in each process.
[0030]
Since the processing procedure of each of the inter-frame difference processing (S2-1 to S2-n) is the same, the inter-frame difference processing (S2-1) will be described as a representative. First, difference processing is performed on image data at a predetermined frame interval (S21). Since the result of the difference in step S21 is multi-valued (for example, 256 gradations), binarization is performed using a predetermined threshold to obtain a binarized image (S22). An area corresponding to the contour of the object obtained from the binarized image is detected as an object existing area (S23) and passed to the next processing.
[0031]
The background difference processing (S3-1 to S3-n) has the same processing procedure, and therefore the background difference processing (S3-1) will be described as a representative. First, background image data is created (S31). Next, a difference process is performed between the currently arriving one-frame image data and the background image data created above (S32). Since the result of the difference in step S32 is multi-valued (for example, 256 gradations), binarization is performed using a predetermined threshold to obtain a binarized image (S33). An area corresponding to the contour of the object obtained from the binarized image is detected as an object existing area (S34) and passed to the next processing.
[0032]
Subsequent to the processing in steps S2 and S3 as described above, in the evaluation in step S4 shown in FIG. 4, the detailed processing shown in FIG. 6 is performed. First, it is determined whether or not there is congestion using the detection areas obtained in steps S2 and S3 (S11). The determination in step S11 is whether or not traffic congestion has occurred in road monitoring, and whether or not many people are present in the monitoring area in intrusion monitoring. Specifically, when the presence of a plurality of detection regions in the image is recognized or when the detection regions overlap, it can be determined that the image is congested.
[0033]
If it is determined in step S11 that there is congestion, the process proceeds to the processing shown in FIG. 7. If it is determined that there is no congestion, the state of the image is input as a video signal to detect the S / N ratio (predetermined or higher). It is detected whether the image state is clear or unclear based on the S / N ratio) and the contrast (the difference between the upper and lower luminance is equal to or more than a predetermined value) (S12). If it is detected in this step S12 that the vehicle is unclear, it is output that the vehicle information cannot be detected (S13), and the process ends. When it is detected that the state of the video is clear, the processing shifts to the processing shown in FIG.
[0034]
The process illustrated in FIG. 7 is a process using the result based on only the inter-frame difference, and the process illustrated in FIG. 8 is a process using the result based on both the inter-frame difference and the background difference. The presence or absence of congestion is detected in step S11 of FIG. 6 described above, and if congestion occurs, the monitoring area is covered with an object, and in view of the fact that the result obtained by the background difference becomes meaningless, the frame shown in FIG. Proper processing is assured by proceeding to processing using the result based only on the inter-difference. Further, the image state is detected in step S12 in FIG. 6 described above, and when the image is unclear, no processing is performed when any processing has no meaning, thereby avoiding generation of an unnecessary load.
[0035]
Next, a description will be given of a process using a result based on only the inter-frame difference shown in FIG. First, a detection area obtained by the inter-frame difference is obtained (S41). As shown in the flowchart of FIG. 5, since a plurality of inter-frame difference processes (S2-1 to S2-n) that can obtain different results are performed, a plurality of detection regions are obtained. Therefore, an optimum area is selected by performing comparison selection and combination of these detection areas (S42). As an example of this, the technology disclosed in Japanese Patent Application No. 2002-106681 already presented by the present inventor can be used.
[0036]
In FIG. 9A, FIG. 10A, and FIG. 11A, an area where a vehicle actually exists in one current frame is an area surrounded by a thick frame. One of the results of the inter-frame difference processing (S2-1 to S2-n) is specified as a region where "1" is gathered, for example, as shown in FIG. Therefore, an area as shown in FIG. 9B is set as a detection area of the monitoring target. Further, another result of the inter-frame difference processing (S2-1 to S2-n) is specified as a region where "1" is gathered, for example, as shown in FIG. Therefore, an area as shown in FIG. 10B is set as a detection area of the monitoring target. Further, still another result of the inter-frame difference processing (S2-1 to S2-n) is specified as a region where "1" is gathered, for example, as shown in FIG. Therefore, an area as shown in FIG. 11B is set as a monitoring target detection area.
[0037]
When there are three types of inter-frame difference processing (S2-1 to S2-n), a plurality of (here, three) regions of the monitoring target are combined. In the synthesis processing here, first, synthesis is simply performed by an OR (logical sum) operation. When the regions shown in FIGS. 9 (b), 10 (b) and 11 (b) have been extracted, as shown in FIG. Is specified.
[0038]
When an optimal detection area is obtained in step S42 in FIG. 7 as described above, the result is output as detection area information of an inter-frame difference (S43). The information of the specific area is used by the image processing unit computer 2 to specify the position of the vehicle in this example and use it for necessary processing, and is also sent to the integrated processing unit computer 3 and the event determination unit computer 4 to be used for event determination. Will be.
[0039]
Next, processing using the results of both the inter-frame difference and the background difference shown in FIG. 8 will be described. First, the detection area obtained by the background difference is obtained (S51). As shown in the flowchart of FIG. 5, since a plurality of background subtraction processes (S3-1 to S3-n) that can obtain different results are performed, a plurality of detection regions are obtained. Next, a detection area obtained by the inter-frame difference is obtained (S52). As shown in the flowchart of FIG. 5, since a plurality of inter-frame difference processes (S2-1 to S2-n) that can obtain different results are performed, a plurality of detection regions are obtained.
[0040]
Therefore, a plurality of detection areas by these two methods are synthesized (S53). Although various methods of this combination can be considered, for example, a simple logical product is obtained. If the n regions are obtained by the inter-frame difference and the n regions are obtained by the background difference, for example, an nxn synthesis result is obtained.
[0041]
Next, the routine proceeds to step 54 shown in FIG. In this step S54, the detection area obtained by only the inter-frame difference created in the processing in FIG. 8, the detection area obtained only by the background difference processing, and the detection area obtained by combining the detection area by the inter-frame difference and the background difference processing Which of the regions is optimal is determined by, for example, detecting a region that best obtains matching between each detection region and a previously held region sample for each vehicle type.
[0042]
In step S54, when it is detected that the detection region obtained only by the inter-frame difference is optimal, a plurality of inter-frame difference processes (S2-1 to S2-n) that obtain different results are performed. Since the detection has been performed, a plurality of detection regions have been obtained, and an optimum region is obtained for these detection regions by using the method described with reference to FIGS. 9 to 12 (S55). If it is detected in step S54 that the detection area obtained only by the background difference processing is optimal, a plurality of background difference processings (S3-1 to S3-n) that obtain different results are performed. Therefore, an optimum area is determined for these detection areas using the method described with reference to FIGS. 9 to 12 (S56). Further, when it is detected that the detection area obtained by combining the detection areas obtained by the inter-frame difference and the background difference processing is optimal, a plurality of detection areas are obtained as described in the description of step S53 in FIG. Therefore, with respect to these, the optimum area is determined using the method described with reference to FIGS. 9 to 12 (S57).
[0043]
Thus, by the processing after step S54, the detection area obtained by only the inter-frame difference, the detection area obtained by only the background difference processing, and the detection area obtained by combining the detection area by the inter-frame difference and the background difference processing are obtained. In addition to selecting a region by a predetermined method according to which one is optimal, when each of the methods is selected, comparison and selection of regions, combination and selection of an optimal region are further performed among them. A suitable region will be selected.
[0044]
In the above-described embodiment, the optimization processing (S6) in FIG. 4 performs the processing as described with reference to FIGS. 9 to 12. However, the processing is not performed, and steps S2, S3, and S3 are performed. One may be selected from the plurality of regions obtained in S53.
[0045]
【The invention's effect】
As described above, according to the present invention, the congestion state of the monitoring target is evaluated based on the results obtained from the inter-frame difference and the background difference, and the quality of image data obtained from the camera is evaluated based on the video signal. Then, based on the evaluation result, the detection area is obtained by selecting whether to use the detection area obtained only by the inter-frame difference or to use the detection area obtained by the inter-frame difference and the background difference. Depending on the external environment where the brightness changes with time or the environment such as a tunnel with sparse lighting and spatial brightness distribution, the area where the monitoring target exists is appropriately specified, and the type of the object and When the situation is image-recognized, a highly accurate result can be guaranteed.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a functional configuration of an image processing apparatus according to the present invention.
FIG. 2 is a configuration diagram of a road monitoring system configured using the image processing device according to the present invention.
FIG. 3 is a block diagram showing a practical configuration example of an image processing apparatus according to the present invention.
FIG. 4 is a flowchart for explaining the operation of the image processing apparatus according to the present invention.
FIG. 5 is a flowchart showing details of a main part of the flowchart shown in FIG. 4;
FIG. 6 is a flowchart showing details of a main part of the flowchart shown in FIG. 4;
FIG. 7 is a flowchart showing details of a main part of the flowchart shown in FIG. 4;
FIG. 8 is a flowchart showing details of a main part of the flowchart shown in FIG. 4;
FIG. 9 is a diagram showing a main part of an area specifying process result for describing image processing by the image processing apparatus according to the present invention.
FIG. 10 is a diagram showing a main part of an area specifying process result for describing image processing by the image processing apparatus according to the present invention.
FIG. 11 is a diagram showing a main part of an area specifying process result for describing image processing by the image processing apparatus according to the present invention.
FIG. 12 is a diagram showing a process for obtaining a final monitoring target object region by combining a plurality of region specifying processing results by the image processing apparatus according to the present invention.
FIG. 13 is a flowchart showing details of a main part of the flowchart shown in FIG. 4;
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Network 2 Image processing part computer 3 Integration processing part computer 4 Event determination part computer 5-1 to 5-n Notification board 6-1 to 6-m Camera 10 Image acquisition means 11 First detection means 12 Second detection means 13 evaluation means 14 selection means 15 optimizing means

Claims (9)

Image acquisition means for capturing monitoring target image data obtained by the camera,
A first detection unit that performs an inter-frame difference on the image data captured by the image acquisition unit to detect an object region;
A second detection unit that performs object subtraction by performing background subtraction on the image data captured by the image acquisition unit;
Evaluating means for evaluating the congestion state of the monitoring target based on the results obtained by the first detecting means and the second detecting means and evaluating the quality of image data obtained from the camera based on a video signal When,
Based on the evaluation result by the evaluation means, a selection is made as to whether to use a detection area obtained only by the first detection means or to use a detection area obtained by the first detection means and the second detection means. An image processing system comprising: a selection unit. 2. The image processing system according to claim 1, further comprising an optimizing unit that selects and combines the detection regions selected by the selecting unit to optimize the region shape. 3. In the first detecting means, the interval between frames for obtaining the difference is changed, and in the second detecting means, the background image is changed to obtain a plurality of detection areas.
The image processing system according to claim 2, wherein the optimizing unit selects and combines the plurality of detection areas to optimize the area shape. An image acquisition step of capturing monitoring target image data obtained by a camera;
A first detection step of performing an inter-frame difference on the image data captured in the image acquisition step to detect an object area;
A second detection step of performing a background subtraction on the image data captured in the image acquisition step to detect an object area;
An evaluation step of evaluating the congestion state of the monitoring target based on the results obtained in the first detection step and the second detection step, and evaluating the quality of image data obtained from the camera based on a video signal When,
Based on the evaluation result of the evaluation step, a selection is made as to whether to use a detection area obtained only by the first detection step or to use a detection area obtained by the first detection step and the second detection step. And a selecting step. 5. The image processing method according to claim 4, further comprising an optimization step of selecting and combining the detection areas selected in the selection step to optimize an area shape. In the first detection step, a plurality of detection areas are obtained by changing an interval between frames for obtaining a difference, and in the second detection step, by changing a background image.
6. The image processing method according to claim 5, wherein in the optimizing step, the plurality of detection areas are selected and combined to optimize the area shape. In an image processing program used in a computer that performs processing by a central processing unit based on a program stored in a main memory and obtains a result,
For the central processing unit,
An image acquisition step of capturing monitoring target image data obtained by a camera;
A first detection step of performing an inter-frame difference on the image data captured in the image acquisition step to detect an object area;
A second detection step of performing a background subtraction on the image data captured in the image acquisition step to detect an object area;
An evaluation step of evaluating the congestion state of the monitoring target based on the results obtained in the first detection step and the second detection step, and evaluating the quality of image data obtained from the camera based on a video signal When,
Based on the evaluation result of the evaluation step, a selection is made as to whether to use a detection area obtained only by the first detection step or to use a detection area obtained by the first detection step and the second detection step. An image processing program for executing a selecting step. The computer-readable storage medium according to claim 7, further comprising executing an optimization step of selecting and combining the detection areas selected in the selection step to optimize an area shape. In the first detection step, a plurality of detection areas are obtained by changing an interval between frames for obtaining a difference, and in the second detection step, by changing a background image.
9. The image processing method according to claim 8, wherein in the optimizing step, the plurality of detection areas are selected and combined to optimize the area shape.

Images