Cable Cross-Section Image Scratch Elimination via Morphological TV Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting the number of conductors in a cable's cross-sectional image are hindered by scratches generated during the sample cutting process, leading to reduced accuracy in quality detection.
Innovation Solution
A method utilizing an improved total variation (TV) algorithm, which involves acquiring a cross-sectional image of a cable, performing dilation and erosion, fusing the images, and applying the TV algorithm to eliminate scratches, thereby enhancing the accuracy of quality detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If conventional image denoising algorithms (spatial-domain filters, transform-domain methods, or basic TV algorithm) are used to eliminate scratches, then the computation is simplified or the algorithm is easier to implement, but the image quality deteriorates with staircase effect, texture loss, or blurring
Solution Approach 1:
The patent combines multiple image processing techniques into a composite algorithm: morphological operations (dilation and erosion) are applied first to remove small noise, followed by the TV algorithm for scratch elimination. This composite approach leverages the strengths of each method while mitigating their individual weaknesses, achieving both ease of implementation and high image quality
Solution Approach 2:
The denoising process is segmented into distinct stages: first applying morphological operations to handle small-scale noise, then applying the TV algorithm specifically targeted at scratch removal. This segmentation allows each algorithm to focus on specific types of defects, improving overall effectiveness while maintaining computational efficiency
2Manufacturing precision
If deep learning-based denoising algorithms are used to eliminate scratches, then the image quality and scratch elimination performance are improved, but the data requirement for model training cannot be met due to limited sampled data and high cost of cable samples
Solution Approach 1:
The patent uses conventional, computationally efficient algorithms (morphological operations and TV algorithm) that do not require expensive deep learning models. These algorithms can be implemented with minimal training data or even without training, making them suitable for applications where sample data is limited and expensive to obtain
Solution Approach 2:
The patent replaces the need for complex deep learning mechanical systems with a combination of mathematical morphology and partial differential equation-based TV algorithm. This substitution achieves comparable or superior scratch elimination performance without requiring large datasets for model training
3Device complexity
If the basic TV algorithm is used to eliminate scratches, then the algorithm is simple and computationally efficient, but the staircase effect causes texture loss and image blurring
Solution Approach 1:
The patent applies morphological operations (dilation and erosion) as a preliminary step before applying the TV algorithm. This preliminary action removes small noise and prepares the image, allowing the TV algorithm to focus specifically on scratch removal without being distracted by small-scale artifacts, thereby reducing the staircase effect and preserving image details
Data Source
AI summary
The present disclosure provides a method for eliminating scratches in a cross-section image of a cable, comprising: acquiring an original image of a cross-section of the cable, the original image having a first modality; performing, by a processor, dilation and erosion on the original image to obtain a processed image, the processed image having a second modality; fusing, by the processor, the original image and the processed image to obtain a fused image; processing, by the processor, the fused image to obtain a scratch eliminated image; and applying the scratch eliminated image in actual quality detection of the cable to improve accuracy of the quality detection. By introducing multimodal features, the method relieves the loss of image information. The method can keep the information of the original image as much as possible, and can significantly improve the accuracy of cable quality detection.