Automatic Optical Inspection and Classification System for Linear Guides

Linear guide is an important mechanical component used to bear loads and to guide linear movement simultaneously. In order to enhance its quality to create added value, its actual size must be strictly inspected to meet the requirement of accurate and quiet operation. The project aims to develop an automated optical inspection system for sliders of accurate linear guides to monitor and record their changes in actual sizes during manufacturing. The manufacturing of the slider can be roughly divided into three steps: (1) incoming material inspection, (2) heat treatment, and (3) grinding. The optical inspection system developed in this project will provide several key sizes for initial screening at the time of incoming inspection, to eliminate defective products, and classify qualified products for subsequent heat treatment. As the dimensions may vary after heat treatment, the system can be used again to filter out defective products and to classify and analyze the trend of tolerances. Based on the analysis, we may design the direction and extent in the following grinding procedure. In order to achieve a resolution of 0.01 mm and cover a measurement range of 80 mm, the project will use a precision X-Y table with positioning accuracy of ±1μm for accurate and repeatable positioning of the guide, together with an optical system with a resolution of 0.007 mm to meet the high precision and large detection range requirements at the same time. In addition to the integration of this mechatronic system, digital image processing plays the major role for automatic inspection, which includes the processing of white balance adjustments, contrast, saturation, sharpening, and de-noising in pre-processing, and processing and calculation to find center, edge, and distances. Further, the system will provide a shape recognition function to identify type of the slider under inspection for automatic interaction with database.

Project ID:PB10703-0458
External Project ID:MOST106-2622-E182-003-CC3