Photographing device and photographing method

Abstract

The present invention relates to the field of shooting equipment control, and in particular to a shooting equipment and a shooting method. The shooting equipment comprises: a shooting combination unit, comprising m groups of shooting modules, and the shooting combination unit is used to shoot a monitoring area and obtain a shot image. A shooting feature picking unit is used to identify shooting features in the shot image. A feature movement identification unit is used to extract a target changing image area and a non-changing image area from a target image set, and one of the shooting modules moves, enlarges, focuses and shoots the target changing image area, and binds and stores the target enlarged image with the corresponding target image. A non-changing image area segmentation unit is used to extract the coverage area coordinates of the non-changing image area, and encode the non-changing image area. By binding and storing the target enlarged image with the corresponding target image, intelligent monitoring can be achieved by viewing the target enlarged image while viewing the target image.

Description

Technical Field

[0001] The present invention relates to the field of photographing equipment control, and in particular to a photographing equipment and a photographing method. Background Art

[0002] With the development of science and technology, people have entered the information age, and monitoring equipment has entered people's daily life and work. A typical TV monitoring system is mainly composed of two parts: front-end equipment and back-end equipment. The front-end equipment usually consists of components such as cameras, manual or electric lenses, pan / tilt heads, protective covers, monitors, alarm detectors and multi-function decoders. They each perform their duties and establish corresponding connections with various devices in the central control system through wired, wireless or optical fiber transmission media (transmitting video / audio signals and control and alarm signals). In actual TV monitoring systems, these front-end devices are not necessarily used at the same time, but cameras and lenses for monitoring on-site image acquisition are indispensable. Back-end equipment can be further divided into central control equipment and sub-control equipment.

[0003] Surveillance equipment is an electronic device used to monitor scenes, usually including cameras, displays, storage devices, remote control devices and other parts. Camera: used to capture image information in the scene, usually closed-circuit television (CCTV), hard disk recorder (DVR) and other types. Display: used to display monitoring images, usually LCD screen. Storage device: used to store monitoring images and data, usually hard disk recorder (DVR) or digital video storage device (NVR). Remote control device: used to remotely control the operation of monitoring equipment, usually wireless transmission equipment, such as mobile phones, tablets, etc. Cloud storage: used to store monitoring data, usually cloud storage services, such as Google Drive, Dropbox, etc. Alarm device: used to monitor abnormal conditions in the scene, usually sound and light alarm devices, such as smoke alarms, carbon monoxide alarms, etc. Through monitoring equipment, you can understand the situation of the scene in real time, discover and deal with problems in time, and ensure the safety and normal operation of the scene.

[0004] Existing camera equipment can only perform conventional shooting and cannot be magnified for monitoring as needed, resulting in the loss of some details and a decrease in monitoring quality. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides a shooting device for shooting a monitoring area, enlarging and tracking it, and then combining and storing it, so as to improve the monitoring quality. The shooting device includes:

[0006] The shooting combination unit includes m groups of shooting modules, where m≥2. The shooting combination unit is used to shoot the monitoring area and obtain the shooting image. The m groups of shooting modules form m-degree shooting overlap for the same monitoring area. The shooting module can be a camera, a high-frequency camera, etc., that is, the shooting of the monitoring area to obtain the shooting image is the monitoring video. The m-degree shooting overlap is that the m groups of shooting modules shoot a monitoring area, and the objects shot by the m shooting modules are the same area. Therefore, when it is fully turned on, the shooting images of this area are also m, and these m shooting images overlap and are defined as m-degree shooting overlap. The shooting module can be a camera or a camera array, which can be specifically determined according to the area of ​​the monitoring area and the coverage area of ​​the camera. If the area of ​​the monitoring area is less than or equal to the coverage area of ​​the camera, only one camera is needed. If the area of ​​the monitoring area is larger than the coverage area of ​​the camera, a combination of multiple cameras is required to cover and lay the monitoring area. When multiple cameras are required to be laid out in combination, the camera array can be set according to the shooting density. The shooting density is not laid out uniformly in a conventional manner, and areas or obstacles that do not need to be monitored need to be avoided, so that the shooting module can be used to the maximum extent and waste can be avoided. The shooting module is a camera with automatic zooming and focusing of the shooting field of view, which can focus on a specific area as needed to increase the specific shooting clarity within the monitoring area.

[0007] The shooting feature picking unit is used to identify shooting features in the shooting images and collect the shooting images with shooting features as the target image set. The shooting features can be set according to the specific monitoring scene, and the shooting features can be set manually in advance. The identification of shooting features is a prior art and will not be described in detail here.

[0008] The feature movement recognition unit is used to extract the target changing image area and the non-changing image area from the target image set, and one of the shooting modules moves the target changing image area, zooms in and focuses on the shooting to obtain the target enlarged image and stores it to obtain the moving target enlarged set, and binds the target enlarged image with the corresponding target image for storage.

[0009] The non-changing image area segmentation unit is used to extract the coverage area coordinates of the non-changing image area, encode the non-changing image area, and bind the original non-changing image area picture frame, the coverage area coordinates of the non-changing image area and the code for storage. Then the image we finally store is the code (coverage area coordinates) corresponding to the non-changing image area, the original non-changing image area picture frame and the target changing image area. When image extraction is required, the code (coverage area coordinates) corresponding to the non-changing image area, the original non-changing image area picture frame and the target changing image area can be reversely translated and combined according to the coordinates to obtain a complete image.

[0010] Preferably: the shooting module needs to be an optical zoom camera or camera, and zooming in will not affect the pixels, because optical zoom relies on lens shift, and the details are not repeated here. The m is preferably 2, but other numbers are not excluded. The larger the number of m, the stronger the monitoring capability, but the greater the investment cost of the equipment. Of course, the number of m can also be designed according to actual monitoring needs. The number of shooting module groups can be m=[[1+sa]exp(c)]+1, where s is the maximum number of detection points in the monitoring area, such as the number of possible parabolic points in the building in the monitoring of high-altitude parabolic objects, the total number of the largest flow of people in the street monitoring area or the maximum applicable number. The a is the overlap occurrence factor, and its value is related to the frequency of occurrence. It can be designed by statistically analyzing the frequency of occurrence. The higher the frequency of occurrence, the larger the value of a. c is a monitoring level parameter, which can be set manually. When there is no abnormal situation for monitoring, the m groups of shooting modules can open one group for routine monitoring. When an abnormality occurs, other groups of shooting modules can be quickly opened. The number of openings needs to be determined according to the number of targets in the monitoring area, and the specific details are not repeated here. The pixels of the shooting modules can be the same or different, and different is preferred. By setting the shooting modules with different pixels, different levels of monitoring can be performed by turning on the shooting modules with different pixels. The number of levels can be m. By monitoring at different levels, the flow and storage space can be reduced. Since it is not the subject of protection of this application, it will not be described in detail here.

[0011] Preferably: the target change image area extraction method may include: sorting and separating each picture frame from the target image according to time, identifying k marking points from the picture frame, k ≥ 3, calculating the distance between each marking point in the previous picture frame, and scaling the next picture frame according to the distance of the marking point in the previous picture frame, so that the k marking points in the picture frame overlap and completely coincide, and then calculating the grayscale value of each pixel point domain, and comparing the grayscale values ​​of the same position of the previous and next picture frames to obtain the grayscale value difference, the area where the grayscale value difference is greater than a preset standard deviation value is defined as the coverage area, and the coverage area is defined as the target change image area. The preset standard deviation value can be pre-set, and its value needs to take into account factors such as device clarity and scene. Under ideal conditions, the preset standard deviation value is 0, and other situations are not described here. The grayscale value calculation is a prior art, and the details are not described here.

[0012] Preferably, the method for obtaining the marking point may include: constructing a plane coordinate system, and implanting the picture frame into the plane coordinate system, wherein the horizontal direction of the picture frame can be used as the x-axis, and the vertical direction of the picture frame can be used as the y-axis, processing the picture frame to obtain the gray value, and using the gray value as the z-axis for dimensionality increase processing, and calculating and The slope points of the texture line are connected to obtain the texture line. The point is taken as the extreme point, k extreme points are selected from the extreme points, and the calculation Where i and j represent the numbers of the extreme points. And calculate Max{l k} corresponding to k extreme value points, these k extreme value points are used as marking points. z' is a preset slope standard value of generally 50-100, which can be calculated based on the variance of the grayscale value of the picture frame, mainly to distinguish the lines with large color difference, which is not the subject of this application, and will not be described in detail here. By finding the texture lines of each picture frame, the deviation caused by slight differences in light or angle can be minimized, the marking points can be better identified, and the marking points can be constructed by the maximum distance, which can minimize the error. The method of scaling according to the distance of the marking point is a prior art and will not be described in detail here. In the m groups of shooting modules, one group of shooting modules continues to shoot, wherein a group of shooting modules in the m-1 group zooms in and focuses on the target changing image area, and the remaining m-2 can track and monitor other moving targets, so that comprehensive monitoring and shooting can be performed. The shooting module tracking and mobile shooting are prior art and will not be described in detail here. The method for obtaining the non-changing image area can be to obtain the overlapping area of ​​the front and rear picture frames, and define the overlapping area minus the part of the target changing image area as the non-changing image area. The target changing image region may be an irregular covering region, and the covering region may be subjected to standardization processing to obtain the target changing image region, wherein standardization processing is preferred to facilitate segmentation and extraction.

[0013] Preferably: the method for standardizing the coverage area may include: calculating and obtaining the coordinates of four points corresponding to Max{x}, Max{y}, Min{x}, and Min{y} of the coverage area, and then drawing a line parallel to the y axis through the coordinate point corresponding to Max{x} and the coordinate point corresponding to Min{x}, and drawing a line parallel to the x axis through the coordinate point corresponding to Max{y} and the coordinate point corresponding to Min{y}, and the rectangular frame area formed by these four parallel lines is the target variable image area. The shooting module moves the center of the shooting area to the center position in the rectangular frame area, and scales the frame of the shooting area of ​​the shooting module to coincide with the boundary line of the rectangular frame area.

[0014] Preferably: the method for scaling the frame of the shooting area of ​​the shooting module to coincide with the boundary line of the rectangular frame area comprises: calculating the length of the vertical side Ly = Max{y}-Min{y}, the length of the horizontal side Lx = Max{x}-Min{x}, and calculating the aspect ratio Is it greater than or equal to the device aspect ratio? If so, the shooting module is scaled vertically; if not, the shooting module is scaled horizontally. The device aspect ratio is the standard aspect ratio of the device for shooting images, which will not be described in detail here.

[0015] Preferably: the method for performing longitudinal scaling of the shooting module comprises: scaling the longitudinal coordinates of the shooting module according to Scaling, horizontal and horizontal proportional scaling, where Ly标 is the longitudinal length of the shooting module, and Y is the longitudinal coordinate of the shooting module.

[0016] Preferably, the method for horizontal scaling of the shooting module comprises: adjusting the horizontal coordinates of the shooting module according to Scaling, vertical and horizontal proportional scaling, where Lx 标 is the horizontal side length of the shooting module, and X is the horizontal coordinate of the shooting module. The specific calculation example is not described here.

[0017] The present invention also provides a shooting method, which comprises:

[0018] S1. Shoot the monitored area to obtain a shot image.

[0019] S2. Identify shooting features in the shot images, and collect the shot images with the shooting features as a target image set.

[0020] S3. Extract the target changed image area and the non-changed image area from the target image set.

[0021] S4, one of the shooting modules is used to move the target moving image area, zoom in and focus on the image to obtain the target zoomed image and store it to obtain a moving target zoom set.

[0022] S5. Binding and storing the target magnified image and the corresponding target image.

[0023] S6. Extract the coverage area coordinates of the non-changing image area and encode the non-changing image area.

[0024] S7. Bind and store the original non-changing image region picture frame, the non-changing image region coverage area coordinates and the code.

[0025] The technical effects and advantages of the present invention are as follows: by identifying the features and movement of the captured images, the moving target is automatically acquired and zoomed, photographed and tracked to obtain a magnified image of the target. The magnified target image is stored in conjunction with the corresponding target image, and the moving target can be automatically identified by viewing the target image at the same time, and the trajectory and details of the moving target can be observed through the mobile target magnification set, thereby realizing intelligent monitoring. During the storage process, there is no complete picture frame or video information, which increases the privacy and security of big data information. At the same time, encoding is used instead of video for storage, and the actual memory occupancy is small, which greatly saves storage space. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a structural block diagram of a shooting device proposed by the present invention.

[0027] Figure 2 The present invention is a flowchart of a shooting method. DETAILED DESCRIPTION

[0028] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present invention are provided for the purpose of illustration and description, and are not intended to be exhaustive or to limit the present invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention, and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

[0029] Example 1

[0030] refer to Figure 1 In this embodiment, a shooting device is proposed, which is used to shoot the monitoring area and enlarge and track it, and then combine and store it to improve the monitoring quality. The shooting device includes:

[0031] The shooting combination unit includes m groups of shooting modules, where m≥2. The shooting combination unit is used to shoot the monitoring area and obtain the shooting image. The m groups of shooting modules form m-degree shooting overlap for the same monitoring area. The shooting module can be a camera, a high-frequency camera, etc., that is, the shooting of the monitoring area to obtain the shooting image is the monitoring video. The m-degree shooting overlap is that the m groups of shooting modules shoot a monitoring area, and the objects shot by the m shooting modules are the same area. Therefore, when it is fully turned on, the shooting images of this area are also m, and these m shooting images overlap and are defined as m-degree shooting overlap. The shooting module can be a camera or a camera array, which can be specifically determined according to the area of ​​the monitoring area and the coverage area of ​​the camera. If the area of ​​the monitoring area is less than or equal to the coverage area of ​​the camera, only one camera is needed. If the area of ​​the monitoring area is larger than the coverage area of ​​the camera, a combination of multiple cameras is required to cover and lay the monitoring area. When multiple cameras are required to be laid in combination, the camera array can be set according to the shooting density. The shooting density is not laid uniformly according to the conventional method. It is necessary to avoid those areas or obstacles that do not need to be monitored, so that the shooting module can be used to the maximum extent and waste can be avoided. The shooting module is a camera that automatically zooms and focuses the shooting field of view. It can focus on a specific area as needed to increase the specific shooting clarity inside the monitoring area. The shooting module needs to be an optical zoom camera or camera. Zooming in will not affect the pixels because the optical zoom relies on the lens shift. The specific details are not repeated here. The m is preferably 2, but other numbers are not excluded. The larger the number of m, the stronger the monitoring capability, but the greater the investment cost of the equipment. Of course, the number of m can also be designed according to actual monitoring needs. The number of shooting module groups can be m=[[1+sa]exp(c)]+1, where s is the maximum number of detection points in the monitoring area, such as the number of possible parabolic points in the building in the monitoring of high-altitude parabolic objects, the total number of people with the largest flow of traffic in the street monitoring area or the maximum applicable number, etc. For example, if a residential building is monitored for objects thrown from high places, and the building has 100 points where objects can be thrown, that is, 100 windows, then s is 100 at this time, and the specific details are not repeated here. The a is the overlap occurrence factor, and its value is related to the frequency of occurrence. It can be designed by statistically analyzing the frequency of occurrence. The higher the frequency of occurrence, the larger the value of a. For example, the frequency of objects thrown from high places is very low, and a can be set to 0.0001 at this time. Of course, other values ​​are not excluded for calculation. Generally, the values ​​of the overlap occurrence factor are very low, so as is basically less than 1, and the specific details are not repeated here.c is a monitoring level parameter, which can be set manually. For example, the monitoring level of a general residential building is a basic monitoring level, and its monitoring level parameter is 0 (basic parameter). The monitoring level parameter is not necessarily a natural number, but can be a decimal, etc. It is set manually for convenience of calculation, and the specific value is not repeated here. [] is the integer symbol, which is the integer part of the calculated value. Through the above example, 100 windows, a can be 0.0001, and the monitoring level parameter is 0 for calculation, then the number of shooting module groups m=2 is calculated. When there is no abnormal situation for monitoring, the m groups of shooting modules can open one group for routine monitoring. When an abnormality occurs, other groups of shooting modules can be quickly opened. The number of openings needs to be determined according to the number of targets in the monitoring area, and the specific details are not repeated here. The pixels of the shooting modules can be the same or different, and different is preferred. They are set by shooting modules with different pixels. By opening shooting modules with different pixels, different levels of monitoring can be performed. The number of levels can be m. By monitoring at different levels, traffic and storage space can be reduced. Since it is not the subject of protection of this application, it is not repeated here.

[0032] The shooting feature picking unit is used to identify the shooting features in the shot images, and collect the shot images with the shooting features as the target image set. The shooting features can be set according to the specific monitoring scene, and the shooting features can be set manually in advance. The recognition of shooting features is a prior art, and the details are not described here. For example, when monitoring an office area, facial features can be extracted from people who have the authority to enter the office area, and the facial features can be used to form a personnel database. At this time, the shooting features can be facial features. Also, when monitoring high-altitude objects, the shooting features can be all moving objects. At this time, the shooting features are not specific object types or shape features, but a morphological feature, which will not be described in detail here.

[0033] The feature movement recognition unit is used to extract the target changing image area and the non-changing image area from the target image set, and one of the shooting modules moves the target changing image area, zooms in and focuses on the image to obtain the target zoomed image and stores it to obtain the moving target zoomed set, and binds the target zoomed image with the corresponding target image for storage. By binding and storing the target zoomed image with the corresponding target image, the moving target can be automatically identified by viewing the target image and the zoomed target image at the same time, and the moving target trajectory and details can be observed through the moving target zoomed set, thereby realizing intelligent monitoring. The target change image area extraction method may include: sorting and separating each picture frame from the target image according to time, identifying k marking points from the picture frame, k≥3, calculating the distance between each marking point in the previous picture frame, and scaling the next picture frame according to the distance of the marking point in the previous picture frame, so that the k marking points in the picture frame overlap and completely coincide, and then calculating the grayscale value of each pixel point domain, and comparing the grayscale values ​​of the same position of the previous and next picture frames to obtain the grayscale value difference, the area where the grayscale value difference is greater than a preset standard deviation value is defined as the coverage area, and the coverage area is defined as the target change image area. The preset standard deviation value can be pre-set, and its value needs to take into account factors such as device clarity and scene. Under ideal conditions, the preset standard deviation value is 0, and other situations are not described here. The grayscale value calculation is a prior art, and the details are not described here. The method for obtaining the marking point may include: constructing a plane coordinate system, and implanting the picture frame into the plane coordinate system, wherein the horizontal direction of the picture frame can be used as the x-axis, and the vertical direction of the picture frame can be used as the y-axis, processing the picture frame to obtain a gray value, and performing dimensionality increase processing with the gray value as the z-axis, and calculating and The slope points of the texture line are connected to obtain the texture line. The point is taken as the extreme point, k extreme points are selected from the extreme points, and the calculation Where i and j represent the numbers of the extreme points. And calculate Max{l k} corresponding to k extreme value points, these k extreme value points are used as marking points. z' is a preset slope standard value of generally 50-100, which can be calculated based on the variance of the grayscale value of the picture frame, mainly to distinguish the lines with large color difference, which is not the subject of this application, and will not be described in detail here. By finding the texture lines of each picture frame, the deviation caused by slight differences in light or angle can be minimized, the marking points can be better identified, and the marking points can be constructed by the maximum distance, which can minimize the error. The method of scaling according to the distance of the marking point is a prior art and will not be described in detail here. In the m groups of shooting modules, one group of shooting modules continues to shoot, wherein a group of shooting modules in the m-1 group zooms in and focuses on the target changing image area, and the remaining m-2 can track and monitor other moving targets, so that comprehensive monitoring and shooting can be performed. The shooting module tracking and mobile shooting are prior art and will not be described in detail here. The method for obtaining the non-changing image area can be to obtain the overlapping area of ​​the front and rear picture frames, and define the overlapping area minus the part of the target changing image area as the non-changing image area. The target changing image area may be an irregular coverage area, and the coverage area can also be standardized to obtain the target changing image area, wherein the standardized processing is preferred to facilitate segmentation and extraction. The method for standardizing the coverage area may include: calculating and obtaining the coordinates of the four points corresponding to Max{x}, Max{y}, Min{x}, and Min{y} of the coverage area, and then drawing a line parallel to the y axis through the coordinate points corresponding to Max{x} and Min{x}, and drawing a line parallel to the x axis through the coordinate points corresponding to Max{y} and Min{y}. The rectangular frame area formed by these four parallel lines is the target changing image area. The shooting module moves the center of the shooting area to the center position in the rectangular frame area, and scales the frame of the shooting area of ​​the shooting module to coincide with the boundary line of the rectangular frame area. The method for scaling the frame of the shooting area of ​​the shooting module to coincide with the boundary line of the rectangular frame area includes: calculating and obtaining the length of the longitudinal side Ly=Max{y}-Min{y}, the length of the transverse side Lx=Max{x}-Min{x}, and calculating the aspect ratio Is it greater than or equal to the device aspect ratio? If so, the camera module is scaled vertically. The device aspect ratio is the standard aspect ratio of the device for capturing images, which is not described in detail here. The camera module vertical scaling method includes: scaling the vertical coordinates of the camera module according to Scaling, horizontal and horizontal proportional scaling, where Ly 标 is the longitudinal length of the camera module, and Y is the longitudinal coordinate of the camera module. is smaller than the aspect ratio of the device, the camera module is horizontally scaled, and the camera module horizontally scaled method includes: scaling the horizontal coordinates of the camera module according to Scaling, vertical and horizontal proportional scaling, where Lx 标 is the horizontal side length of the shooting module, and X is the horizontal coordinate of the shooting module. The specific calculation example is not described here.

[0034] The non-changing image area segmentation unit is used to extract the coverage area coordinates of the non-changing image area, encode the non-changing image area, and bind the original non-changing image area picture frame, the coverage area coordinates of the non-changing image area and the code for storage. Then the image we finally store is the code (coverage area coordinates) corresponding to the non-changing image area, the original non-changing image area picture frame and the target changing image area. When image extraction is required, the code (coverage area coordinates) corresponding to the non-changing image area, the original non-changing image area picture frame and the target changing image area can be reversely translated and combined according to the coordinates to obtain a complete image. During the storage process, there is no complete picture frame or video information, which increases the privacy and security of big data information. At the same time, the code is used instead of the video for storage, and the actual memory occupancy is small, which greatly saves storage space.

[0035] Example 2

[0036] refer to Figure 2 In this embodiment, a shooting method is proposed, and the shooting method includes:

[0037] S1. Shoot the monitored area to obtain a shot image.

[0038] S2. Identify shooting features in the shot images, and collect the shot images with the shooting features as a target image set.

[0039] S3. Extract the target changed image area and the non-changed image area from the target image set.

[0040] S4, one of the shooting modules is used to move the target moving image area, zoom in and focus on the image to obtain the target zoomed image and store it to obtain a moving target zoom set.

[0041] S5. Binding and storing the target magnified image and the corresponding target image.

[0042] S6. Extract the coverage area coordinates of the non-changing image area and encode the non-changing image area.

[0043] S7. Bind and store the original non-changing image region picture frame, the non-changing image region coverage area coordinates and the code.

[0044] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field and related fields without creative work should fall within the scope of protection of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention are implemented according to the conventional means in the field unless otherwise specified and limited.

Claims

1. A photographing device, characterized in that: The shooting equipment includes: A shooting combination unit, comprising m groups of shooting modules, wherein m≥2, the shooting combination unit is used to shoot the monitoring area and obtain the shooting image, and the m groups of shooting modules form m degrees of shooting overlap for the same monitoring area; A shooting feature picking unit, which is used to identify shooting features in the shot images and collect the shot images with the shooting features as a target image set; A feature movement recognition unit is used to extract a target changing image area and a non-changing image area from a target image set, and use one of the shooting modules to move, zoom in, focus and shoot the target changing image area to obtain a target zoomed image and store it to obtain a moving target zoomed set, and bind the target zoomed image with the corresponding target image for storage; A non-changing image area segmentation unit is used to extract the coordinates of the coverage area of ​​the non-changing image area, encode the non-changing image area, and bind and store the original non-changing image area picture frame, the coordinates of the coverage area of ​​the non-changing image area and the code; The target change image region extraction method includes: sorting and separating each picture frame from the target image according to time, identifying k marking points from the picture frame, k≥3, calculating the distance between each marking point in the previous picture frame, and scaling the next picture frame according to the marking point distance in the previous picture frame, so that the k marking points in the picture frame overlap and completely coincide, then calculating the gray value of each pixel point domain, and comparing the gray value of the same position of the previous and next picture frames to obtain the gray value difference, and the area where the gray value difference is greater than a preset standard deviation is defined as the coverage area, and the coverage area is defined as the target change image region; The method for obtaining the marking point includes: constructing a plane coordinate system, implanting the picture frame into the plane coordinate system, processing the picture frame to obtain a gray value, performing dimensionality upscaling processing with the gray value as the z-axis, and calculating to obtain a plurality of slope points, connecting the slope points to obtain a texture line, and calculating the texture line. The points are taken as extreme points, k extreme points are selected from these extreme points, and the calculation Where i and j represent the numbers of the extreme points, and Max{l k } corresponding to the k extreme points, which are used as marking points.

2. A photographing device according to claim 1, characterized in that: The m is 2.

3. A photographing device according to claim 1, characterized in that: The number of shooting module groups m = [[1+sa]exp(c)]+1, where s is the maximum number of detection points in the monitoring area, a is the overlap occurrence factor, and its value is related to the frequency of occurrence of the monitored items. The higher the frequency of occurrence, the larger the value of the overlap occurrence factor a; c is the monitoring level parameter. When there is no abnormal situation for monitoring, one of the m groups of shooting modules is turned on for routine monitoring. When an abnormality occurs, the other groups of shooting modules are quickly turned on. The number of turned on modules needs to be determined according to the number of targets in the monitoring area.

4. A photographing device according to claim 1, characterized in that: The method for obtaining the non-changing image area includes obtaining the overlapping area of ​​the front and rear picture frames, and defining the overlapping area minus the part of the target changing image area as the non-changing image area.

5. A photographing device according to claim 1, characterized in that: The slope point is calculated as follows: and 6. A photographing device according to claim 1, characterized in that: The method for standardizing the coverage area includes: calculating and obtaining the coordinates of four points corresponding to Max{x}, Max{y}, Min{x}, and Min{y} of the coverage area, and then drawing a line parallel to the y axis through the coordinate point corresponding to Max{x} and the coordinate point corresponding to Min{x}, and drawing a line parallel to the x axis through the coordinate point corresponding to Max{y} and the coordinate point corresponding to Min{y}. The rectangular frame area formed by these four parallel lines is the target variable image area. The shooting module moves the center of the shooting area to the center position in the rectangular frame area, and scales the frame of the shooting area of ​​the shooting module to coincide with the boundary line of the rectangular frame area.

7. A photographing device according to claim 6, characterized in that: The method for scaling the frame of the shooting area of ​​the shooting module to coincide with the boundary line of the rectangular frame area includes: calculating the length of the vertical side Ly = Max{y}-Min{y}, the length of the horizontal side Lx = Max{x}-Min{x}, and calculating the aspect ratio Whether it is greater than or equal to the aspect ratio of the device, if so, the shooting module is scaled vertically, if not, the shooting module is scaled horizontally.

8. A shooting method, characterized in that: The shooting method comprises: S1. photographing the monitored area to obtain a photographic image; S2, identifying shooting features in the shot images, and collecting the shot images with the shooting features as a target image set; S3, extracting the target changed image area and the non-changed image area from the target image set; S4, one of the shooting modules moves the target moving image area, zooms in and focuses on the image to obtain the target zoomed image and stores the image to obtain a moving target zoomed set; S5, binding and storing the target magnified image and the corresponding target image; S6, extracting the coverage area coordinates of the non-changing image area, and encoding the non-changing image area; S7, binding and storing the original non-changing image area picture frame, the non-changing image area coverage area coordinates and the code; The target change image region extraction method includes: sorting and separating each picture frame from the target image according to time, identifying k marking points from the picture frame, k≥3, calculating the distance between each marking point in the previous picture frame, and scaling the next picture frame according to the marking point distance in the previous picture frame, so that the k marking points in the picture frame overlap and completely coincide, then calculating the gray value of each pixel point domain, and comparing the gray value of the same position of the previous and next picture frames to obtain the gray value difference, and the area where the gray value difference is greater than a preset standard deviation is defined as the coverage area, and the coverage area is defined as the target change image region; The method for obtaining the marking point includes: constructing a plane coordinate system, implanting the picture frame into the plane coordinate system, processing the picture frame to obtain a gray value, performing dimensionality upscaling processing with the gray value as the z-axis, and calculating to obtain a plurality of slope points, connecting the slope points to obtain a texture line, and calculating the texture line. The point is taken as the extreme point, k extreme points are selected from the extreme points, and the calculation Where i and j represent the numbers of the extreme points, and calculate Max{l k } corresponding to the k extreme points, which are used as marking points.

Images