Battery Tab Corner Detection for Misalignment Inspection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current tab misalignment detection methods in lithium-ion polymer batteries, particularly using template matching algorithms, suffer from poor accuracy and robustness due to environmental factors and noise, leading to unreliable detection results.
Innovation Solution
A method involving key point detection on tab images taken under front irradiation to identify corner points, using key point position maps to determine the actual width of the tab, and comparing it with a preset width to detect misalignment, employing a key point detection model trained on high-resolution images to enhance accuracy and robustness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If template matching algorithm is used for tab misalignment detection, then the detection process can be implemented, but the accuracy and robustness are poor due to environmental factors and noise
Solution Approach 1:
The patent replaces the traditional template matching algorithm (which relies on geometric pattern recognition) with a deep learning-based key point detection model. This substitution of the detection mechanism enables the system to achieve higher accuracy and robustness by training on extensive data that includes various environmental conditions, lighting variations, and tab configurations, thereby resolving the contradiction between detection implementability and detection reliability
Solution Approach 2:
The patent changes the fundamental parameters of the detection approach by transitioning from manual template definition to automated key point detection through neural networks. The model learns to identify corner points and calculate tab dimensions directly from images, adapting to different environmental factors and noise levels, thus improving both accuracy and robustness simultaneously
2Measurement precision
If tab image is taken under back irradiation, then the tab structure can be visualized, but the corner point identification is difficult and accuracy is reduced
Solution Approach 1:
The patent inverts the traditional imaging approach by using front irradiation instead of back irradiation to capture tab images. This inversion allows the light to pass through the separator and illuminate the tab from the front, creating clearer images with better contrast for corner point identification. The inverted imaging strategy eliminates the harmful effects of back irradiation while maintaining the ability to visualize tab structure, thereby resolving the contradiction between visualization capability and measurement precision
3Measurement precision
If key point detection model is trained on high-resolution images, then the detection accuracy is improved, but the processing time and computational resources increase
Solution Approach 1:
The patent applies preliminary action by pre-training the key point detection model on high-resolution images during the development phase. Once trained, the model can efficiently process new images without requiring real-time high-resolution processing, as the complex feature extraction and pattern recognition have already been learned during training. This preliminary training approach resolves the contradiction by preparing the system in advance to handle high-accuracy requirements without incurring the full computational cost during actual detection
Data Source
AI summary
This application provides a tab misalignment detection method performed by a computer device, The method comprises: performing key point detection on a tab image of a current tab of a lithium-ion battery cell to obtain multiple key point position maps indicating multiple corner points of the current tab respectively; determining, based on the multiple key point position maps, position information of the current tab, the position information indicating positions of the multiple corner points of the current tab in the tab image; determining, based on the position information, an actual width of the current tab; and determining that the current tab is misaligned when a difference between the actual width and a preset width that is greater than a difference threshold.