Anti-interference industrial vision counting method and device based on spatio-temporal context perception

By identifying the intersection ratio in real-time video streams and analyzing target video segments, the problems of miscounting and duplicate counting in industrial vision counting are solved, and accurate counting based on spatiotemporal context awareness is achieved.

CN122199550APending Publication Date: 2026-06-12SHANGHAI DONGFANG HOPE SOFTWARE TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHANGHAI DONGFANG HOPE SOFTWARE TECH CO LTD
Filing Date
2026-05-15
Publication Date
2026-06-12

AI Technical Summary

⚠Technical Problem

Existing industrial vision counting methods cannot accurately count the number of target objects, especially when the target objects are in a complex state of motion, which can easily lead to miscounting or double counting.

⚗Method used

By acquiring real-time video streams of the monitored area, the system identifies and detects targets and calculates the intersection ratio. Targets with an intersection ratio greater than or equal to a threshold are identified as targets for counting and their videos are recorded. The system analyzes the instantaneous information in the target video segments and generates counting results based on the decision displacement direction and category.

🎯Benefits of technology

It achieves accurate counting of target objects, eliminates counting deviations caused by image jitter, reflection, or occlusion, and improves the accuracy and reliability of counting.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122199550A_ABST
    Figure CN122199550A_ABST
Patent Text Reader

Abstract

The application provides an anti-interference industrial visual counting method and device based on space-time context perception, which comprises the following steps: acquiring real-time video stream of a monitoring area and recognizing a detection target; calculating intersection proportion of the detection target and a target area; triggering video recording to acquire a target video segment when the intersection proportion is greater than or equal to an intersection threshold; extracting instantaneous position and instantaneous category of a counting target in each video frame of the target video segment; generating a position sequence based on the instantaneous position and calculating a decision displacement direction; statistically analyzing the instantaneous category and determining a decision category; comparing the decision displacement direction and the decision category with preset effective direction and effective category to generate a detection result; and counting the number of effective results in a preset detection time to obtain a counting result. The application realizes classified and directional industrial visual counting by analyzing the motion state of the counting target in continuous time.
Need to check novelty before this filing date? Find Prior Art