Target tracking method, apparatus, and monitoring system

Abstract

The present application relates to a target tracking method, an apparatus, and a monitoring system. The method comprises: acquiring an image position of a target object in a monitoring image collected by a camera in an image processing device, matching a radar position of a moving object and the image position, and when the matching is successful, on the basis of the image position, controlling the camera to track the target object so as to obtain a tracking image of the target object; and on the basis of the image position of the target object in any frame of monitoring image of the tracking image, controlling the camera to perform zooming, so as to obtain a zooming image of the target object. In the method, a radar position of a moving object is matched with an image position of a target object obtained on the basis of visual detection, and when the matching is successful, it can be determined that the target object is a moving object focused by a user, so as to control a camera to track the target object, thereby effectively reducing the occurrences of false tracking, tracking switching, and missed tracking, and thus improving the accuracy of target tracking.

Description

Target tracking method, device and monitoring system

[0001] Related applications

[0002] The present application claims priority to the Chinese patent application No. 2024110756517, filed on August 6, 2024, and entitled "Target tracking method, device and monitoring system", the contents of which are hereby incorporated by reference in its entirety. TECHNICAL FIELD

[0003] The present application relates to the technical field of monitoring, in particular to a target tracking method, device and monitoring system. BACKGROUND

[0004] With the development of science and technology, monitoring systems have been increasingly applied in people's daily production, life and study.

[0005] In related technologies, a monitoring system can detect a target in a monitoring image collected by a camera, such as AI detection, determine the position of the target, and control the camera to follow the target to move, thereby realizing target tracking.

[0006] However, the accuracy of target tracking is poor due to various factors. SUMMARY

[0007] In a first aspect, the present application provides a target tracking method applied to a monitoring system, the monitoring system comprising a radar and an image processing device, and the method comprising:

[0008] obtaining an image position of a target object in a monitoring image collected by a camera in the image processing device;

[0009] matching the radar position of the moving object with the image position;

[0010] in the case of successful matching, controlling the camera to track the target object according to the image position, to obtain a tracking image of the target object;

[0011] controlling the camera to zoom according to the image position of the target object in any frame of the tracking image, to obtain a zoom image of the target object.

[0012] In one of the embodiments, matching the radar position of the moving object with the image position comprises:

[0013] converting the image position to a radar coordinate system to obtain an image converted position;

[0014] matching the image converted position with the radar position to obtain a matching result.

[0015] In one of the embodiments, the image conversion position comprises a center conversion position, the center conversion position being a coordinate position of a center point of the detection frame of the target object in the radar coordinate system; the image conversion position is matched with the radar position to obtain a matching result, comprising:

[0016] an interval distance between the center conversion position and the radar position is obtained;

[0017] in a case where the interval distance is less than a preset distance threshold, it is determined that the matching result is a matching success;

[0018] in a case where the interval distance is greater than or equal to the preset distance threshold, it is determined that the matching result is a matching failure.

[0019] In one of the embodiments, the image position is converted to the radar coordinate system to obtain the image conversion position, comprising:

[0020] the image position is converted to a camera coordinate system of the camera in the current attitude to obtain a standby position;

[0021] a conversion relationship between the camera coordinate system and the radar coordinate system is obtained;

[0022] the standby position is converted to the radar coordinate system according to the conversion relationship to obtain the image conversion position.

[0023] In one of the embodiments, the image position comprises a first diagonal position and a second diagonal position of the detection frame, and the image conversion position comprises a first conversion position converted from the first diagonal position and a second conversion position converted from the second diagonal position; the image conversion position is matched with the radar position to obtain a matching result, comprising:

[0024] a reference region in the radar coordinate system is determined according to the first conversion position and the second conversion position;

[0025] the matching result is determined according to the reference region and the radar position.

[0026] In one of the embodiments, the matching result is determined according to the reference region and the radar position, comprising:

[0027] in a case where the radar position is located within the reference region, it is determined that the matching result is a matching success;

[0028] in a case where the radar position is located outside the reference region, it is determined that the matching result is a matching failure.

[0029] In one of the embodiments, the camera is controlled to track the target object according to the image position to obtain a tracking image of the target object, comprising:

[0030] determine a reference distance and a reference direction of the image position relative to the center point position according to the image position and the center point position of the monitoring image;

[0031] control the camera to track the target object according to the reference distance and the reference direction, and obtain a tracking image of the target object.

[0032] In one of the embodiments, the control of the camera to track the target object according to the reference distance and the reference direction, and the obtaining of the tracking image of the target object, comprises:

[0033] determine a motion speed according to the reference distance;

[0034] control the camera to move along the reference direction at the motion speed, and obtain the tracking image of the target object.

[0035] In one of the embodiments, the control of the camera to zoom according to the image position of the target object in any frame of the monitoring image of the tracking image, and the obtaining of the zoomed image of the target object, comprises:

[0036] for any frame of the monitoring image in the tracking image, determine a zoomed region in the monitoring image according to the image position of the target object in the monitoring image;

[0037] control the camera to zoom according to the zoomed region, and obtain the zoomed image of the target object.

[0038] In one of the embodiments, the image position comprises a first diagonal position and a second diagonal position of a detection frame; the determination of the zoomed region in the monitoring image according to the image position of the target object in the monitoring image, comprises:

[0039] determine a target region in the monitoring image according to the first diagonal position and the second diagonal position;

[0040] determine a region comprising the target region in the monitoring image as the zoomed region.

[0041] In one of the embodiments, the control of the camera to zoom according to the zoomed region, and the obtaining of the zoomed image of the target object, comprises:

[0042] determine a zooming multiple according to the monitoring image and the zoomed region;

[0043] control the camera to zoom the zoomed region by the zooming multiple, and obtain the zoomed image of the target object.

[0044] In a second aspect, the application further provides a target tracking device, comprising:

[0045] a position obtaining module, configured to obtain an image position of a target object in a monitoring image collected by a camera;

[0046] The position matching module is configured to match the radar position of the moving object with the image position.

[0047] The target tracking module is configured to control the camera to track the target object according to the image position in a case of successful matching, to obtain a tracking image of the target object.

[0048] The zoom control module is configured to control the camera to zoom according to any frame of the tracking image, to obtain a zoom image of the target object.

[0049] In one of the embodiments, the position matching module comprises:

[0050] The coordinate conversion submodule is configured to convert the image position to the radar coordinate system to obtain an image conversion position.

[0051] The position matching submodule is configured to match the image conversion position with the radar position to obtain a matching result.

[0052] In one of the embodiments, the image conversion position comprises a center conversion position, which is a coordinate position of a center point of the detection frame of the target object in the radar coordinate system; and the position matching submodule comprises:

[0053] The distance acquisition unit is configured to acquire an interval distance between the center conversion position and the radar position.

[0054] The matching success unit is configured to determine that the matching result is successful matching in a case that the interval distance is less than a preset distance threshold.

[0055] The matching failure unit is configured to determine that the matching result is failed matching in a case that the interval distance is greater than or equal to the preset distance threshold.

[0056] In one of the embodiments, the coordinate conversion submodule comprises:

[0057] The attitude conversion unit is configured to convert the image position to a camera coordinate system of the camera in a current attitude to obtain a backup position.

[0058] The relationship acquisition unit is configured to acquire a conversion relationship between the camera coordinate system and the radar coordinate system.

[0059] The position conversion unit is configured to convert the backup position to the radar coordinate system according to the conversion relationship to obtain the image conversion position.

[0060] In one of the embodiments, the image position comprises a first diagonal position and a second diagonal position of the detection frame, and the image conversion position comprises a first conversion position converted from the first diagonal position and a second conversion position converted from the second diagonal position; and the position matching submodule comprises:

[0061] reference region unit, configured to determine a reference region in a radar coordinate system according to the first conversion position and the second conversion position;

[0062] a matching result unit, configured to determine a matching result according to the reference region and the radar position.

[0063] In one of the embodiments, the matching result unit comprises:

[0064] a success sub-unit, configured to determine the matching result as a matching success in a case that the radar position is located within the reference region;

[0065] a failure sub-unit, configured to determine the matching result as a matching failure in a case that the radar position is located outside the reference region.

[0066] In one of the embodiments, the target tracking module comprises:

[0067] a reference determination sub-module, configured to determine a reference distance and a reference direction of the image position relative to a center point position of the monitoring image according to the image position and the center point position;

[0068] a tracking control sub-module, configured to control the camera to track the target object according to the reference distance and the reference direction, and obtain a tracking image of the target object.

[0069] In one of the embodiments, the reference determination sub-module comprises:

[0070] a speed determination unit, configured to determine a motion speed according to the reference distance;

[0071] a motion control unit, configured to control the camera to move along the reference direction at the motion speed, and obtain the tracking image of the target object.

[0072] In one of the embodiments, the zoom control module comprises:

[0073] a zoom determination sub-module, configured to determine a zoom region in the monitoring image according to the image position of the target object in the monitoring image for any frame of the tracking image;

[0074] a region zoom sub-module, configured to control the camera to zoom according to the zoom region, and obtain a zoom image of the target object.

[0075] In one of the embodiments, the image position comprises a first diagonal position and a second diagonal position of a detection frame; and the zoom determination module comprises:

[0076] a target determination unit, configured to determine a target region in the monitoring image according to the first diagonal position and the second diagonal position;

[0077] A region determining unit is configured to determine a region including the target region as the zoom region in the monitoring image.

[0078] In one embodiment, the region zoom sub-module includes:

[0079] A multiple determining unit is configured to determine a zoom multiple according to the monitoring image and the zoom region;

[0080] A zoom control unit is configured to control the camera to zoom in the zoom region by the zoom multiple and acquire a zoom image of the target object.

[0081] In a third aspect, the present application provides a monitoring system including a radar and an image processing device, the image processing device including a camera, a posture sensor, a memory and a processor, the memory storing a computer program, and the processor implementing the steps of any of the above methods when executing the computer program.

[0082] The details of one or more embodiments of the application are set forth in the accompanying drawings and the description below. Other features, objects, and advantages of the application will be apparent from the description and drawings, and from the claims. BRIEF DESCRIPTION OF DRAWINGS

[0083] In order to more clearly illustrate the technical solutions in the embodiments of the present application or in the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced below. Obviously, the drawings in the following description are only embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort on the basis of the disclosed drawings.

[0084] FIG. 1 is a structural schematic diagram of a monitoring system in one embodiment;

[0085] FIG. 2 is a flow schematic diagram of a target tracking method in one embodiment;

[0086] FIG. 3 is a flow schematic diagram of obtaining a matching result in one embodiment;

[0087] FIG. 4 is a flow schematic diagram of obtaining an image conversion position in one embodiment;

[0088] FIG. 5 is a flow schematic diagram of determining a matching result in one embodiment;

[0089] FIG. 6 is a flow schematic diagram of determining a matching result in another embodiment;

[0090] FIG. 7 is a flow schematic diagram of obtaining a tracking image in one embodiment;

[0091] FIG. 8 is a flow schematic diagram of obtaining a tracking image in another embodiment;

[0092] FIG. 9 is a flowchart of a process of obtaining a zoomed image in an embodiment;

[0093] FIG. 10 is a flowchart of a process of obtaining a zoomed image in another embodiment;

[0094] FIG. 11 is a flowchart of a process of determining a zoomed region in an embodiment;

[0095] FIG. 12 is a flowchart of a process of target tracking in another embodiment;

[0096] FIG. 13 is a block diagram of a structure of a target tracking device in an embodiment. DETAILED DESCRIPTION

[0097] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and should not be used to limit the present application.

[0098] The target tracking method provided by the embodiments of the present application can be applied to a monitoring system as shown in FIG. 1. The radar 102 and the image processing device 104 communicate with each other. The image processing device 104 includes a camera and a posture sensor connected to the camera. The image processing device 104 further includes a microcontroller and a main chip. The microcontroller is configured to obtain the posture of the camera collected by the posture sensor and feed back to the main chip. The main chip is configured to receive the radar position of the moving object sent by the radar 102, and perform visual detection on the monitoring image collected by the camera to obtain the image position of the target object in the monitoring image. The main chip can implement the above-mentioned target tracking method based on the radar position of the moving object detected by the radar and the image position of the target object obtained by visual detection.

[0099] The above-mentioned target tracking method can be applied to various scenes in the security field, such as indoor, street, forest, farm and the like.

[0100] Those skilled in the art can understand that the structure shown in FIG. 1 is only a block diagram of part of the structure related to the present application scheme, and does not constitute a limitation on the monitoring system to which the present application scheme is applied. The specific monitoring system can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0101] In an embodiment, as shown in FIG. 2, a target tracking method is provided. Taking the image processing device in FIG. 1 as an example, the method is applied to the main chip in the image processing device, and includes the following steps:

[0102] S210, obtaining the image position of the target object in the monitoring image collected by the camera in the image processing device.

[0103] The image position of the target object in the monitoring image is an example of a visual position of the target object. The image position can be a two-dimensional coordinate of the target object in the monitoring image.

[0104] Optionally, the image processing device is equipped with a camera. The camera can capture a monitoring image or a monitoring video within the monitoring coverage. For the captured monitoring image or at least one frame of the monitoring video, the image processing device can detect the target object in the monitoring image captured by the camera to determine the image position of the target object in the monitoring image. For example, the image processing device can use the diagonal coordinates of the detected bounding box as the image position of the target object, and can also obtain the center point coordinates of the bounding box as the image position of the target object.

[0105] S220, matching the radar position of the moving object with the image position.

[0106] The radar position of the moving object is a spatial position of the moving object determined by the radar based on radar data. For example, the radar position can be a three-dimensional coordinate of the moving object in a radar coordinate system.

[0107] Optionally, the image processing device communicates with the radar. The image processing device can directly receive the radar position of the moving object fed back by the radar, or can receive radar data collected by the radar and determine the radar position of the moving object based on the radar data, and then match the radar position of the moving object with the detected image position of the target object to obtain a matching result.

[0108] For example, the radar data is point cloud data. The image processing device or the radar can cluster the collected point cloud data to divide points with similar distances in the point cloud data into a category, and obtain the center point coordinates of the point cloud data in the category as the radar coordinates of the corresponding moving object.

[0109] It should be noted that the image position of the target object is determined based on single visual detection, and tracking is performed. Mis-detection is likely to occur, and a static target not focused on by the user is tracked. In addition, target switching is likely to occur in the tracking process due to overlapping of the positions of the target objects, which causes tracking errors. The radar position of the moving object is determined based on single radar detection, and tracking is performed. Only the tracking of the moving object can be achieved, and the target object actually focused on by the user cannot be accurately tracked. Therefore, the radar detection of the moving object and the visual detection of the target object focused on by the user can be fused, the radar position of the moving object obtained by the radar detection is matched with the image position of the target object focused on by the user obtained by the visual detection, and the tracking of the target object is performed according to the matching result.

[0110] Wherein, the radar position matches the image position successfully, indicating that the target object detected by the vision is the mobile object focused by the user, and tracking is needed; otherwise, the radar position fails to match the image position, indicating that the target object detected by the vision is not the mobile object focused by the user, and accordingly tracking is not needed.

[0111] S230, in the case of successful matching, the camera is controlled to track the target object according to the image position, and a tracking image of the target object is obtained.

[0112] Optionally, the image processing device, in the case of successful matching of the radar position of the mobile object and the image position of the target object in the monitoring image, controls the camera to move according to the image position of the target object in the monitoring image to track the target object, and a tracking image of the target object is collected.

[0113] Illustratively, the image processing device can obtain the image positions of the target object in two adjacent monitoring images at the collection time, respectively, and determine the moving direction of the target object according to the obtained two image positions to control the camera to move in the moving direction, thereby realizing tracking of the target object. Specifically, the direction from the image position of the target object in the former monitoring image to the image position of the target object in the latter monitoring image can be obtained as the moving direction of the target object.

[0114] In the case of failure of matching of the radar position of the mobile object and the image position of the target object in the monitoring image, the image processing device does not trigger tracking of the target object. In the case of detection of multiple target objects in the monitoring image, if the radar position of the mobile object does not match the image position of one of the target objects, the image processing device can replace the target object to match the radar position of the mobile object with the image positions of other target objects until successful matching or all target objects are matched.

[0115] S240, the camera is controlled to zoom according to the image position of the target object in any frame of the tracking image, and a zoomed image of the target object is obtained.

[0116] Wherein, the control of the camera to zoom can be zooming out of the region in the monitoring image where the target object is located, or zooming in of the region in the monitoring image where the target object is located.

[0117] In the case of zooming in on the region where the target object is located, the obtained zoomed image of the target object includes a partial picture of the target object and filters out environmental interference, so that the user can focus on the motion of the partial target object, thereby improving the concentration of the target tracking on the target object. In the case of zooming out on the region where the target object is located, the obtained zoomed image of the target object includes an overall picture of the motion of the target object in the environment, so that more environmental pictures are provided, and the user can view the overall environmental changes in the motion process of the target object, thereby improving the comprehensiveness of the information provided by the target tracking.

[0118] Optionally, for any frame of the monitoring image in the tracking image, the image processing device can obtain the image position of the target object in the monitoring image, control the camera to zoom in or out on the region where the target object is located in the monitoring image with the image position of the target object as the visual center, and continuously capture the zoomed image of the target object processed by the camera.

[0119] In the embodiments of the present application, the image position of the target object in the monitoring image captured by the camera in the image processing device is obtained, and the radar position of the moving object is matched with the image position, so that in the case of successful matching, the camera is controlled to track the target object according to the image position, the tracking image of the target object is obtained, and the camera is controlled to zoom in or out according to the image position of the target object in any frame of the monitoring image in the tracking image, so as to obtain the zoomed image of the target object. In the above method, the radar position of the moving object is matched with the image position of the target object obtained based on visual detection, and in the case of successful matching, the target object can be determined as the moving object concerned by the user, so that the camera is controlled to track the target object, which can effectively reduce the occurrence of false tracking, tracking switching and missed tracking, thereby improving the accuracy of the target tracking.

[0120] The radar position and the image position need to be converted to the same coordinate system. Based on this, in one of the embodiments, as shown in FIG. 3, the above S220, matching the radar position of the moving object with the image position, comprises:

[0121] S310, converting the image position to the radar coordinate system to obtain an image converted position.

[0122] The radar coordinate system is a three-dimensional coordinate system used by the radar. For example, the radar coordinate system is a spherical coordinate system.

[0123] Optionally, the image position is a two-dimensional coordinate, and the image processing device can read a pre-calibrated conversion relationship from two-dimensional to three-dimensional, so as to convert the image position to the radar coordinate system by using the conversion relationship, obtain the three-dimensional coordinates of the image position in the radar coordinate system as the image converted position.

[0124] S320, match the image conversion position and the radar position to obtain a matching result.

[0125] Optionally, after obtaining the image conversion position and the radar position in the radar coordinate system, the image processing device can directly match the image conversion position and the radar position to obtain a matching result.

[0126] Illustratively, the image position of the target object is a center point coordinate of a detection frame, and a corresponding image conversion position of the center point coordinate is a center conversion position. The image processing device can obtain an interval distance between the center conversion position and the radar position, and determine a matching result according to the interval distance. In a case where the interval distance is less than a preset distance threshold, it is determined that the matching result is a matching success; otherwise, in a case where the interval distance is greater than or equal to the preset distance threshold, it is determined that the matching result is a matching failure.

[0127] In the embodiments of the present application, the image position is converted to the radar coordinate system to obtain an image conversion position, so as to match the image conversion position and the radar position to obtain a matching result. In the above method, the coordinate systems of the image position and the radar position are unified, and the accuracy and reliability of the matching result are correspondingly improved.

[0128] In order to improve the accuracy of the image conversion position, in one of the embodiments, as shown in FIG. 4, S310, converting the image position to the radar coordinate system to obtain an image conversion position, includes:

[0129] S410, converting the image position to a camera coordinate system of the camera in a current attitude to obtain a backup position.

[0130] The camera coordinate system is a three-dimensional coordinate system constructed by the camera in the current attitude. Illustratively, the camera coordinate system is a spherical coordinate system.

[0131] In the process of target tracking, the attitude of the camera changes, and the camera coordinate system also changes. The camera coordinate system can be determined by the attitude of the camera, the intrinsic parameters and the extrinsic parameters. The attitude of the camera can be obtained by an attitude sensor connected to the camera. Illustratively, the attitude sensor can be an acceleration sensor (G-sensor).

[0132] Optionally, the image processing device can construct a three-dimensional coordinate system of the camera in the current attitude as the camera coordinate system, and convert a two-dimensional coordinate position of the target object in the monitoring image, i.e., an image position, to the camera coordinate system according to the intrinsic parameters, the extrinsic parameters and the current attitude of the camera to obtain a three-dimensional coordinate of the image position in the camera coordinate system, and take it as a backup position. The camera coordinate system can be represented as:

[0133] C(x, y, z) = F(f x, f y, L, Rd, Rr, img x, img y)

[0134] C(x, y, z) represents the position coordinates of any point in the camera coordinate system; f x, f y represent the intrinsic parameters of the camera; L represents the extrinsic parameters of the camera, specifically the installation height of the camera; Rd and Rr are the pitch angle and the rotation angle of the camera respectively, representing the attitude of the camera; img x is the x coordinate of any point in the camera coordinate system in the two-dimensional picture, and img y is the y coordinate of any point in the camera coordinate system in the two-dimensional picture.

[0135] Exemplarily, the image position of the target object is the diagonal coordinates (pl, pr) of the detection frame, pl represents the upper left corner coordinates (img pl x, img pl y) of the detection frame in the monitoring image, and pr represents the lower right corner coordinates (img pr x, img pr y) of the detection frame in the monitoring image. Conversion to the camera coordinate system under the current attitude of the camera respectively obtains the positions of the upper left corner and the lower right corner of the corresponding detection frame in the spherical coordinate system: P1(x, y, z) = F(f x, f y, L, Rd, Rr, img pl x, img pl y), P2(x, y, z) = F(f x, f y, L, Rd, Rr, img pr x, img pr y).

[0136] S420, obtain the conversion relationship between the camera coordinate system and the radar coordinate system.

[0137] Optionally, after the image processing device determines the camera coordinate system under the current attitude of the camera, the radar coordinate system can be read from the radar end to perform coordinate system registration analysis, and the conversion relationship between the camera coordinate system and the radar coordinate system is obtained.

[0138] S430, convert the standby position to the radar coordinate system according to the conversion relationship to obtain the image conversion position.

[0139] Optionally, after obtaining the conversion relationship between the camera coordinate system and the radar coordinate system, the image processing device can convert the standby position in the camera coordinate system obtained in the foregoing to the radar coordinate system according to the conversion relationship, to obtain the three-dimensional coordinates of the standby position in the radar coordinate system as the image conversion position.

[0140] Continuing the above example, after converting the standby positions P1(x, y, z) and P2(x, y, z) to the radar coordinate system, the image conversion positions Pr1(x, y, z) = K*P1(x, y, z), Pr2(x, y, z) = K*P2(x, y, z) are obtained, and K represents the conversion relationship between the camera coordinate system and the radar coordinate system.

[0141] In the embodiments of the present application, the image position is converted to the camera coordinate system of the camera in the current posture to obtain a backup position, the conversion relationship between the camera coordinate system and the radar coordinate system is obtained, and the backup position is converted to the radar coordinate system according to the conversion relationship to obtain the image conversion position. In the above method, the image position is converted to the camera coordinate system of the camera in the current posture, the actual motion of the camera is taken into account, the camera coordinate system is dynamically updated, the influence of the motion of the camera is reduced, and the accuracy of the image conversion position obtained by conversion is improved.

[0142] In a case where the image position includes a first diagonal position and a second diagonal position of the detection frame, and the image conversion position includes a first conversion position converted from the first diagonal position and a second conversion position converted from the second diagonal position, in one of the embodiments, as shown in FIG. 5, the S320, matching the image conversion position with the radar position to obtain a matching result, includes:

[0143] S510, determining a reference region in the radar coordinate system according to the first conversion position and the second conversion position.

[0144] The reference region is a spatial region formed by the detection frame in the radar coordinate system.

[0145] Optionally, the image processing device can connect the line between the first conversion position and the second conversion position as the diagonal line of the spatial region to form the spatial region in the radar coordinate system as the reference region. For example, the reference region can be a cuboid or a square spatial region, or a spherical spatial region.

[0146] S520, determining a matching result according to the reference region and the radar position.

[0147] Optionally, after obtaining the reference region, the image processing device can perform position matching between the radar position and the reference region to determine the matching result.

[0148] In an optional embodiment, as shown in FIG. 6, the S520, determining a matching result according to the reference region and the radar position, includes:

[0149] S610, in a case where the radar position is located within the reference region, determining that the matching result is matching success.

[0150] S620, in a case where the radar position is located outside the reference region, determining that the matching result is matching failure.

[0151] Wherein, the radar position is located within the reference region, indicating that the spatial position of the moving object detected by the radar coincides with the target object detected by the vision, and accordingly indicating that the radar position and the image position match successfully, also indicating that the moving object detected by the radar and the target object detected by the vision are the same object; on the contrary, the radar position is located outside the reference region, indicating that the spatial position of the moving object detected by the radar does not coincide with the target object detected by the vision, and accordingly indicating that the radar position and the image position match unsuccessfully, also indicating that the moving object detected by the radar and the target object detected by the vision are different objects.

[0152] In the embodiment of the application, the image position includes a first diagonal position and a second diagonal position of the detection frame, the image conversion position includes a first conversion position converted from the first diagonal position and a second conversion position converted from the second diagonal position, the reference region can be determined according to the first conversion position and the second conversion position, and then the matching result can be determined according to the reference region and the radar position. Specifically, in the case that the radar position is located within the reference region, it is determined that the matching result is matching successfully; in the case that the radar position is located outside the reference region, it is determined that the matching result is matching unsuccessfully. In the above method, the spatial region formed by the detection frame in the radar coordinate system, that is, the reference region, can accurately and comprehensively represent the spatial position of the target object, and the matching based on the reference region and the radar position to determine the matching result can correspondingly improve the accuracy of the matching result.

[0153] To realize accurate tracking of the target object, in one of the embodiments, as shown in FIG. 7, the image position is used to control the camera to track the target object in S230, and the tracking image of the target object is obtained, including:

[0154] S710, determining a reference distance and a reference direction of the image position relative to the center point position according to the image position and the center point position of the monitoring image.

[0155] Wherein, the reference distance of the image position relative to the center point position is used to represent the distance between the target object and the center point in the monitoring image; and the reference direction of the image position relative to the center point position is the direction of the target object relative to the center point in the monitoring image.

[0156] Optionally, after the image processing device detects the image position of the target object in the monitoring image based on the vision, the reference distance of the image position relative to the center point position can be obtained according to the image position and the center point position in the monitoring image, and the direction of the center point position pointing to the image position is obtained as the reference direction of the image position relative to the center point position.

[0157] Exemplarily, in a case where the image position comprises a first diagonal position and a second diagonal position of the detection frame, the image processing device can acquire a midpoint position of a line connecting the first diagonal position and the second diagonal position, and acquire a distance between the midpoint position and the center point position as the reference distance.

[0158] S720, control the camera to track the target object according to the reference distance and the reference direction, to obtain a tracking image of the target object.

[0159] Optionally, after obtaining the reference distance and the reference direction of the image position relative to the center point position, the image processing device can adjust the reference distance according to a preset ratio to obtain a target distance, and control the camera to move towards the reference direction by the target distance, so as to track the target object by the moving camera and obtain the tracking image of the target object. Wherein, controlling the camera to move towards the reference direction by the target distance can make the target object in the middle region of the monitoring image.

[0160] Exemplarily, the image processing device can reduce the reference distance according to the preset ratio to obtain the target distance.

[0161] In the embodiments of the present application, the reference distance and the reference direction of the image position relative to the center point position are determined according to the image position and the center point position of the monitoring image, so as to control the camera to track the target object according to the reference distance and the reference direction, to obtain the tracking image of the target object. In the above method, the reference distance and the reference direction can accurately reflect the position of the target object in the monitoring image, and controlling the camera to move to track the target object based on the reference distance and the reference direction can improve the tracking accuracy of the target object and reduce the target loss probability.

[0162] The reference distance determines the movement speed of the camera. Therefore, in one of the embodiments, as shown in FIG. 8, the above S720, controlling the camera to track the target object according to the reference distance and the reference direction to obtain the tracking image of the target object, comprises:

[0163] S810, determining the movement speed according to the reference distance.

[0164] Wherein, the movement speed is the movement speed of the camera. The reference distance is negatively related to the movement speed. The greater the reference distance is, the smaller the movement speed is; the smaller the reference distance is, the greater the movement speed is.

[0165] It should be noted that in the target tracking process, the size of the reference distance between the image position of the target object in the monitoring image and the center point position reflects the speed of movement of the target object. If the reference distance is large, it indicates that the target object moves faster, and the camera needs to move at a faster speed to keep up with the target object. Conversely, if the reference distance is small, it indicates that the target object moves slower, and the camera can keep up with the target object by moving at a slower speed.

[0166] Optionally, after obtaining the reference distance, the image processing device can determine the speed corresponding to the reference distance according to a preset distance-speed correspondence relationship, as the movement speed of the camera.

[0167] S820, control the camera to move along the reference direction at the movement speed, and collect the tracking image of the target object.

[0168] Optionally, after obtaining the movement speed of the camera, the image processing device can control the camera to move along the reference direction at the movement speed, so as to track the target object by the moving camera, and collect the tracking image of the target object.

[0169] In the embodiments of the present application, the movement speed is determined according to the reference distance, and the camera is controlled to move along the reference direction at the movement speed, and the tracking image of the target object is collected. In the above method, the movement speed of the camera is determined based on the reference distance, which can adapt to the actual movement of the target object, flexibly adjust the movement speed of the camera to track the target object, and thus improve the flexibility and accuracy of target tracking.

[0170] To meet the zooming requirement, in one of the embodiments, as shown in FIG. 9, the above S240, controlling the camera to zoom according to the image position of the target object in any frame of the monitoring image of the tracking image, to obtain the zoomed image of the target object, includes:

[0171] S910, for any frame of the monitoring image in the tracking image, determining a zoomed region in the monitoring image according to the image position of the target object in the monitoring image.

[0172] The zoomed region in the monitoring image is a partial region in the monitoring image including the target object, rather than the entire monitoring image.

[0173] Optionally, for any frame of the monitoring image in the tracking image, the image processing device can obtain the image position of the target object in the monitoring image, and determine a partial region including the target object in the monitoring image as the zoomed region according to the image position. Illustratively, the image position is the position of the target object in the monitoring image, and the image processing device can determine a partial region including the image position in the monitoring image as the zoomed region.

[0174] S920, control the camera to zoom according to the zoom region, to obtain the zoom image of the target object.

[0175] Optionally, after obtaining the zoom region in the monitoring region, the image processing device can control the camera to zoom the zoom region in the monitoring image to obtain the zoom image of the target object.

[0176] It should be noted that the process of the above zoom processing is also performed in the case that the radar position of the moving object detected by the radar and the image position of the target object detected by the vision are successfully matched, and the camera coordinate system is dynamically updated in the matching process, which correspondingly improves the reliability of the matching result. In the case of successful matching, zooming is performed based on the image position of the target object, which reduces the influence of camera motion on zooming, and thus improves the zooming accuracy.

[0177] In the target tracking process, it is usually necessary to enlarge the region where the target object is located in the monitoring image. Based on this, in an optional embodiment, as shown in FIG. 10, the above S920, control the camera to zoom according to the zoom region, to obtain the zoom image of the target object, includes:

[0178] S1010, determine the zooming multiple according to the monitoring image and the zoom region.

[0179] Optionally, the image processing device can determine the zooming multiple according to the specification of the monitoring image and the specification of the zoom region. Exemplarily, the specification can be represented by resolution. The image processing device can obtain the ratio of the resolution of the monitoring image to the resolution of the zoom region, and take it as the zooming multiple.

[0180] S1020, control the camera to enlarge the zoom region by the zooming multiple, and collect the zoom image of the target object.

[0181] Optionally, after the image processing device determines the zooming multiple and the zoom region in the monitoring region, the camera can be controlled to enlarge the zoom region by the zooming multiple, so as to collect the zoom image of the target object through the zoomed camera.

[0182] In the embodiments of the present application, for any frame of monitoring image in the tracking image, the zoom region in the monitoring image is determined according to the image position of the target object in the monitoring image, so as to control the camera to zoom according to the zoom region, to obtain the zoom image of the target object. Specifically, the zooming multiple can be determined according to the monitoring image and the zoom region, so as to control the camera to enlarge the zoom region by the zooming multiple, and collect the zoom image of the target object. In the above method, automatic zooming based on the target object is realized, and the zooming efficiency and accuracy are improved.

[0183] The image position includes a first diagonal position and a second diagonal position of the detection frame. In one embodiment, the zoom region in the monitoring image is determined according to the image position of the target object in the monitoring image in S910, as shown in FIG. 11, including:

[0184] S1110, determining a target region in the monitoring image according to the first diagonal position and the second diagonal position.

[0185] The target region in the monitoring image is a region including the target object.

[0186] Optionally, the image processing device can restore the detection frame of the rectangle in the monitoring image according to the first diagonal position and the second diagonal position, and take the region in the monitoring image where the detection frame is located as the target region.

[0187] S1120, determining a region including the target region in the monitoring image as the zoom region.

[0188] The part region in the monitoring image including the target region can be directly the target region, or an image region in the monitoring image including and larger than the target region. The zoom region is a part image region in the monitoring image.

[0189] Optionally, after obtaining the target region in the monitoring region, the image processing device can directly obtain the target region as the zoom region, or can obtain a region including and larger than the target region in the monitoring image as the zoom region.

[0190] In the embodiments of the present application, the image position includes the first diagonal position and the second diagonal position of the detection frame, the target region in the monitoring image can be determined according to the first diagonal position and the second diagonal position, and a region including the target region in the monitoring image is determined as the zoom region. In the above method, the target region is determined based on the first diagonal position and the second diagonal position of the detection frame, the target region can cover the entire target object, and the determined zoom region correspondingly covers the entire target object, so that the integrity of the zoomed image of the target object obtained after zooming is improved.

[0191] In order to facilitate the understanding of those skilled in the art, the target tracking method provided by the present application is described in detail below, as shown in FIG. 12, which can include:

[0192] S1201, obtaining an image position of a target object in a monitoring image captured by a camera; the image position includes a first diagonal position and a second diagonal position of a detection frame;

[0193] S1202, converting the image position to a camera coordinate system of the camera in a current posture to obtain a standby position;

[0194] S1203, convert the standby position to the radar coordinate system according to the conversion relationship between the camera coordinate system and the radar coordinate system, to obtain an image conversion position; the image conversion position includes a first conversion position converted from the first diagonal position and a second conversion position converted from the second diagonal position;

[0195] S1204, determine a reference region in the radar coordinate system according to the first conversion position and the second conversion position;

[0196] S1205, in a case where the radar position of the moving object is located within the reference region, determine that the radar position of the moving object matches the image position successfully; in a case where the radar position of the moving object is located outside the reference region, determine that the radar position of the moving object fails to match the image position;

[0197] S1206, in a case of successful matching, determine a reference distance and a reference direction of the image position relative to the center point position of the monitoring image according to the image position and the center point position of the monitoring image;

[0198] S1207, determine the motion speed according to the reference distance;

[0199] S1208, control the camera to move along the reference direction at the motion speed, and acquire a tracking image of the target object;

[0200] S1209, for any frame of the monitoring image in the tracking image, determine a target region in the monitoring image according to the first diagonal position and the second diagonal position;

[0201] S1210, determine a region including the target region in the monitoring image as a zoom region;

[0202] S1211, determine a zoom factor according to the monitoring image and the zoom region;

[0203] S1212, control the camera to zoom in the zoom region by the zoom factor, and acquire a zoom image of the target object.

[0204] It should be noted that the description in S1201-S1212 above can refer to the description of the related embodiments above, and the effects are similar, and the present embodiment will not be repeated here.

[0205] It should be understood that although the steps in the flowcharts involved in the embodiments described above are shown in sequence according to the arrows, the steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, the execution of the steps is not strictly limited in sequence, and the steps can be executed in other orders. Moreover, at least some of the steps in the flowcharts involved in the embodiments described above can include multiple steps or multiple stages, which are not necessarily executed at the same time but can be executed at different times, and the execution order of the steps or stages is not necessarily sequential but can be alternately or alternately executed with at least part of other steps or steps or stages in other steps.

[0206] Based on the same inventive concept, the embodiments of the present application also provide a projection correction device for implementing the projection correction method described above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more projection correction device embodiments provided below can refer to the limitations of the projection correction method described above, which will not be repeated here.

[0207] In one embodiment, as shown in FIG. 13, a target tracking device is provided, comprising: a position acquisition module 1301, a position matching module 1302, a target tracking module 1303, and a zoom control module 1304, wherein:

[0208] The position acquisition module 1301 is configured to acquire an image position of a target object in a monitoring image captured by a camera in an image processing device;

[0209] The position matching module 1302 is configured to match the radar position of the moving object with the image position;

[0210] The target tracking module 1303 is configured to, in the case of successful matching, control the camera to track the target object according to the image position, and obtain a tracking image of the target object;

[0211] The zoom control module 1304 is configured to control the camera to zoom according to the image position of the target object in any frame of the tracking image, and obtain a zoom image of the target object.

[0212] In one embodiment, the position matching module 1302 comprises:

[0213] A coordinate conversion submodule is configured to convert the image position to a radar coordinate system to obtain an image conversion position;

[0214] A position matching submodule is configured to match the image conversion position with the radar position to obtain a matching result.

[0215] In one of the embodiments, the image conversion position comprises a center conversion position, the center conversion position being a coordinate position of a center point of the detection box of the target object in the radar coordinate system; the position matching sub-module comprises:

[0216] a distance acquisition unit configured to acquire an interval distance between the center conversion position and the radar position;

[0217] a matching success unit configured to determine that the matching result is matching success in a case that the interval distance is less than a preset distance threshold;

[0218] a matching failure unit configured to determine that the matching result is matching failure in a case that the interval distance is greater than or equal to the preset distance threshold.

[0219] In one of the embodiments, the coordinate conversion sub-module comprises:

[0220] a posture conversion unit configured to convert the image position to a camera coordinate system of the camera in the current posture to obtain a backup position;

[0221] a relationship acquisition unit configured to acquire a conversion relationship between the camera coordinate system and the radar coordinate system;

[0222] a position conversion unit configured to convert the backup position to the radar coordinate system according to the conversion relationship to obtain the image conversion position.

[0223] In one of the embodiments, the image position comprises a first diagonal position and a second diagonal position of the detection box, and the image conversion position comprises a first conversion position converted from the first diagonal position and a second conversion position converted from the second diagonal position; the position matching sub-module comprises:

[0224] a reference region unit configured to determine a reference region in the radar coordinate system according to the first conversion position and the second conversion position;

[0225] a matching result unit configured to determine the matching result according to the reference region and the radar position.

[0226] In one of the embodiments, the matching result unit comprises:

[0227] a success sub-unit configured to determine that the matching result is matching success in a case that the radar position is located within the reference region;

[0228] a failure sub-unit configured to determine that the matching result is matching failure in a case that the radar position is located outside the reference region.

[0229] In one of the embodiments, the target tracking module 1303 comprises:

[0230] The reference determining sub-module is configured to determine a reference distance and a reference direction of the image position relative to the center point position according to the image position and the center point position of the monitoring image.

[0231] The tracking control sub-module is configured to control the camera to track the target object according to the reference distance and the reference direction, and obtain a tracking image of the target object.

[0232] In one of the embodiments, the reference determining sub-module comprises:

[0233] The speed determining unit is configured to determine the motion speed according to the reference distance.

[0234] The motion control unit is configured to control the camera to move along the reference direction at the motion speed, and obtain the tracking image of the target object.

[0235] In one of the embodiments, the zoom control module 1304 comprises:

[0236] The zoom determining sub-module is configured to determine a zoom region in the monitoring image according to the image position of the target object in the monitoring image for any frame of the tracking image.

[0237] The region zoom sub-module is configured to control the camera to zoom according to the zoom region, and obtain a zoom image of the target object.

[0238] In one of the embodiments, the image position comprises a first diagonal position and a second diagonal position of the detection frame; and the zoom determining module comprises:

[0239] The target determining unit is configured to determine a target region in the monitoring image according to the first diagonal position and the second diagonal position.

[0240] The region determining unit is configured to determine a region comprising the target region in the monitoring image as the zoom region.

[0241] In one of the embodiments, the region zoom sub-module comprises:

[0242] The multiple determining unit is configured to determine a zoom multiple according to the monitoring image and the zoom region.

[0243] The zoom control unit is configured to control the camera to zoom the zoom region by the zoom multiple, and obtain the zoom image of the target object.

[0244] The above modules in the target tracking device can be realized by software, hardware and combinations thereof in whole or in part. The above modules can be embedded in or independent of the processor in the computer device in hardware form, or stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to the above modules.

[0245] In one embodiment, a monitoring system is provided, comprising a radar and an image processing device, the image processing device comprising a camera, a posture sensor, a memory and a processor, the memory storing a computer program, and the processor implementing the following steps when executing the computer program:

[0246] obtaining an image position of a target object in a monitoring image captured by the camera in the image processing device; matching the radar position of the moving object with the image position; in the case of successful matching, controlling the camera to track the target object according to the image position, to obtain a tracking image of the target object; and controlling the camera to zoom according to the image position of the target object in any frame of the tracking image, to obtain a zoomed image of the target object.

[0247] In one of the embodiments, the processor further implements the following steps when executing the computer program:

[0248] converting the image position to a radar coordinate system to obtain an image converted position; and matching the image converted position with the radar position to obtain a matching result.

[0249] In one of the embodiments, the image converted position comprises a center converted position, which is a coordinate position of a center point of a detection frame of the target object in the radar coordinate system; and the processor further implements the following steps when executing the computer program:

[0250] obtaining an interval distance between the center converted position and the radar position;

[0251] in the case that the interval distance is less than a preset distance threshold, determining that the matching result is matching success;

[0252] in the case that the interval distance is greater than or equal to the preset distance threshold, determining that the matching result is matching failure.

[0253] In one of the embodiments, the processor further implements the following steps when executing the computer program:

[0254] converting the image position to a camera coordinate system of the camera in the current posture to obtain a backup position; obtaining a conversion relationship between the camera coordinate system and the radar coordinate system; and converting the backup position to the radar coordinate system according to the conversion relationship to obtain the image converted position.

[0255] In one of the embodiments, the image position comprises a first diagonal position and a second diagonal position of the detection frame, and the image converted position comprises a first converted position converted from the first diagonal position and a second converted position converted from the second diagonal position; and the processor further implements the following steps when executing the computer program:

[0256] The reference region in the radar coordinate system is determined according to the first conversion position and the second conversion position; and the matching result is determined according to the reference region and the radar position.

[0257] In one of the embodiments, the processor, when executing the computer program, further implements the following steps:

[0258] When the radar position is within the reference region, the matching result is determined as a matching success; and when the radar position is outside the reference region, the matching result is determined as a matching failure.

[0259] In one of the embodiments, the processor, when executing the computer program, further implements the following steps:

[0260] The reference distance and the reference direction of the image position relative to the center point position are determined according to the image position and the center point position of the monitoring image; and the camera is controlled to track the target object according to the reference distance and the reference direction, so as to obtain the tracking image of the target object.

[0261] In one of the embodiments, the processor, when executing the computer program, further implements the following steps:

[0262] The motion speed is determined according to the reference distance; and the camera is controlled to move along the reference direction at the motion speed, so as to obtain the tracking image of the target object.

[0263] In one of the embodiments, the processor, when executing the computer program, further implements the following steps:

[0264] For any frame of the monitoring image in the tracking image, the zoom region in the monitoring image is determined according to the image position of the target object in the monitoring image; and the camera is controlled to zoom according to the zoom region, so as to obtain the zoom image of the target object.

[0265] In one of the embodiments, the image position includes a first diagonal position and a second diagonal position of the detection frame; and the processor, when executing the computer program, further implements the following steps:

[0266] The target region in the monitoring image is determined according to the first diagonal position and the second diagonal position; and a region including the target region in the monitoring image is determined as the zoom region.

[0267] In one of the embodiments, the processor, when executing the computer program, further implements the following steps:

[0268] The zoom multiple is determined according to the monitoring image and the zoom region; and the camera is controlled to magnify the zoom region by the zoom multiple, so as to obtain the zoom image of the target object.

[0269] In one of the embodiments, a computer readable storage medium is provided, and the computer readable storage medium stores a computer program. The computer program is executed by a processor to implement the following steps:

[0270] obtaining an image position of a target object in a monitoring image captured by a camera in an image processing device; matching the radar position of the moving object with the image position; in the case of successful matching, controlling the camera to track the target object according to the image position, to obtain a tracking image of the target object; and controlling the camera to zoom according to the image position of the target object in any frame of the tracking image, to obtain a zoomed image of the target object.

[0271] In one of the embodiments, the computer program, when executed by the processor, further implements the following steps:

[0272] converting the image position to a radar coordinate system to obtain an image converted position; and matching the image converted position with the radar position to obtain a matching result.

[0273] In one of the embodiments, the image converted position includes a center converted position, which is a coordinate position of a center point of the detection frame of the target object in the radar coordinate system; and the computer program, when executed by the processor, further implements the following steps:

[0274] obtaining an interval distance between the center converted position and the radar position;

[0275] in the case that the interval distance is less than a preset distance threshold, determining that the matching result is a matching success;

[0276] in the case that the interval distance is greater than or equal to the preset distance threshold, determining that the matching result is a matching failure.

[0277] In one of the embodiments, the computer program, when executed by the processor, further implements the following steps:

[0278] converting the image position to a camera coordinate system of the camera in the current posture to obtain a backup position; obtaining a conversion relationship between the camera coordinate system and the radar coordinate system; and converting the backup position to the radar coordinate system according to the conversion relationship to obtain the image converted position.

[0279] In one of the embodiments, the image position includes a first diagonal position and a second diagonal position of the detection frame, and the image converted position includes a first converted position converted from the first diagonal position and a second converted position converted from the second diagonal position; and the computer program, when executed by the processor, further implements the following steps:

[0280] determining a reference region in the radar coordinate system according to the first converted position and the second converted position; and determining the matching result according to the reference region and the radar position.

[0281] In one of the embodiments, the computer program, when executed by the processor, further implements the following steps:

[0282] In a case where the radar position is located within the reference region, it is determined that the matching result is a matching success; in a case where the radar position is located outside the reference region, it is determined that the matching result is a matching failure.

[0283] In one of the embodiments, the computer program, when executed by the processor, further implements the following steps:

[0284] According to the image position and the center point position of the monitoring image, a reference distance and a reference direction of the image position relative to the center point position are determined; and the camera is controlled to track the target object according to the reference distance and the reference direction, so as to obtain a tracking image of the target object.

[0285] In one of the embodiments, the computer program, when executed by the processor, further implements the following steps:

[0286] According to the reference distance, a movement speed is determined; and the camera is controlled to move along the reference direction at the movement speed, so as to obtain the tracking image of the target object.

[0287] In one of the embodiments, the computer program, when executed by the processor, further implements the following steps:

[0288] For any frame of the monitoring image in the tracking image, a zoom region in the monitoring image is determined according to the image position of the target object in the monitoring image; and the camera is controlled to zoom according to the zoom region, so as to obtain a zoom image of the target object.

[0289] In one of the embodiments, the image position includes a first diagonal position and a second diagonal position of the detection frame; and the computer program, when executed by the processor, further implements the following steps:

[0290] According to the first diagonal position and the second diagonal position, a target region in the monitoring image is determined; and a region including the target region in the monitoring image is determined as the zoom region.

[0291] In one of the embodiments, the computer program, when executed by the processor, further implements the following steps:

[0292] According to the monitoring image and the zoom region, a zoom multiple is determined; and the camera is controlled to magnify the zoom region by the zoom multiple, so as to obtain the zoom image of the target object.

[0293] In one of the embodiments, a computer program product is provided, including a computer program, which, when executed by a processor, implements the following steps:

[0294] Obtaining an image position of a target object in a monitoring image collected by a camera in an image processing device; matching a radar position of a moving object with the image position; in a case of successful matching, controlling the camera to track the target object according to the image position, and obtaining a tracking image of the target object; and controlling the camera to zoom according to the image position of the target object in any frame of the tracking image, and obtaining a zoomed image of the target object.

[0295] In one of the embodiments, the computer program, when executed by the processor, further implements the following steps:

[0296] Converting the image position to a radar coordinate system to obtain an image converted position; and matching the image converted position with the radar position to obtain a matching result.

[0297] In one of the embodiments, the image converted position includes a center converted position, which is a coordinate position of a center point of a detection frame of the target object in the radar coordinate system; and the computer program, when executed by the processor, further implements the following steps:

[0298] Obtaining an interval distance between the center converted position and the radar position;

[0299] In a case where the interval distance is less than a preset distance threshold, determining that the matching result is matching success;

[0300] In a case where the interval distance is greater than or equal to the preset distance threshold, determining that the matching result is matching failure.

[0301] In one of the embodiments, the computer program, when executed by the processor, further implements the following steps:

[0302] Converting the image position to a camera coordinate system of the camera in a current posture to obtain a backup position; obtaining a conversion relationship between the camera coordinate system and the radar coordinate system; and converting the backup position to the radar coordinate system according to the conversion relationship to obtain the image converted position.

[0303] In one of the embodiments, the image position includes a first diagonal position and a second diagonal position of the detection frame, and the image converted position includes a first converted position converted from the first diagonal position and a second converted position converted from the second diagonal position; and the computer program, when executed by the processor, further implements the following steps:

[0304] Determining a reference region in the radar coordinate system according to the first converted position and the second converted position; and determining the matching result according to the reference region and the radar position.

[0305] In one of the embodiments, the computer program, when executed by the processor, further implements the following steps:

[0306] In a case that the radar position is located within the reference region, it is determined that the matching result is a matching success; in a case that the radar position is located outside the reference region, it is determined that the matching result is a matching failure.

[0307] In one of the embodiments, the computer program, when executed by the processor, further implements the following steps:

[0308] According to the image position and the center point position of the monitoring image, a reference distance and a reference direction of the image position relative to the center point position are determined; and the camera is controlled to track the target object according to the reference distance and the reference direction, so as to obtain a tracking image of the target object.

[0309] In one of the embodiments, the computer program, when executed by the processor, further implements the following steps:

[0310] According to the reference distance, a movement speed is determined; and the camera is controlled to move along the reference direction at the movement speed, so as to obtain the tracking image of the target object.

[0311] In one of the embodiments, the computer program, when executed by the processor, further implements the following steps:

[0312] For any frame of the monitoring image in the tracking image, a zoom region in the monitoring image is determined according to the image position of the target object in the monitoring image; and the camera is controlled to zoom according to the zoom region, so as to obtain a zoom image of the target object.

[0313] In one of the embodiments, the image position includes a first diagonal position and a second diagonal position of a detection frame; and the computer program, when executed by the processor, further implements the following steps:

[0314] According to the first diagonal position and the second diagonal position, a target region in the monitoring image is determined; and a region including the target region in the monitoring image is determined as the zoom region.

[0315] In one of the embodiments, the computer program, when executed by the processor, further implements the following steps:

[0316] According to the monitoring image and the zoom region, a zoom multiple is determined; and the camera is controlled to zoom the zoom region by the zoom multiple, so as to obtain the zoom image of the target object.

[0317] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (Read-Only Memory, ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric memory (Ferroelectric Random Access Memory, FRAM), phase change memory (Phase Change Memory, PCM), graphene memory, etc. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0318] Any combination of the technical features of the above embodiments can be made. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist contradictory, it should be considered as the scope of the present application.

[0319] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent of the present application. It should be noted that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of protection of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A target tracking method, wherein, The method is applied to a monitoring system comprising a radar and an image processing device, and comprises: obtaining an image position of a target object in a monitoring image captured by a camera in the image processing device; matching a radar position of a moving object with the image position; in a case of successful matching, controlling the camera to track the target object according to the image position, to obtain tracking video of the target object; controlling the camera to zoom according to the image position of the target object in any frame of the tracking video, to obtain zoom video of the target object.

2. The method of claim 1, wherein, The matching of the radar position of the moving object with the image position comprises: converting the image position to a radar coordinate system to obtain an image converted position; matching the image converted position with the radar position to obtain a matching result.

3. The method of claim 2, wherein, The image converted position comprises a center converted position, which is a coordinate position of a center point of a detection box of the target object in the radar coordinate system. The matching of the image converted position with the radar position to obtain a matching result comprises: obtaining an interval distance between the center converted position and the radar position; in a case that the interval distance is less than a preset distance threshold, determining that the matching result is matching success; in a case that the interval distance is greater than or equal to the preset distance threshold, determining that the matching result is matching failure.

4. The method of claim 2, wherein, The conversion of the image position to the radar coordinate system to obtain an image converted position comprises: converting the image position to a camera coordinate system of the camera in a current posture to obtain a backup position; obtaining a conversion relationship between the camera coordinate system and the radar coordinate system; converting the backup position to the radar coordinate system according to the conversion relationship to obtain the image converted position.

5. The method of claim 4, wherein, The image position comprises a first diagonal position and a second diagonal position of a detection box, and the image converted position comprises a first converted position converted from the first diagonal position and a second converted position converted from the second diagonal position; the matching of the image converted position with the radar position to obtain a matching result comprises: determining a reference region in the radar coordinate system according to the first converted position and the second converted position; determining the matching result according to the reference region and the radar position.

6. The method of claim 5, wherein, The determination of the matching result according to the reference region and the radar position comprises: in a case that the radar position is located within the reference region, determining that the matching result is matching success; in a case that the radar position is located outside the reference region, determining that the matching result is matching failure.

7. The method of any one of claims 1-6, wherein, The control of the camera to track the target object according to the image position to obtain tracking video of the target object comprises: determining a reference distance and a reference direction of the image position relative to a center point position of the monitoring image according to the image position and the center point position. Control the camera to track the target object according to the reference distance and the reference direction, and obtain a tracking image of the target object.

8. The method of claim 7, wherein, The control of the camera to track the target object according to the reference distance and the reference direction, and the obtaining of the tracking image of the target object, comprise: Determine a motion speed according to the reference distance; Control the camera to move along the reference direction at the motion speed, and obtain the tracking image of the target object.

9. The method of any one of claims 1-6, wherein, The control of the camera to zoom according to the image position of the target object in any frame of the tracking image, and the obtaining of the zoom image of the target object, comprise: For any frame of the tracking image, determine a zoom region in the frame according to the image position of the target object in the frame; Control the camera to zoom according to the zoom region, and obtain the zoom image of the target object.

10. The method of claim 9, wherein, The image position comprises a first diagonal position and a second diagonal position of a detection frame; the determination of the zoom region in the frame according to the image position of the target object in the frame, comprise: Determine a target region in the frame according to the first diagonal position and the second diagonal position; Determine a region comprising the target region in the frame as the zoom region.

11. The method of claim 9, wherein, The control of the camera to zoom according to the zoom region, and the obtaining of the zoom image of the target object, comprise: Determine a zoom multiple according to the frame and the zoom region; Control the camera to magnify the zoom region by the zoom multiple, and obtain the zoom image of the target object.

12. A target tracking device, wherein, The device comprises: A position obtaining module, configured to obtain an image position of a target object in a monitoring image collected by a camera in an image processing device; A position matching module, configured to match a radar position of a moving object with the image position; A target tracking module, configured to, in a case of successful matching, control the camera to track the target object according to the image position, and obtain a tracking image of the target object; A zoom control module, configured to control the camera to zoom according to the image position of the target object in any frame of the tracking image, and obtain a zoom image of the target object.

13. The apparatus of claim 12, wherein, The position matching module comprises: A coordinate conversion submodule, configured to convert the image position to a radar coordinate system, and obtain a converted image position; A position matching submodule, configured to match the converted image position with the radar position, and obtain a matching result.

14. The apparatus of claim 13, wherein, The converted image position comprises a center converted position, which is a coordinate position of a center point of a detection frame of the target object in the radar coordinate system; The position matching submodule comprises: A distance obtaining unit, configured to obtain an interval distance between the center converted position and the radar position; A matching success unit, configured to, in a case that the interval distance is less than a preset distance threshold, determine that the matching result is matching success. The matching failure unit is configured to determine that the matching result is a matching failure when the interval distance is greater than or equal to the preset distance threshold.

15. The apparatus of claim 13, wherein, The coordinate conversion submodule includes: The attitude conversion unit is configured to convert the image position to a camera coordinate system of the camera in a current attitude to obtain a backup position. The relationship acquisition unit is configured to acquire a conversion relationship between the camera coordinate system and the radar coordinate system. The position conversion unit is configured to convert the backup position to the radar coordinate system according to the conversion relationship to obtain the image conversion position.

16. The apparatus of claim 15, wherein, The image position includes a first diagonal position and a second diagonal position of a detection frame, and the image conversion position includes a first conversion position converted from the first diagonal position and a second conversion position converted from the second diagonal position. The position matching submodule includes: The reference region unit is configured to determine a reference region in the radar coordinate system according to the first conversion position and the second conversion position. The matching result unit is configured to determine the matching result according to the reference region and the radar position.

17. The apparatus of claim 16, wherein, The matching result unit includes: The success subunit is configured to determine that the matching result is a matching success when the radar position is located within the reference region. The failure subunit is configured to determine that the matching result is a matching failure when the radar position is located outside the reference region.

18. The apparatus of any one of claims 12-17, wherein, The target tracking module includes: The reference determination submodule is configured to determine a reference distance and a reference direction of the image position relative to a center point position of the monitoring image according to the image position and the center point position. The tracking control submodule is configured to control the camera to track the target object according to the reference distance and the reference direction to obtain a tracking image of the target object.

19. The apparatus of claim 18, wherein, The reference determination submodule includes: The speed determination unit is configured to determine a motion speed according to the reference distance. The motion control unit is configured to control the camera to move along the reference direction at the motion speed and acquire the tracking image of the target object.

20. The apparatus of any one of claims 12-17, wherein, The zoom control module includes: The zoom determination submodule is configured to determine a zoom region in any frame monitoring image of the tracking image according to an image position of the target object in the monitoring image. The region zoom submodule is configured to control the camera to zoom according to the zoom region to obtain a zoom image of the target object.

21. The apparatus of claim 20, wherein, The image position includes a first diagonal position and a second diagonal position of a detection frame; and the zoom determination module includes: The target determination unit is configured to determine a target region in the monitoring image according to the first diagonal position and the second diagonal position. The region determination unit is configured to determine a region including the target region in the monitoring image as the zoom region.

22. The apparatus of claim 20, wherein, The region zoom submodule includes: The multiple determination unit is configured to determine a zoom multiple according to the monitoring image and the zoom region. The zoom control unit is configured to control the camera to magnify the zoom region by the zoom multiple and acquire the zoom image of the target object.

23. A monitoring system comprising a radar and an image processing device, the image processing device comprising a camera, an attitude sensor, a memory and a processor, the memory storing a computer program, wherein, The processor, when executing the computer program, implements the steps of the method according to any one of claims 1 to 11.

Images