AI Vision Check System for Industrial Process Monitoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current check systems in industrial production processes are prone to human errors, are not versatile, and existing automated systems are ineffective, complicated, and unreliable.
Innovation Solution
A check system utilizing machine vision and artificial intelligence, specifically deep learning, to monitor and control industrial processes in real-time by analyzing 2D and 3D images from cameras, ensuring the correctness and safety of activities without direct interference, and providing objective assessments of process quality and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If human operators perform process checks, then flexibility and adaptability are maintained, but human errors occur reducing reliability
Solution Approach 1:
The patent replaces human operators with an automated check system using machine vision and artificial intelligence. Image acquisition devices capture visual data of the production process, and AI algorithms automatically analyze the images to detect deviations from execution rules, eliminating human errors while maintaining adaptability through programmable rule sets
Solution Approach 2:
The system enables self-monitoring of the production process through automated image analysis. The check system independently executes monitoring tasks using predefined execution rules stored in memory, automatically comparing actual process states against rules without requiring human intervention, thereby achieving both reliability and adaptability
2Reliability
If automated check systems with sensors and alarms are implemented, then reliability improves, but device complexity and installation difficulty increase
Solution Approach 1:
The patent replaces complex physical sensors and alarms with a simplified machine vision-based system. Instead of multiple specialized sensors for different parameters, a single or few image acquisition devices capture comprehensive visual information that AI algorithms then analyze to detect various process deviations, reducing hardware complexity while maintaining reliability
Solution Approach 2:
The image acquisition devices serve multiple functions: capturing process state images, detecting object positions, monitoring tool usage, and identifying process deviations. This multi-functional approach eliminates the need for separate specialized sensors for each monitoring task, thereby reducing device complexity and installation difficulty while maintaining comprehensive monitoring capability
3Adaptability or versatility
If traditional automated check systems are used, then some automation is achieved, but versatility and adaptability to different processes are limited
Solution Approach 1:
The system employs dynamic execution rules that can be easily modified to适应 different production processes. The AI algorithms adapt to various process types by loading appropriate execution rules from memory, allowing the same hardware platform to monitor diverse processes such as assembly operations, quality inspection, and safety monitoring without physical reconfiguration
Solution Approach 2:
The system uses image copies of the actual production process for analysis. Instead of directly interfering with the physical process, the machine vision system creates digital representations through image acquisition, and the AI algorithms analyze these copies to detect deviations, enabling versatile monitoring across different processes without physical modifications to the production line
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
A check procedure of a production process of an artifact (6), wherein image acquisition means (4) acquire 2D and 3D images from a work area (1). Said images are processed in an analysis check (100), using neural networks, to extract data that are sent to a check cycle (300) that checks said data in order to inform alarm situations by means of signaling means (5).