CN-114782381-B - Camera erection quality analysis method and device

CN114782381BCN 114782381 BCN114782381 BCN 114782381BCN-114782381-B

Abstract

The embodiment of the invention provides a camera erection quality analysis method and device, which relate to the technical field of video monitoring, and the method comprises the steps of determining a target statistical mode set for a target camera; the target statistical mode is a preset mode for analyzing multiple faces, a designated image shot by a target camera in the target statistical mode is determined, a deflection angle of the faces in the designated image about a designated direction is acquired, and an analysis result aiming at the target camera set quality is generated based on the acquired deflection angle. Through the scheme, the camera frame quality analysis can be more accurately and efficiently carried out.

Inventors

ZHAO BIYUAN
XU BIN
XIE DI
PU SHILIANG

Assignees

杭州海康威视数字技术股份有限公司

Dates

Publication Date: 20260505
Application Date: 20220427

Claims (12)

1. The camera erection quality analysis method is characterized by comprising the following steps of: Determining a target statistical mode set for a target camera, wherein the target statistical mode is a preset mode for analyzing multiple faces, and the images acquired in the target statistical mode for analyzing the scene where the target camera is located contain the multiple faces; Acquiring a designated image shot by the target camera in the target statistical mode, wherein the designated image is a multi-frame image shot by the target camera in a collection time range under the condition that the target statistical mode is a multi-frame mode, and the collection time range is an image selection parameter corresponding to the multi-frame mode; The method comprises the steps of detecting a deflection angle of a face in a specified image in a specified direction through an omnidirectional face detection algorithm, wherein the deflection angle of the face in the specified direction is a rolling angle, a detection result for face detection obtained through the omnidirectional face detection algorithm is characterized by using a six-dimensional array, the six-dimensional array comprises coordinates of a center point of a positive rectangular frame, the width of the positive rectangular frame, the height of the positive rectangular frame and the offset of a detection frame of the face relative to the center point of the positive rectangular frame, the positive rectangular frame is a circumscribed rectangular frame of the detection frame of the face, and the deflection angle of the face in the specified image in the specified direction is calculated based on the width of the positive rectangular frame, the height of the positive rectangular frame and the offset of the detection frame of the face relative to the center point of the positive rectangular frame; and carrying out weighted average processing on the acquired deflection angles to obtain the deflection angles of the target camera during erection, wherein the weight of the deflection angles acquired based on the designated images acquired in the peak period of the human flow is higher than the weight of the deflection angles acquired based on the designated images acquired in other times.
2. The method according to claim 1, wherein the obtaining the specified image captured by the target camera in the target statistics mode includes: Determining image selection parameters corresponding to the target statistical modes, wherein different statistical modes correspond to different image selection parameters; And selecting the designated image shot by the target camera based on the image selection parameters.
3. The method of claim 2, wherein the target statistical mode is a multi-frame mode or a single-frame mode; the image selection parameters corresponding to the multi-frame mode also comprise image frame numbers; the image selection parameters corresponding to the single frame mode comprise a threshold value of the number of faces existing in the image.
4. The method according to claim 2, wherein the target statistics mode is a fixed face number mode or a fixed duration mode; the image selection parameters corresponding to the fixed face number mode comprise the number of target faces; the image selection parameters corresponding to the fixed duration mode comprise acquisition duration.
5. A method according to claim 3, wherein selecting the specified image captured by the target camera based on the image selection parameter comprises: and if the target statistical mode is a multi-frame mode, selecting multi-frame images which accord with the image frame number and the acquisition time range from the images shot by the target camera, and obtaining multi-frame designated images.
6. The method according to claim 5, wherein, before selecting a plurality of frames of images conforming to the number of frames of the image and the acquisition time range from the images captured by the target camera to obtain a plurality of frames of designated images, the method further comprises: determining auxiliary parameters corresponding to the multi-frame mode, wherein the auxiliary parameters comprise a threshold value of the number of faces existing in an image; Selecting a plurality of frames of images which accord with the image frame number and the acquisition time range from the images shot by the target camera to obtain a plurality of frames of designated images, wherein the method comprises the following steps of: and selecting a plurality of frames of images which are in line with the image frame number and the acquisition time range and contain the number of faces which are not lower than the number threshold value from the images shot by the target camera, and obtaining a plurality of frames of designated images.
7. A method according to claim 3, wherein selecting the specified image captured by the target camera based on the image selection parameter comprises: And if the target statistical mode is a single-frame mode, selecting a frame of image with the number of the faces not lower than the number threshold from the images shot by the target camera, and obtaining a frame of designated image.
8. The method according to claim 4, wherein selecting the specified image captured by the target camera based on the image selection parameter includes: And if the target statistical mode is a fixed face number mode, selecting at least one frame of image with the total accumulated face number being the target face number from the images acquired by the target camera, and obtaining a designated image.
9. The method according to claim 4, wherein selecting the specified image captured by the target camera based on the image selection parameter includes: And if the target statistical mode is a fixed duration mode, selecting at least one frame of image acquired by the target camera within the acquisition duration from the images shot by the target camera, and obtaining at least one frame of designated image.
10. The method of any one of claims 1-8, wherein the face deflection angle with respect to the specified direction comprises a face deflection angle with respect to a vertical direction of the image, a face deflection angle with respect to a plumb line, or a face deflection angle with respect to a vertical direction of the image display screen.
11. The method according to claim 1, wherein the method further comprises: when the deviation angle exceeds a preset threshold, determining an adjustment direction and/or an adjustment angle corresponding to the deviation angle, and outputting a camera adjustment notification containing the adjustment direction and/or the adjustment angle; Or alternatively And outputting a camera adjustment notification containing the deviation angle when the deviation angle exceeds a preset threshold value.
12. A camera-set-up mass analysis device, the device comprising: The system comprises a first determining module, a first judging module, a second judging module and a third judging module, wherein the first determining module is used for determining a target statistical mode set for a target camera, the target statistical mode is a preset mode for analyzing a plurality of faces, and the image acquired in the target statistical mode and used for analyzing the scene where the target camera is located contains the plurality of faces; The system comprises a target camera, a second determining module, a first determining module and a second determining module, wherein the target camera is used for acquiring a specified image shot by the target camera in the target statistical mode, the specified image is a multi-frame image shot by the target camera in an acquisition time range when the target statistical mode is a multi-frame mode, and the acquisition time range is an image selection parameter corresponding to the multi-frame mode; The face detection system comprises an acquisition module, a six-dimensional array, a correction module and a correction module, wherein the acquisition module is used for detecting the deflection angle of a face in a specified image in a specified direction through an omnidirectional face detection algorithm, the deflection angle of the face in the specified direction is a rolling angle, a detection result for face detection obtained through the omnidirectional face detection algorithm is characterized by using the six-dimensional array, the six-dimensional array comprises coordinates of a center point of a positive rectangular frame, the width of the positive rectangular frame, the height of the positive rectangular frame and the offset of a detection frame of the face relative to the center point of the positive rectangular frame, the positive rectangular frame is a circumscribed rectangular frame of the detection frame of the face, and the deflection angle of the face in the specified image in the specified direction is calculated based on the width of the positive rectangular frame, the height of the positive rectangular frame and the offset of the detection frame of the face relative to the center point of the positive rectangular frame; the generation module is used for carrying out weighted average processing on the acquired deflection angles to obtain the deflection angles of the target camera during erection, wherein the weight of the deflection angles acquired based on the designated images acquired in the peak period of the human flow is higher than the weight of the deflection angles acquired based on the designated images acquired in other times.

Description

Camera erection quality analysis method and device Technical Field The invention relates to the technical field of video monitoring, in particular to a camera erection quality analysis method and device. Background Today, cameras have been widely deployed in various locations, such as locations with high traffic or locations with less traffic. Moreover, after the camera is erected, the angle may deviate, so that the photographed scene picture is inclined. In order to optimize the erection of the camera, it is generally necessary to analyze the erection quality of the camera, so as to adjust the angle of the camera based on the analysis result. It can be seen that how to accurately and efficiently perform camera-set quality analysis is a problem to be solved. Disclosure of Invention The embodiment of the invention aims to provide a camera erection quality analysis method and device so as to accurately and efficiently carry out camera erection quality analysis. The specific technical scheme is as follows: in a first aspect, an embodiment of the present invention provides a method for analyzing quality of camera erection, including: determining a target statistical mode set for a target camera, wherein the target statistical mode is a preset mode for analyzing multiple faces; acquiring a designated image shot by the target camera under the target statistical mode; acquiring a deflection angle of a face in the appointed image about an appointed direction; Based on the acquired deflection angle, an analysis result for the target camera set-up quality is generated. Optionally, the obtaining the specified image obtained by shooting by the target camera in the target statistics mode includes: Determining image selection parameters corresponding to the target statistical modes, wherein different statistical modes correspond to different image selection parameters; And selecting the designated image shot by the target camera based on the image selection parameters. Optionally, the target statistical mode is a multi-frame mode or a single-frame mode; the image selection parameters corresponding to the multi-frame mode comprise image frame number and/or acquisition time range; the image selection parameters corresponding to the single frame mode comprise a threshold value of the number of faces existing in the image. Optionally, the target statistics mode is a fixed face number mode or a fixed duration mode; the image selection parameters corresponding to the fixed face number mode comprise the number of target faces; the image selection parameters corresponding to the fixed duration mode comprise acquisition duration. Optionally, the selecting, based on the image selection parameter, the designated image captured by the target camera includes: and if the target mode is a multi-frame mode, selecting multi-frame images conforming to the image frame number and/or the acquisition time range from the images shot by the target camera, and obtaining multi-frame designated images. Optionally, before selecting a plurality of frames of images according with the image frame number and/or the acquisition time range from the images obtained by shooting the target camera to obtain a plurality of frames of designated images, the method further includes: determining auxiliary parameters corresponding to the multi-frame mode, wherein the auxiliary parameters comprise a threshold value of the number of faces existing in an image; Selecting a multi-frame image conforming to the image frame number and/or the acquisition time range from the images shot by the target camera to obtain a multi-frame designated image, wherein the method comprises the following steps of: And selecting a plurality of frames of images which are in line with the frame number and/or the acquisition time range of the images and the number of faces contained in each image is not lower than the number threshold value from the images shot by the target camera, and obtaining a plurality of frames of designated images. Optionally, the selecting, based on the image selection parameter, the designated image captured by the target camera includes: And if the target statistical mode is a single-frame mode, selecting a frame of image with the number of the faces not lower than the number threshold from the images shot by the target camera, and obtaining a frame of designated image. Optionally, the selecting, based on the image selection parameter, the designated image captured by the target camera includes: And if the target statistical mode is a fixed face number mode, selecting at least one frame of image with the total accumulated face number being the target face number from the images acquired by the target camera, and obtaining a designated image. Optionally, the selecting, based on the image selection parameter, the designated image captured by the target camera includes: And if the target statistical mode is a fixed duration mode, selecting at least one frame of imag