Integrated Work Video Browsing by Scenario Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video image systems fail to efficiently browse video images of multiple scenarios with low reliability, making it difficult to view desired scenarios.
Innovation Solution
A video image integrating device that detects time-series data of worker's skeleton or region, judges work conditions for each scenario, and generates integrated video images based on extraction settings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If video images of segments with low reliability are displayed, then the system provides detailed work analysis, but the efficiency of browsing multiple scenarios deteriorates
Solution Approach 1:
The patent segments video images into multiple scenario categories (e.g., normal work, abnormal work, specific task types) and selectively integrates only relevant segments based on the manager's browsing needs. This allows efficient navigation through categorized scenarios rather than displaying all low-reliability segments uniformly, thus improving browsing efficiency while maintaining detection precision for each category.
2Quantity of substance
If all video segments are integrated regardless of scenario type, then comprehensive work coverage is achieved, but the ability to efficiently browse specific scenarios deteriorates
Solution Approach 1:
The patent applies local quality by integrating video segments with different characteristics into separate scenario categories. Each category has its own integration criteria and presentation format, allowing managers to efficiently browse specific scenario types (e.g., only abnormal work scenarios or only specific task types) while maintaining comprehensive coverage of all video content across the different categories.
3Device complexity
If video images are integrated without scenario classification, then integration simplicity is maintained, but the ability to extract and browse multiple scenario types deteriorates
Solution Approach 1:
The patent adds a scenario classification dimension to the video integration process. Instead of a single linear integration sequence, video segments are organized into multiple dimensional categories (scenario types, reliability levels, time periods). This multi-dimensional organization enables efficient extraction and browsing of specific scenario types while maintaining a structured integration process that isn't overly complex.
Data Source
AI summary
The present disclosure provides a video image integrating device including an acquiring section acquiring video images in which work of a worker is captured; a detecting section; a judging section; and a generating section.


