Hon Hai Precision Industry has filed a patent for a method of detecting defects in product images. The method involves acquiring an original image and automatically detecting any apparent defects. If defects are found, the original image is cut into partial images centered on each defect. Each partial image is then analyzed to determine if it indicates a real defect or a false defect. Based on the results, the original image is classified as desired or undesired. The patent also includes a defect detection system that applies this method. GlobalData’s report on Hon Hai Precision Industry gives a 360-degree view of the company including its patenting strategy. Buy the report here.
Smarter leaders trust GlobalData
Access deeper industry intelligence
Experience unmatched clarity with a single platform that combines unique data, AI, and human expertise.
According to GlobalData’s company profile on Hon Hai Precision Industry, device power optimization was a key innovation area identified from patents. Hon Hai Precision Industry's grant share as of September 2023 was 58%. Grant share is based on the ratio of number of grants to total number of patents.
Method for detecting defects in images of products
A recently filed patent (Publication Number: US20230316497A1) describes a method and system for detecting defects in images. The system includes an image acquiring apparatus, an automatic optical detection apparatus, and a defect rechecking device. The method involves acquiring an original image using the image acquiring apparatus and automatically detecting the image to determine if it contains any defects. If defects are found, the original image is divided into partial images, each containing one defect. If no defects are found, the original image is considered a desired image. The method then determines whether each partial image is a desired image and outputs a result. Based on this result, the method determines whether the original image is a desired image.
The method further includes determining whether the original image is a desired image based on the result obtained from analyzing the partial images. If all the partial images are determined to be desired images, the original image is considered a desired image. However, if at least one of the partial images is determined to be a defective image, the original image is considered a defective image.
To determine whether each partial image is a desired image, the method utilizes a plurality of deep learning models. Each partial image is inputted into a corresponding deep learning model, which determines whether the image is a desired image based on the output result. Each deep learning model is configured for a different false defect feature, corresponding to a specified type of false defect feature.
The method also includes preprocessing the partial images for enlargement before determining whether they are desired images and outputting a result.
The patent also describes a defect detection system that implements the method. The system includes an image acquiring apparatus, an automatic optical detection apparatus, and a defect rechecking device. The image acquiring apparatus acquires the original image, and the automatic optical detection apparatus detects defects in the image. If defects are found, the defect rechecking device cuts the original image into partial images, each containing one defect. The defect rechecking device then determines whether each partial image is a desired image and outputs a result. Based on this result, the defect rechecking device determines whether the original image is a desired image.
The defect rechecking device utilizes a plurality of deep learning models to determine whether each partial image is a desired image. Each deep learning model is configured for a different false defect feature, corresponding to a specified type of false defect feature. The defect rechecking device inputs each partial image into the deep learning models and determines the desired image status based on the output result.
Overall, this patent presents a method and system for detecting defects in images using deep learning models and partial image analysis. The system aims to improve defect detection accuracy and efficiency.
To know more about GlobalData’s detailed insights on Hon Hai Precision Industry, buy the report here.
Data Insights
From
The gold standard of business intelligence.
Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.
GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.
GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.
