Defect image generation method, device, electronic device and storage medium

By acquiring workpiece images, text prompt information and mask images, the defect image generation model is used to extract generation features from the workpiece images and text prompt information, and shape features from the mask image to generate a target defect image that meets the mask information constraints. This solves the problem in the existing technology that the defect generation model cannot accurately control the shape and position, and meets the actual needs of the defect image.

CN120198549BActive Publication Date: 2025-09-09SHENZHEN XINRUN FULIAN DIGITAL TECH CO LTD
1 Cites 0 Cited by

Patent Information

Application Number
CN202510683309.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-05-26
Publication Date
2025-09-09
Estimated Expiration
2045-05-26

AI Technical Summary

Technical Problem

The defect generation model in the existing technology cannot accurately control the shape and position of the defect, resulting in the generated defect image not meeting actual needs and cannot be used as an ideal negative sample.

Method used

By obtaining the workpiece image, text prompt information and mask image, and using the pre-trained defect image generation model, the generation features are extracted from the workpiece image and text prompt information, and the shape features are extracted from the mask image. Combined with the shape constraint sub-module and the stable diffusion sub-model, the target defect image that meets the mask information constraints is generated.

Benefits of technology

Precise control of defect shape and position is achieved, so that the generated defect image meets actual needs and can be used as an ideal negative sample.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The present application relates to a defect image generation method, device, electronic device, and storage medium. The method includes: obtaining a workpiece image, text prompt information, and a mask image, wherein the mask image is generated based on the workpiece image and the mask information, and the mask information is used to characterize the shape and position of the defect to be generated on the workpiece image; inputting the workpiece image, the text prompt information, and the mask image into a pre-trained defect image generation model to obtain a target defect image, wherein the defect image generation model is used to extract a first generation feature from the workpiece image and the text prompt information, extract a first shape feature from the mask image, and generate a target defect image based on the first generation feature and the first shape feature, wherein the defect generated in the target defect image satisfies the constraints of the mask information on the defect shape and defect position. In this way, precise control of the defect shape and defect position can be achieved, so that the target defect image meets actual needs.
Need to check novelty before this filing date? Find Prior Art