iNEMI Call for Participation: Development of an Imaging AOI + AI Ecosystem Project
Development of an Imaging AOI + AI Ecosystem
September 1 & 2, 2020
iNEMI’s Development of an Imaging AOI + AI Ecosystem project will address the electronics manufacturing industry’s need for comprehensive image libraries (data sets) to train AI models for AOI and AXI. Insufficient data can result in inaccurate AI models for image classification. Presently, companies such as electronic manufacturing services (EMS) providers maintain their own image data, which is private and isolated. The collection of image data — data acquisition, identification, labeling, etc. — is time consuming and there is currently no collaboration model in the industry for sharing image data that can balance benefits with risk mitigation.
This project will evaluate and propose an ecosystem collaboration model for the electronics manufacturing industry to collect and share image data from collaborating companies in a commonly acceptable protocol. The data sets will be contributed to a central data repository where data is consolidated into different categories/types and can provide valuable input for AI training cycles. The initial focus of the project will be on the EMS sector, which could benefit quickly from leveraging more data to enhance their AOI capabilities.
The project team will investigate technical, business and security issues with the aim of developing an acceptable protocol and business model that will allow for the appropriate level of sharing of images to enhance manufacturing capability without impinging on intellectual property or confidentiality.
Registration
Join us for this call-for-participation webinar to understand how the Development of an Imaging AOI + AI Ecosystem project can benefit your organization. Two sessions are scheduled (with the same content) and the webinar is open to industry. Click on a registration link (below) to register. For additional information, contact Mark Schaffer (marks@inemi.org).
Session 1: Thursday, September 1
10:00-11:00 a.m. EDT (U.S)
4:00-5:00 p.m. CEST (Europe)
10:00-11:00 p.m. CST (China)
Session 2: Thursday, September 2
9:00-10:00 a.m. CST (China)
9:00-10:00 p.m. EDT (U.S.) on September 1
