AI Optical Fiber Splice Verification for Abnormality Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the fusion-spliced state of optical fibers are inadequate, leading to difficulties in identifying abnormal splices, which can hinder communication and require time-consuming rework, especially when workers and management servers are from different companies.
Innovation Solution
A verification apparatus and method using artificial intelligence to analyze images of spliced optical fibers, classifying defects by type, calculating similarities, and determining abnormality based on set ranges and correlations, with a work server providing real-time feedback to workers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual inspection of spliced optical fibers is performed, then workers can identify abnormal splices, but the process is time-consuming and requires rework before workers leave the job site
Solution Approach 1:
The patent replaces manual visual inspection with an automated image analysis system using AI algorithms. The system captures images of spliced optical fibers and automatically analyzes them to determine splice quality, eliminating the time-consuming manual inspection process and enabling rapid verification before workers leave the site
Solution Approach 2:
The patent introduces a work server as an intermediary between the splicing operation and management servers. This work server performs real-time image analysis and communicates results back to workers, providing immediate feedback without requiring workers to wait for management server verification, thus reducing time loss
2Measurement precision
If simple defect detection is performed on spliced portions, then abnormalities can be identified, but all optical lines with any defects may be incorrectly flagged as abnormal requiring rework
Solution Approach 1:
The patent changes the evaluation parameters from simple defect presence/absence to a comprehensive analysis considering multiple factors including defect type classification, defect location, defect size, and correlation between multiple defects. This allows the system to distinguish between critical defects requiring rework and acceptable defects that do not impact splice quality
Solution Approach 2:
The patent applies different evaluation criteria to different types of defects based on their specific characteristics. Instead of treating all defects uniformly, the system analyzes each defect type (e.g., misalignment, gap, contamination) with appropriate thresholds and weightings, allowing accurate assessment that maintains productivity by avoiding false positives
3Extent of automation
If work servers are provided separately from splicers, then real-time analysis can be performed, but communication between workers and management servers increases complexity
Solution Approach 1:
The patent implements a feedback mechanism where the work server analyzes splice images in real-time and immediately communicates results to workers through the splicer interface. This closed-loop feedback system automates the verification process while maintaining simple worker interaction, as workers receive direct pass/fail indications and guidance without navigating complex communication protocols
Data Source
AI summary
A verification apparatus is provided. The verification apparatus includes: an acquisition unit configured to acquire a photographed image of an optical line for which a splicing operation is completed; an analysis unit configured to extract a defect of a spliced portion of the optical line through analysis of the image; and a determination unit configured to determine the presence or absence of an abnormality in the spliced portion based on the extracted defect.