- Research Name Vision System for Automated Compliance Evaluation of Production Part Features
- Client ATCO Industries, Inc.
- Category PRACTICUM PROJECTS
- RESEARCH YEAR 2019
ATCO Industries’ primary business is quality inspection and containment services. A major solution they deliver to customers is quality sorting services.
The goal of this project was to create a model that will identify if a picture of the reject part has the part in the correct orientation matching the template good picture(s) generated by the PC (ATCO employee) to a certain degree. The matching criteria was determined by the key points and orientation of the part in the image.
The main scope of the project was for the model to be quickly and easily updatable to be used with new/future parts ATCO will inspect and return real time feedback if the picture/image capture is a good picture. In order to accomplish this, the feature matching method, BRISK along with brute force, was used for the model implementation.
Matthew Drouillard, Anthony Leo, Hannah Mcquade, and Shravani Mandalaparty