A background modeling method, device and computer storage medium

Abstract

The application provides a background modeling method and device and a computer storage medium. The background modeling method comprises the following steps: acquiring a plurality of video frames of a traffic scene, and acquiring a foreground region of each video frame; determining a background region of each video frame based on the foreground region of each video frame, and acquiring a pixel value parameter of the background region; and fusing the pixel value parameters of the plurality of video frames to form a background picture. In the manner, the background region is determined based on the foreground region, the background part hidden by the object in the foreground region can be restored, and a pure background picture is obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of image processing, in particular to a background modeling method and device and computer storage medium. BACKGROUND

[0002] With the rapid development of society, traffic travel becomes more and more convenient, but the traffic scene also becomes more and more complex, and traffic events occur from time to time.

[0003] Traffic marking detection is one of the main businesses of traffic event detection, and a large number of technologies can realize lane line detection. However, in real traffic scenes, traffic markings are often blocked by motor vehicles, non-motor vehicles and pedestrians on the road, and the existing technologies that identify frame by frame can only restore part of the traffic markings that are not blocked, resulting in the inability to determine whether a vehicle runs a red light or illegally crosses a line and other illegal behaviors according to the traffic markings. Video frame result fusion through pre-recognition and post-fusion can solve the problem of occlusion to some extent, but the fundamental problem is that the target is blocked, resulting in unstable recognition results and easy misidentification or missed detection. SUMMARY

[0004] To solve the above technical problems, the present application provides a background modeling method, device and computer storage medium.

[0005] To solve the above technical problems, the present application provides a background modeling method, which comprises:

[0006] Obtain a plurality of video frames of a traffic scene, and obtain the foreground area of each video frame; determine the background area of each video frame based on the foreground area of each video frame, and obtain the pixel value parameter of the background area; fuse the pixel value parameters of the plurality of video frames to form a background picture.

[0007] Wherein, the foreground area of each video frame is obtained by using target detection or target segmentation to obtain the area where the moving target is located; each individual area or connected area where the moving target is located is taken as the foreground area of each video frame; wherein the moving target includes moving pedestrians, non-motor vehicles and / or motor vehicles.

[0008] Wherein, the pixel value parameters of the plurality of video frames are fused to form a background picture, which comprises: obtaining the pixel value parameter of each pixel point position; obtaining the pixel value mean of the pixel value parameter, and taking the pixel value mean as the final pixel value of the background picture at the pixel point position.

[0009] Wherein, after obtaining the pixel value parameter of each pixel point position, the background modeling method further comprises: judging whether the pixel value parameter of each pixel point position of the background picture is empty; if not, continue to obtain the supplementary video frame of the traffic scene until the pixel value parameter of the background region of the supplementary video frame is combined, and the pixel value parameter of each pixel point position of the background picture is not empty.

[0010] Wherein, the determining the background region of each video frame based on the foreground region of each video frame, and obtaining the pixel value parameter of the background region, comprises: obtaining the pixel value parameter of the background region of the first frame video frame according to the time sequence of the plurality of video frames; performing an added pixel value parameter obtaining operation on the subsequent video frames to obtain the pixel value parameter of the added background region of each video frame after the first frame video frame; the added pixel value parameter obtaining operation comprises: comparing the background region of the previous frame video frame and the background region of the current frame video frame to obtain the added background region of the current frame video frame; obtaining the pixel value parameter of the added background region of the current frame video frame; wherein, the added background region is the region in which the previous frame video frame is the foreground region and the current frame video frame is the background region.

[0011] Wherein, the obtaining the plurality of video frames of the traffic scene comprises: obtaining a traffic scene video; equally dividing the traffic scene video along the time axis into a plurality of video segments; and cutting the same number of video frames from each video segment to form the plurality of video frames of the traffic scene.

[0012] Wherein, after fusing the pixel value parameters of the plurality of video frames to form a background picture, the background modeling method further comprises: using target detection or target segmentation to obtain the markers and / or traffic markings in the background picture; and labeling the markers and / or the traffic markings on the background picture.

[0013] Wherein, after using target detection or target segmentation to obtain the markers and / or traffic markings in the background picture, the background modeling method further comprises: generating a lane region in the background picture based on the traffic markings and a preset lane logic; and dividing the lane region into a drivable region and a non-drivable region based on the markers and a preset traffic rule logic.

[0014] To solve the above technical problems, the present application provides a background modeling device, which comprises a memory and a processor coupled with the memory; wherein the memory is used to store program data, and the processor is used to execute the program data to realize the above background modeling method.

[0015] To solve the above technical problems, the application further provides a computer storage medium, which is used for storing program data, and the program data is used to realize the background modeling method when being executed by a computer.

[0016] Compared with the prior art, the application has the beneficial effects that the background modeling device acquires a plurality of video frames of a traffic scene, and acquires a foreground region of each video frame; based on the foreground region of each video frame, the background region of each video frame is determined, and a pixel value parameter of the background region is acquired; the pixel value parameters of the plurality of video frames are fused to form a background picture. Through the above manner, the background part blocked by the object in the foreground region is restored to obtain a pure background picture by determining the background region based on the foreground region. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0018] Among them:

[0019] Figure 1 is a flowchart of the first embodiment of the background modeling method provided by the application;

[0020] Figure 2 is a flowchart of the sub-step of step S11 in the first embodiment of the background modeling method provided by the application;

[0021] Figure 3 is a flowchart of the second embodiment of the background modeling method provided by the application;

[0022] Figure 4 is a flowchart of the sub-step of step S13 in the first embodiment provided by the application;

[0023] Figure 5 is a flowchart of the third embodiment of the background modeling method provided by the application;

[0024] Figure 6 is a flowchart of the whole process of the background modeling method provided by the application;

[0025] Figure 7 is a structural diagram of an embodiment of the background modeling device provided by the application;

[0026] Figure 8 is a structural diagram of an embodiment of the computer storage medium provided by the application. DETAILED DESCRIPTION

[0027] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0028] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the application described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0029] Please refer to details. Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the background modeling method provided in this application.

[0030] The background modeling method of this application is applied to a background modeling device, wherein the background modeling device can be a server, a local terminal, or a system in which the server and the local terminal cooperate with each other. Accordingly, the various parts of the background modeling device, such as various units, sub-units, modules, and sub-modules, can all be set in the server, all in the local terminal, or separately in the server and the local terminal.

[0031] Furthermore, the aforementioned server can be either hardware or software. When the server is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or as a single server. When the server is software, it can be implemented as multiple software programs or software modules, such as software or software modules used to provide distributed server functionality, or as a single software program or software module; no specific limitations are made here.

[0032] like Figure 1 As shown, the specific steps are as follows:

[0033] Step S11: Acquire several video frames of the traffic scene and acquire the foreground region of each video frame.

[0034] Specifically, the background modeling device obtains a traffic scene video recording from a road monitoring video or a road monitoring video database, equally divides the traffic scene video into a plurality of video segments along a time axis, takes the same number of video frames from each video segment, and forms a plurality of video frames of the traffic scene. The traffic scene video segments facilitate further inter-frame processing.

[0035] Further, the present application provides a sub-step of step S11 for obtaining a foreground region of each video frame. For details, see Figure 2 , Figure 2 FIG. 1 is a flowchart of a sub-step of step S11 in the first embodiment of the background modeling method provided by the present application.

[0036] As shown in Figure 2 , the specific steps are as follows:

[0037] Step S111: using target detection or target segmentation to obtain a region where a moving target of each video frame is located.

[0038] The moving target includes a moving change pedestrian, a non-motor vehicle, and / or a motor vehicle.

[0039] Specifically, the background modeling device uses a Gaussian mixture model, a frame difference processing model, or a motion object recognition deep neural network model to detect targets from the plurality of video frames and obtain a region where a moving target of each video frame is located.

[0040] In an embodiment of the present application, the size of the region where the moving target is located is determined according to the size of the moving target, and the region can be marked by a target detection box or a coordinate box. The background modeling device can change the size of the target detection box to enlarge or reduce the size of the region where the moving target is located by using a preset index or in response to a user control instruction.

[0041] Step S112: taking each individual region or connected region where a moving target is located as a foreground region of each video frame.

[0042] Specifically, in an embodiment of the present application, the background modeling device divides each region including a moving target as an individual region, and each foreground region obtained contains one moving target. The foreground region of each video frame can include a plurality of individual regions, i.e., a plurality of moving targets. The moving object can be a pedestrian, a motor vehicle, or a non-motor vehicle, or an independent whole composed of a pedestrian and a non-motor vehicle. In other embodiments of the present application, when the distance between the moving targets in the video frame exceeds a preset distance or overlap occurs, the background modeling device connects the regions where the moving targets are located and takes the connected regions as the foreground region of each video frame.

[0043] The foreground region can be one or more.

[0044] Step S12: determining the background region of each video frame based on the foreground region of each video frame, and obtaining the pixel value parameter of the background region.

[0045] Specifically, the background modeling device removes the foreground region of each video frame by calculating through a Gaussian mixture model, a frame difference calculation model or a background detection deep neural network model, determines the part removed from the foreground region as the background region of each video frame, and obtains the pixel value parameter of the background region.

[0046] The pixel value parameter is a pixel value set of all constituent pixel points of the background region of the first video frame.

[0047] The background region includes but is not limited to all parts removed from the foreground, and also includes a blocked background part in the foreground region which is misrecognized into the foreground region due to the blocking of a moving target. Further, an embodiment is proposed in the present application to specifically restore the background region blocked by the moving target to the original position, please refer to Figure 3 , Figure 3 is a flowchart of a second embodiment of the background modeling method proposed in the present application.

[0048] As shown in Figure 3 , the specific steps are as follows:

[0049] Step S21: obtaining the pixel value parameter of the background region of the first video frame in the time sequence of a plurality of video frames.

[0050] Specifically, the background modeling device obtains the pixel value parameter of the background region of the first video frame by calculation in the time sequence of a plurality of video frames. It should be noted that the first video frame can be the first frame at which the traffic scene video starts in time, or can be the starting frame from which the background image is to be extracted, which is not specifically limited here.

[0051] The pixel value parameter is a pixel value set of all constituent pixel points of the background region of the first video frame.

[0052] Step S22: performing an added pixel value parameter obtaining operation on the subsequent video frames to obtain the pixel value parameter of the added background region of each video frame after the first video frame.

[0053] The added background region is a region in which the previous video frame is the foreground region and the current video frame is the background region.

[0054] Specifically, the background modeling device performs an added pixel obtaining operation on the subsequent video frames of the first video frame, and the added pixel obtaining operation includes the following steps:

[0055] The background modeling device compares the background region of the previous frame of video frames with the background region of the current frame of video frames to obtain the newly added background region of the current frame of video frames.

[0056] Specifically, the foreground target region is input into the background modeling device, and the detected foreground region is directly ignored by the background modeling device during the background modeling process, and no parameter updating is performed. After the next frame of image is input, the foreground region changes, and therefore the part of the region that is foreground in the previous frame and is not foreground in this frame is the newly added background region.

[0057] The background modeling device obtains the pixel value parameter of the newly added background region of the current frame of video frames.

[0058] Specifically, the background modeling device calculates the newly added background region of the current frame of video frames to obtain the pixel value parameter of the newly added background region of the current frame of video frames.

[0059] Further, the background modeling device adds the pixel value parameter of the newly added background region of the current frame of video frames to the pixel point at the same position in the previous frame of video frames. Due to the shielding of the foreground, the pixel point at the same position in the previous frame of video frames does not have a pixel value parameter. The pixel value parameter of the newly added background region is added to the position where the pixel value is empty in the previous frame of video frames, so that the pixel value of the newly added background region is filled back to the previous frame of video frames, and the shielded background region is supplemented. After a certain time of iteration, the background model will tend to be complete, that is, there is no pixel point with an empty pixel value in the background image. At this time, the background modeling ends, and the background modeling obtains a pure and complete background picture from the background model.

[0060] For example, the pixel value parameter of the newly added background region of the current frame is 180, and the pixel value parameter of the background region of the previous frame of video frames is empty. Therefore, the pixel value parameter of the shielded background region is 180.

[0061] Through steps S21-S22, the newly added background region is obtained by comparing the background region of the previous frame of video frames with the background region of the current frame of video frames, and the pixel value parameter of the moving target shielded in the foreground region can be effectively extracted. Only the pixel value parameter shielded by the moving target is recovered and fused, and there is no need to update each pixel value in the background image, thereby improving the operation efficiency of the background modeling device and reducing the operation amount of the background modeling device.

[0062] Step S13: The pixel value parameters of a plurality of video frames are fused to form a frame of background pictures.

[0063] Specifically, the background modeling device calculates by using a fusion algorithm, a weighting algorithm, or a mixture Gaussian model to fuse the pixel values of a plurality of video frames to form a pure background picture.

[0064] The pixel value parameters of several video frames are fused, so that each pixel point of the formed background picture has a pixel value parameter, that is, the background picture does not have an occluded area.

[0065] Further, in an embodiment of the present application, a mixed Gaussian model is established for each background pixel point to determine the final pixel value by the mean value. For details, please refer to Figure 4 , Figure 4 is a sub-step flowchart of step S13 in the first embodiment provided by the present application.

[0066] As Figure 4 shown, the specific steps are as follows:

[0067] Step S131: Obtain the pixel value parameter of each pixel point position.

[0068] Specifically, the background modeling device obtains the pixel value parameter of each pixel point position in the background picture extracted from each video frame.

[0069] Further, after obtaining the pixel value parameter of each pixel point position, it is determined whether the pixel value parameter of each pixel point position of the background picture is empty; if it is empty, continue to obtain the supplementary video frame of the traffic scene until the pixel value parameter of the background area of the supplementary video frame is combined, and the pixel value parameter of each pixel point position of the background picture is not empty.

[0070] Step S132: Obtain the pixel value mean of the pixel value parameter, and take the pixel value mean as the final pixel value of the background picture at the pixel point position.

[0071] Specifically, the background modeling device performs weighted average operation on the pixel value parameters of the same pixel point position in each background picture to obtain the pixel value mean of the pixel point position, and takes the pixel value mean as the final pixel value of the background picture at the pixel point position.

[0072] Through steps S131-S132, the final pixel value of the pixel point position is determined by the pixel value mean of the pixel value parameters of the pixel point position of several video frames, the occluded part is restored to the background picture through multi-frame average processing, and a pure and complete background picture is further extracted.

[0073] In the above manner, the background part occluded by the object in the foreground region is restored to obtain a pure background picture by determining the background region through the foreground region.

[0074] In order to further restore the traffic scene by the background modeling method, the present application further provides an embodiment, please refer to Figure 5 , Figure 5 is a flowchart of the third embodiment of the background modeling method provided by the present application.

[0075] As Figure 5 shown, the specific steps are as follows:

[0076] Step S31: Obtain the markers and / or traffic markings in the background picture using target detection or target segmentation.

[0077] Specifically, target detection or target segmentation is performed on the markers or traffic markings or both on the pure background picture, and the results can be represented by target detection boxes or masks.

[0078] The markers include but are not limited to guide arrows, guide strips, road shoulders, guide strips, etc. The traffic markings include but are not limited to yellow and white solid lines, yellow and white dashed lines, yellow and white dashed and solid lines, yellow and white double lines, etc. Lane lines and zebra crossings, stop lines, etc. Marker detection and traffic marking detection can be achieved by target detection or image segmentation methods. The target detection methods include but are not limited to YOLO, Faster R-CNN, etc. The image segmentation methods include but are not limited to Mask R-CNN, SpatialEmbeddings, and end-to-end method PINet, etc.

[0079] After step S31, the following steps are further included: based on the traffic markings and the preset lane logic, generating a lane area in the background picture.

[0080] The preset lane logic is set by the background modeling device according to the setting rules of the traffic road. For example: the lane is composed of left and right lane lines and stop lines.

[0081] Based on the markers and the preset traffic rule logic, the lane area is divided into drivable areas and non-drivable areas.

[0082] The markers include but are not limited to road signs, road barriers, etc. The preset traffic rule logic can include that according to road conditions and traffic needs, the road is divided into motor vehicle lanes, non-motor vehicle lanes, and pedestrian lanes, and motor vehicles, non-motor vehicles, and pedestrians implement separate traffic. Without dividing motor vehicle lanes, non-motor vehicle lanes, and pedestrian lanes, motor vehicles travel in the middle of the road, and non-motor vehicles and pedestrians travel on both sides of the road, etc.

[0083] Step S32: Label the markers and / or traffic markings on the background picture.

[0084] Specifically, the application scenarios of background modeling include but are not limited to crossroads and road sections. The background modeling device outputs the final structured scene information by labeling the markers and / or traffic markings on the background picture.

[0085] The background modeling device obtains semantic information such as drivable areas, non-drivable areas, lanes, lane-to-lane relationships, and the like in a scene. The reconstruction combines the basic elements recognized in step S31, such as lane lines, zebra crossings, road shoulders, and the like, through logical combination according to prior knowledge known by the national standard, combines, deletes, supplements, and the like of lane lines and road boundaries through the logic that a lane is composed of a left lane line and a right lane line and a stop line to obtain a lane, and further reconstructs drivable areas and non-drivable areas through spatial and logical relationships between different areas, thereby outputting final structured scene information.

[0086] Through steps S31-S32, the original background in the target scene can be obtained, and the real scene can be restored to the greatest extent. After the background picture is obtained, no lane completion operation or the like needs to be performed.

[0087] As shown in Figure 6 , Figure 6 is a schematic diagram of the overall process of the background modeling method provided by the present application.

[0088] As shown in Figure 6 , the specific process is as follows:

[0089] If the pixel value of each pixel point position is empty, the supplementary video frame of the traffic scene is continuously obtained until the pixel value parameter of each pixel point position of the background picture is not empty in combination with the pixel value parameter of the background area of the supplementary video frame.

[0090] Specifically, the background modeling device inputs the foreground target area into the background modeling, and in the process of the background modeling, the detected foreground area is directly ignored and no parameter updating is performed. After the next frame of image is input, the foreground area has changed, and therefore the parameter updating is performed on the area that was foreground in the last frame but is not foreground in this frame. After a certain time of iteration, the background model will tend to be complete, the background modeling is ended, and the background picture is obtained from the background model.

[0091] In the above manner, target detection and reconstruction are not performed on each frame of image, and the lane line scene is reconstructed only after the pure background image is obtained through the background modeling, and therefore the time consumption is short.

[0092] To implement the background modeling method described above, the present application further provides a background modeling device, and the specific process is described below with reference to Figure 7 , Figure 7 is a structural schematic diagram of an embodiment of the background modeling device provided by the present application.

[0093] The background modeling device 400 of the embodiment includes a processor 41, a memory 42, an input / output device 43, and a bus 44.

[0094] The processor 41, the memory 42 and the input and output device 43 are connected to the bus 44, the memory 42 stores program data, and the processor 41 is used to execute the program data to realize the background modeling method in the above embodiment.

[0095] In the embodiment of the present application, the processor 41 can also be referred to as a CPU (Central Processing Unit). The processor 41 can be an integrated circuit chip with a signal processing capability. The processor 41 can also be a general-purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor or the processor 41 can also be any conventional processor.

[0096] The present application also provides a computer storage medium, please continue to refer to Figure 8 , Figure 8 is a structural schematic diagram of an embodiment of the computer storage medium provided by the present application. The computer storage medium 500 stores a computer program 51, and the computer program 51 is used to realize the background modeling method in the above embodiment when executed by a processor.

[0097] When the embodiment of the present application is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the whole or part of the technical solutions that essentially contribute to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the method described in the various embodiments of the present application. The foregoing storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0098] The above merely describes the embodiments of the present application, and does not limit the patent scope of the present application. Equivalent structures or equivalent process transformations made by the content of the present application specification and drawings, or direct or indirect application in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A background modeling method, characterized in that, The background modeling method includes: Acquire several video frames of the traffic scene and obtain the foreground region of each video frame; Based on the foreground region of each video frame, determine the background region of each video frame, and obtain the pixel value parameters of the background region; The pixel value parameters of the aforementioned video frames are merged to form a background image. The process of determining the background region of each video frame based on the foreground region of each video frame and obtaining the pixel value parameters of the background region includes: According to the time sequence of the aforementioned video frames, obtain the pixel value parameters of the background region of the first video frame; Perform an operation to obtain newly added pixel value parameters for subsequent video frames, so as to obtain the pixel value parameters of the newly added background area in each video frame after the first video frame; The operation of obtaining the new pixel value parameter includes: Compare the background area of ​​the previous video frame with the background area of ​​the current video frame to obtain the newly added background area of ​​the current video frame; obtain the pixel value parameters of the newly added background area of ​​the current video frame. The newly added background area is the area where the previous video frame is the foreground area and the current video frame is the background area.

2. The background modeling method according to claim 1, characterized in that, The step of obtaining the foreground region of each video frame includes: The region where the moving target is located in each video frame is obtained using target detection or target segmentation. The individual region or connected region where each moving target is located is taken as the foreground region of each video frame. The moving targets include pedestrians, non-motorized vehicles, and / or motorized vehicles in motion.

3. The background modeling method according to claim 1, characterized in that, The step of fusing the pixel value parameters of the plurality of video frames to form a background image includes: Get the pixel value parameter at the position of each pixel; Obtain the average pixel value of the pixel value parameter, and use the average pixel value as the final pixel value of the background image at that pixel position.

4. The background modeling method according to claim 3, characterized in that, After obtaining the pixel value parameters of each pixel location, the background modeling method further includes: Determine whether the pixel value parameter at each pixel position of the background image is not empty; If not, continue acquiring supplementary video frames of the traffic scene until, when combined with the pixel value parameters of the background area of ​​the supplementary video frames, the pixel value parameters of each pixel position in the background image are not empty.

5. The background modeling method according to claim 1, characterized in that, The acquisition of several video frames of the traffic scene includes: Acquire traffic scene videos; The traffic scene video is equally divided into several video segments along the time axis; The same number of video frames are extracted from each video segment to form several video frames of the traffic scene.

6. The background modeling method according to claim 1, characterized in that, After fusing the pixel value parameters of the several video frames to form a background image, the background modeling method further includes: Use object detection or object segmentation to obtain landmarks and / or traffic markings in the background image; The sign and / or the traffic markings are marked on the background image.

7. The background modeling method according to claim 6, characterized in that, After obtaining landmarks and / or traffic markings in the background image using object detection or object segmentation, the background modeling method further includes: Based on the traffic markings and preset lane logic, a lane area is generated in the background image; Based on the aforementioned signs and preset traffic rule logic, the lane area is divided into a drivable area and a non-drivable area.

8. A background modeling device, characterized in that, The background modeling device includes a memory and a processor coupled to the memory; The memory is used to store program data, and the processor is used to execute the program data to implement the background modeling method as described in any one of claims 1 to 7.

9. A computer storage medium, characterized in that, The computer storage medium is used to store program data, which, when executed by the computer, is used to implement the background modeling method as described in any one of claims 1 to 7.

Images