Optical Design for Automatic Inspection System of Defects on Disposable Surgical Scalpel Blades Using Machine Vision

Nowadays, many products in the medical area are disposable, such as needles, daily contact lenses, disposable ultra-thin surgical scalpel blades, infusion tubes. With the increasing demand in the market, manufacturers are striving to enhance production output while maintaining acceptable quality standards. Consequently, the implementation of an automatic inspection system for these products has become imperative. The main challenge with automatic quality inspection systems is the input data, consisting mainly of images captured during production. Image processing algorithms or deep learning models are used to analyze these images, classifying the products as “pass” or “fail” based on the inspection results. To minimize inspection time, high-quality input images are crucial, clearly distinguishing defects from the background. The optical system used for inspection must be carefully designed, considering the unique characteristics of each product, such as surface properties and external features that may pose challenges. In this article, we present a research study in which an optical system is designed using Dark Field, Coaxial and Dome lighting techniques to highlight defects on disposable surgical scalpel blades with metallic reflection surface. The Dark Field technique effectively reveals missing portions, cracks, or fractures on the cutting edge, while Coaxial lighting detects faults on the blade body. The Dome lighting technique provides illumination, preventing reflections on the metallic surface. We employed the Dome technique to inspect the quality of printed details on the aluminum foil sheet of the sealed blade’s pocket. The resulting images enable efficient defect detection due to their well-highlighted nature.