CN-121999433-A - Production flow monitoring method, equipment and medium

CN121999433ACN 121999433 ACN121999433 ACN 121999433ACN-121999433-A

Abstract

The application relates to the technical field of production process monitoring and discloses a production process monitoring method, equipment and a medium. The method comprises the steps of obtaining a plurality of video frame images in a production monitoring video, extracting motion state characteristics of production materials in the video frame images to obtain material motion state characteristics comprising position characteristics and speed characteristics of the production materials, extracting operation state characteristics of production operations in the video frame images to obtain operation comprehensive state characteristics comprising operation type characteristics, compliance degree characteristics and abnormal type characteristics of the production operations, carrying out characteristic compliance evaluation on the material motion state characteristics and the operation comprehensive state characteristics, and generating flow compliance evaluation results aiming at production flows based on the characteristic compliance evaluation results. The embodiment of the application can improve the monitoring efficiency of the production flow monitoring and the sequential logic coherence analysis performance.

Inventors

Lai Peizhou
TANG JUNZHE
LIN YIQIN
LI SHUKAI
ZHANG WEIGANG

Assignees

深圳智眸未来科技有限公司

Dates

Publication Date: 20260508
Application Date: 20260114

Claims (10)

1. A method of monitoring a production process, comprising: acquiring a plurality of video frame images in a production monitoring video; Extracting motion state characteristics of the production materials in the video frame images to obtain material motion state characteristics comprising position characteristics and speed characteristics of the production materials; extracting operation state characteristics of production operation in the video frame image to obtain operation comprehensive state characteristics comprising operation type characteristics, procedure degree characteristics and abnormal type characteristics of the production operation; And carrying out feature compliance evaluation on the material motion state features and the operation comprehensive state features, and generating a flow compliance evaluation result aiming at the production flow based on the feature compliance evaluation result.
2. The method for monitoring a production process according to claim 1, wherein the extracting motion state features of the production materials in the video frame image comprises: extracting the position characteristics of the production materials in the video frame images to obtain the position characteristics of the production materials; Based on the position characteristics of the production materials and the time sequence characteristics of the video frame images, extracting the speed characteristics of the production materials in the video frame images to obtain the speed characteristics of the production materials; Fitting the position characteristic and the speed characteristic of the production material to obtain the material motion state characteristic.
3. The method of claim 2, wherein the extracting the speed feature of the production material in the video frame image based on the position feature of the production material and the time sequence feature of the video frame image comprises: Predicting a motion trail of the production material based on the position characteristics of the production material and the time sequence characteristics of the video frame image; and extracting the speed characteristics of the production materials in the video frame images based on the motion track of the production materials and the time sequence characteristics of the video frame images to obtain the speed characteristics of the production materials.
4. The method of claim 1, wherein the extracting the operation state feature of the production operation in the video frame image comprises: extracting space-time state characteristics of production operation in the video frame image to obtain operation space-time state characteristics representing the space-time characteristics of the production operation; performing multi-task classification processing on the operation space-time state characteristics to obtain operation type characteristics, procedure combination degree characteristics and abnormal type characteristics of the production operation; fitting the operation type feature, the combining procedure degree feature and the abnormal type feature to obtain the operation comprehensive state feature.
5. The method of claim 4, wherein the performing the space-time state feature extraction on the production operation in the video frame image comprises: Based on a slow branch of SlowFast networks, carrying out continuous convolution operation on the production operation in the video frame image to obtain the spatial domain characteristics of the production operation; Based on the quick branch of the SlowFast network, carrying out continuous convolution operation on the production operation in the video frame image to obtain the time domain characteristics of the production operation; fitting the spatial domain features and the temporal domain features to obtain the operation space-time state features.
6. The process flow monitoring method of claim 1, wherein the performing feature compliance evaluation on the material movement state feature and the operation integrated state feature comprises: performing motion state compliance evaluation on the material motion state characteristics to obtain characteristic compliance evaluation results of the material motion state characteristics; and carrying out local compliance evaluation and global compliance evaluation on the operation comprehensive state characteristics to obtain characteristic compliance evaluation results of the operation comprehensive state characteristics.
7. The method of claim 6, wherein said evaluating the motion state compliance of the material motion state features comprises: Determining the retention time length characteristic of the production materials in a preset area, the displacement distance characteristic in the preset time length and the material quantity characteristic in the preset area based on the material motion state characteristic; And determining a feature compliance evaluation result of the retention time length feature, the displacement distance feature and the material quantity feature based on the deviation between the retention time length feature, the displacement distance feature and the material quantity feature and the corresponding motion state compliance feature.
8. The process flow monitoring method of claim 6, wherein said performing a local compliance assessment and a global compliance assessment on said operational composite status feature comprises: determining a feature compliance evaluation result of the operation type feature, the compliance degree feature and the abnormality type feature based on the deviation between the operation type feature, the compliance degree feature and the abnormality type feature of the production operation and the corresponding local compliance feature; And determining a feature compliance evaluation result of the operation type feature based on the deviation between the operation type feature sequence and a preset compliance operation feature sequence, wherein the operation type feature sequence is obtained by arranging the operation type feature of the production operation based on the time sequence feature of the video frame image.
9. An electronic device comprising a memory storing a computer program and a processor that when executing the computer program implements the production flow monitoring method of any one of claims 1 to 8.
10. A computer readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the production flow monitoring method of any one of claims 1 to 8.

Description

Production flow monitoring method, equipment and medium Technical Field The application relates to the technical field of production process monitoring, in particular to a production process monitoring method, equipment and a medium. Background In modern industrial production, especially in the discrete manufacturing fields of automobile assembly, electronic equipment assembly and the like, compliance, efficiency and stability of production circulation and assembly flow are core elements for determining product quality and cost control. In the related art, the production process monitoring method mainly relies on manual inspection and paper recording, or adopts an isolated sensor network or single-function video monitoring, which has low efficiency and lacks the continuous analysis performance of sequential logic of the whole production process. Disclosure of Invention The application aims to provide a production flow monitoring method, equipment and medium, which aim to improve the monitoring efficiency and sequential logic coherence analysis performance of production flow monitoring. The embodiment of the application provides a production flow monitoring method, which comprises the following steps: acquiring a plurality of video frame images in a production monitoring video; Extracting motion state characteristics of the production materials in the video frame images to obtain material motion state characteristics comprising position characteristics and speed characteristics of the production materials; extracting operation state characteristics of production operation in the video frame image to obtain operation comprehensive state characteristics comprising operation type characteristics, procedure degree characteristics and abnormal type characteristics of the production operation; And carrying out feature compliance evaluation on the material motion state features and the operation comprehensive state features, and generating a flow compliance evaluation result aiming at the production flow based on the feature compliance evaluation result. In some embodiments, the extracting motion state features of the production materials in the video frame image includes: extracting the position characteristics of the production materials in the video frame images to obtain the position characteristics of the production materials; Based on the position characteristics of the production materials and the time sequence characteristics of the video frame images, extracting the speed characteristics of the production materials in the video frame images to obtain the speed characteristics of the production materials; Fitting the position characteristic and the speed characteristic of the production material to obtain the material motion state characteristic. In some embodiments, the extracting the speed feature of the production material in the video frame image based on the position feature of the production material and the time sequence feature of the video frame image includes: Predicting a motion trail of the production material based on the position characteristics of the production material and the time sequence characteristics of the video frame image; and extracting the speed characteristics of the production materials in the video frame images based on the motion track of the production materials and the time sequence characteristics of the video frame images to obtain the speed characteristics of the production materials. In some embodiments, the performing operation state feature extraction on the production operation in the video frame image includes: extracting space-time state characteristics of production operation in the video frame image to obtain operation space-time state characteristics representing the space-time characteristics of the production operation; performing multi-task classification processing on the operation space-time state characteristics to obtain operation type characteristics, procedure combination degree characteristics and abnormal type characteristics of the production operation; fitting the operation type feature, the combining procedure degree feature and the abnormal type feature to obtain the operation comprehensive state feature. In some embodiments, the performing spatio-temporal state feature extraction of production operations in the video frame image includes: Based on a slow branch of SlowFast networks, carrying out continuous convolution operation on the production operation in the video frame image to obtain the spatial domain characteristics of the production operation; Based on the quick branch of the SlowFast network, carrying out continuous convolution operation on the production operation in the video frame image to obtain the time domain characteristics of the production operation; fitting the spatial domain features and the temporal domain features to obtain the operation space-time state features. In some embodiments, the performing feature compliance asses