CN-121999250-A - Load examination judging method and device based on image recognition, equipment and medium

CN121999250ACN 121999250 ACN121999250 ACN 121999250ACN-121999250-A

Abstract

The application provides a load examination judging method, a device, equipment and a medium based on image recognition, which belong to the field of image recognition, wherein the method comprises the steps of obtaining a target image; if the definition of the target image is smaller than or equal to a preset definition threshold, compensating the target image based on a standard compensation image in a standard compensation image library determined in advance to obtain a compensation image corresponding to the target image, extracting image features of the compensation image to obtain real-time image features and virtual-time image features, calculating the matching degree of the real-time image features and a standard examination feature library to obtain a first matching degree, calculating the matching degree of the virtual-time image features and the standard examination feature library to obtain a second matching degree, and judging whether the load examination is illegal or not according to the first matching degree and the second matching degree. The load examination judging method, the load examination judging device, the load examination judging equipment and the load examination judging medium based on the image recognition can improve the accuracy and the efficiency of the load examination of special equipment.

Inventors

WANG AIXIANG
XIN JING
WANG JIUXIA
LIU HUIMIN

Assignees

河北省特种设备监督检验研究院

Dates

Publication Date: 20260508
Application Date: 20260130

Claims (10)

1. The load examination judging method based on image recognition is characterized by comprising the following steps of: acquiring a target image, wherein the target image is a ground image acquired on a designated examination path in the load examination process; If the definition of the target image is smaller than or equal to a preset definition threshold, compensating the target image based on a standard compensation image in a predetermined standard compensation image library to obtain a compensation image corresponding to the target image; Extracting image features of the compensation image to obtain real-time image features and virtual-time image features, wherein the real-time image features are image features contained in the target image, and the virtual-time image features are image features contained in the standard compensation image; Calculating the matching degree of the real-time image features and a standard examination feature library to obtain a first matching degree; calculating the matching degree of the virtual image features and the standard examination feature library to obtain a second matching degree; Judging whether the load examination is illegal or not according to the first matching degree and the second matching degree.
2. The image recognition-based load test determination method according to claim 1, wherein the compensating the target image based on the standard compensation image in the predetermined standard compensation image library comprises: determining a blurred region of the target image based on the sharpness; If the duty ratio of the key region in the fuzzy region is larger than or equal to a preset duty ratio threshold, dynamic filtering is carried out on the key region based on the definition of the fuzzy region; Compensating the filtered target image based on standard compensation images in a predetermined standard compensation image library; the key area is an area including an appointed examination path image in the load examination process.
3. The method for determining a loading examination based on image recognition according to claim 2, wherein the dynamically filtering the key region based on the sharpness of the blurred region comprises: calculating a standard deviation of a gaussian filter based on a sharpness of the blurred region, the standard deviation increasing with decreasing sharpness; and determining a Gaussian filtering range based on the standard deviation, and dynamically filtering the key region based on the Gaussian filtering range.
4. The image recognition-based load test decision method as recited in claim 3, wherein said calculating a standard deviation of gaussian filtering based on the sharpness of said blurred region comprises: Dividing the fuzzy region into a plurality of subareas, and determining the definition of each subarea; calculating a normalized ambiguity for the ambiguous region based on the sharpness of each sub-region; and calculating the standard deviation of Gaussian filtering according to the normalized ambiguity.
5. The method for determining a load test based on image recognition according to claim 1, wherein the real-time image features correspond to a first weight, the virtual-image features correspond to a second weight, and the first weight is greater than the second weight; The calculating the matching degree of the real-time image features and the standard examination feature library to obtain a first matching degree comprises the following steps: and calculating the matching degree of the real-time image features and the standard examination feature library based on the first weight to obtain a first matching degree.
6. The method for determining load test based on image recognition according to claim 5, wherein the calculating the matching degree between the image feature of the virtual image and the standard test feature library to obtain the second matching degree comprises: And calculating the matching degree of the virtual image features and the standard examination feature library based on the second weight to obtain a second matching degree.
7. The method for determining a loading test based on image recognition according to claim 1, wherein the determining whether the loading test is illegal according to the first matching degree and the second matching degree comprises: Calculating the sum of the first matching degree and the second matching degree; if the sum is greater than or equal to a preset matching degree threshold value, judging that the load test is compliant; and if the sum is smaller than a preset matching degree threshold value, judging that the load examination is illegal.
8. A load test determination device based on image recognition, comprising: The system comprises an image acquisition module, a load examination module and a load examination module, wherein the image acquisition module is used for acquiring a target image, wherein the target image is a ground image acquired on a specified examination path in the load examination process; The image compensation module is used for compensating the target image based on standard compensation images in a predetermined standard compensation image library if the definition of the target image is smaller than or equal to a preset definition threshold value, so as to obtain a compensation image corresponding to the target image; The feature extraction module is used for extracting the image features of the compensation image to obtain real-image features and virtual-image features, wherein the real-image features are the image features contained in the target image, and the virtual-image features are the image features contained in the standard compensation image; The first calculation module is used for calculating the matching degree of the real-time image features and the standard examination feature library to obtain a first matching degree; The second calculation module is used for calculating the matching degree of the image features of the virtual image and the standard examination feature library to obtain a second matching degree; And the examination judging module is used for judging whether the load examination is illegal or not according to the first matching degree and the second matching degree.
9. An electronic device comprising a processor and a memory, wherein the memory stores a computer program that is loaded and executed by the processor to implement the image recognition-based load test determination method of any one of claims 1 to 7.
10. A computer-readable storage medium, in which a computer program is stored, the computer program being loaded and executed by a processor to implement the image recognition-based load test determination method of any one of claims 1 to 7.

Description

Load examination judging method and device based on image recognition, equipment and medium Technical Field The application belongs to the technical field of image recognition, and particularly relates to a load examination judging method, device, equipment and medium based on image recognition. Background In practical operation examination of special equipment operators, crane drivers, forklift drivers and the like are generally required to move loads according to a specified route and accurately place the loads to specified positions. The examination process has clear requirements on the normalization of the load running path, the correctness of the running direction and the accuracy of the final placement position. Although the existing part adopts a camera or a sensor to detect the operation process of special equipment, the special equipment is mainly used for industrial positioning or safety monitoring, most of monitoring is the final drop point position, and the requirement of real-time monitoring and identification in examination scenes is difficult to meet. Disclosure of Invention The application provides a load examination judging method, a device, equipment and a medium based on image recognition, which can improve the accuracy and instantaneity of operation examination judgment of special equipment through image recognition. According to an aspect of the embodiment of the present application, there is provided a load examination judging method based on image recognition, including: acquiring a target image, wherein the target image is a ground image acquired on a designated examination path in the load examination process; if the definition of the target image is smaller than or equal to a preset definition threshold, compensating the target image based on a standard compensation image in a predetermined standard compensation image library to obtain a compensation image corresponding to the target image; Extracting image features of the compensation image to obtain real-image features and virtual-image features, wherein the real-image features are image features contained in the target image, and the virtual-image features are image features contained in the standard compensation image; calculating the matching degree of the real-time image features and the standard examination feature library to obtain a first matching degree; calculating the matching degree of the image features of the virtual image and the standard examination feature library to obtain a second matching degree; judging whether the load examination is illegal or not according to the first matching degree and the second matching degree. According to an aspect of the embodiment of the present application, there is provided a load examination determining device based on image recognition, including: The system comprises an image acquisition module, a load examination module and a load examination module, wherein the image acquisition module is used for acquiring a target image, wherein the target image is a ground image acquired on a specified examination path in the load examination process; The image compensation module is used for compensating the target image based on a standard compensation image in a predetermined standard compensation image library if the definition of the target image is smaller than or equal to a preset definition threshold value, so as to obtain a compensation image corresponding to the target image; The feature extraction module is used for extracting the image features of the compensation image to obtain real-image features and virtual-image features, wherein the real-image features are the image features contained in the target image, and the virtual-image features are the image features contained in the standard compensation image; The first calculation module is used for calculating the matching degree of the real-time image features and the standard examination feature library to obtain a first matching degree; The second calculation module is used for calculating the matching degree of the image features of the virtual image and the standard examination feature library to obtain a second matching degree; And the examination judging module is used for judging whether the load examination is illegal or not according to the first matching degree and the second matching degree. According to an aspect of an embodiment of the present application, there is provided an electronic device including a processor and a memory, in which a computer program is stored, the computer program being loaded and executed by the processor to implement the above-described load test determination method based on image recognition. According to one aspect of an embodiment of the present application, the computer program product comprises a computer program stored in a computer readable storage medium. The processor of the electronic device reads the computer program from the computer-readable storage medium, and the proces