Textile Visual Quality Control Using Image Sub-areas
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing quality control processes for warning protection clothing are time-consuming, subjective, and lack objective precision, particularly in assessing the luminosity and reflection capacity of textiles after washing and wear, which can lead to non-compliance with functional standards.
Innovation Solution
A procedure and device for visual quality control that involves selecting specific sub-areas of the textile for analysis based on image processing techniques, including edge recognition, morphological processing, and color-based exemptions, using multiple images to evaluate pixel brightness and reflection, and comparing these evaluations against threshold values for objective assessment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual visual inspection by trained personnel is used, then quality control can be performed with simple equipment, but the process is subjective and lacks precision
Solution Approach 1:
The patent replaces manual visual inspection by trained personnel with an automated image processing system. The system captures images of the textile, processes them through algorithms to detect defects, and provides objective quality assessment. This substitution eliminates subjectivity while maintaining operational simplicity through software-based analysis rather than complex hardware.
2Reliability
If automated camera systems are used for quality control, then objectivity and precision are improved, but the inspection process becomes time-consuming
Solution Approach 1:
The patent segments the quality control process into distinct image processing steps: capturing multiple images at different exposure settings, selectively processing relevant areas, and analyzing specific quality parameters. This segmentation allows parallel processing of different image sets and focuses computational resources on critical defect detection areas, significantly reducing total inspection time while maintaining comprehensive quality assessment.
Solution Approach 2:
The system performs preliminary actions by capturing multiple images at different exposure settings simultaneously before the actual quality analysis begins. This pre-capture of various exposure conditions allows the subsequent processing to work with pre-prepared data, eliminating the need for sequential shooting and reducing overall inspection time while ensuring all necessary quality parameters are captured.
3Measurement precision
If multiple images with different exposure settings are captured, then measurement precision is improved, but the complexity of image processing increases
Solution Approach 1:
The patent applies local quality by processing different areas of the textile image with different exposure settings according to their specific requirements. Bright areas use underexposed images to prevent overexposure artifacts, while dark areas use overexposed images to ensure sufficient brightness for defect detection. This localized processing approach improves measurement precision for each specific area while reducing overall processing complexity by avoiding uniform processing of the entire image.
Data Source
Figure 1~2
Figure 3~5
Figure 6~9
AI summary
The present invention relates to a device, a computer program, and a method for the visual quality control of textiles, comprising the steps of: - capturing at least one image (100, 102) of a textile object (10), - selecting at least one sub-area (112, 116, 120) of the captured textile object in the image (100, 102) based on at least one predetermined selection criterion, - evaluating at least some image points in the selected sub-area (112, 116, 120) and determining an evaluation value for the image points based on at least one predetermined evaluation criterion, - comparing the evaluation value with a predetermined threshold value, or - calculating an averaged evaluation value for all evaluated image points and comparing the averaged evaluation value with the predetermined threshold value.