Image Defect Detection with Mask-Guided Pixel Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing defect detection methods in industrial manufacturing require numerous manual operations and suffer from low detection accuracy due to subjective factors, leading to inefficiencies and high labor costs.
Innovation Solution
An image defect detection method utilizing mask image groups with complementary binary images and an image completion model to generate restored defect-free images, enabling accurate defect location at the pixel level without manual intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual operations are used for defect detection, then detection process can be performed, but detection accuracy is low and labor costs are high
Solution Approach 1:
The patent replaces manual mechanical inspection with an automated image processing system that uses mask image groups and completion models to detect defects, thereby eliminating subjective factors and improving detection accuracy while reducing labor requirements
Solution Approach 2:
The patent introduces mask image groups as an intermediary tool between the original image and defect detection results. These masks guide the completion model to focus on specific regions, enabling automated high-precision defect localization without manual intervention
2Measurement precision
If traditional defect detection methods are used, then detection can be performed, but location accuracy of defect area is insufficient
Solution Approach 1:
The patent segments the image processing task by using multiple mask image groups that divide the original image into different regions of interest. Each mask focuses the completion model on specific areas, enabling precise defect location while managing system complexity through modular processing
Solution Approach 2:
The patent adds a mask dimension to the traditional image processing approach. By superimposing mask image groups on the original image, the system creates a multi-dimensional processing space that enhances defect location accuracy without proportionally increasing overall system complexity
3Productivity
If manual defect detection is performed, then detection can be completed, but productivity is low due to numerous manual operations
Solution Approach 1:
The patent implements a self-service detection system where the completion model automatically processes images and identifies defects without requiring manual operations. The system serves itself by using the input image and mask groups to generate defect locations autonomously, dramatically improving productivity and reducing time loss
Solution Approach 2:
The patent performs preliminary action by pre-processing the original image with mask image groups before defect detection. This preparation step organizes the data structure and highlights potential defect regions, enabling the completion model to work more efficiently and reduce overall detection time
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
Provided are an image defect detection method and apparatus, an electronic device, a storage medium and a product, relating to the field of image processing technology and, in particular, to the field of defect detection technology. The solution includes acquiring a to-be-detected image; obtaining a restored image corresponding to the to-be-detected image based on the to-be-detected image, at least one mask image group and a plurality of defect-free positive sample images, where each mask image group includes at least two binary images having a complementary relationship, and different mask image groups have different image sizes; and locating a defect of the to-be-detected image based on the to-be-detected image and each restored image. The solution solves the problem in which a related defect detection method requires numerous manual operations and has a low detection accuracy due to subjective factors of a worker. With the solution, the defect of the image can be located at the pixel level, the detection accuracy is relatively high, and a large amount of labor cost can be saved.