By introducing the AI image inspection system, we achieved a test result accuracy of 97%! After implementing it across all lines, the number of inspection workers was reduced by 70%.
The inspection system DEEP INSPECTION improves the speed and accuracy of the entire system and reduces the burden of visual inspections. Rist has developed a mirror surface inspection system using Deep Learning technology for the product inspection process of Murakami Kaimeido Co., Ltd., which boasts a high market share in rearview mirrors. This system has achieved an improvement in inspection accuracy (from 60% to 97%) and reduced the burden on inspectors performing visual checks. Rist's approach utilizes a convolutional network for multi-class classification. They developed a unique "Deep Inspection Confidence" and achieved a testing result accuracy of 97%. 【Issues with Conventional Detection Systems】 ■ Difficult to quantify image color tones, etc. ■ Ambiguity in standards due to human determination of thresholds ■ Inspection accuracy around 60% ■ Ultimately, inspectors confirm visually 【Features】 ■ After full-line implementation, the number of inspection workers is reduced by 70% ■ With an eye on overseas expansion, the aim is to optimize upstream processes based on data obtained from the automation of inspections. *For more details, please refer to the PDF document or feel free to contact us.
Inquire About This Product
basic information
For more details, please refer to the PDF document or feel free to contact us.
Price range
Delivery Time
Applications/Examples of results
For more details, please refer to the PDF document or feel free to contact us.
catalog（1）
Download All Catalogs
Company information
Rist Inc. is a company that develops and provides solutions and products using AI technology, primarily focused on the manufacturing industry, under the mission of "Bringing brilliance to Japan's future with AI." We possess the technical expertise to address the unique challenges of the manufacturing industry and develop custom solutions tailored to our clients' needs, including appearance inspection systems utilizing Deep Learning, data analysis, and predictive models. By mastering new technologies such as Deep Learning, which are being researched worldwide, we continue to challenge ourselves to solve the issues faced by our clients and society through the systematization of tasks that have traditionally been performed by humans and the accumulation and analysis of know-how.