In capital goods industry, there are some components that are employed for safety purposes and, due to this fact, partsare subjected to high quality control demands. This is especially relevant for the case of safety components that containwelds because of the inherent complex process and the likelihood of defects appearance. In this context, this workpresents a machine vision system that was employed for replacing costly quality control procedures based on visualinspection. This was possible thanks to the proper design of all the machine vision system components including theimage processing algorithm. As a special feature of the system, it has to be highlighted the low cycle time of theproduction process (<2s), which stablished some requirements on the image processing algorithms. During theinspection system development, the main efforts were concentrated for obtaining a reliable and balanced database ofdefective and non-defective parts images useful to train the classification model. At this respect, the main contributionsconsisted of image analysis software development and visual curation of data. As a result, tailor made filters weredeveloped that allowed together with color information the identification of common flaws, as Lacks of Fusions (LoF).Due to the high amount of inspected samples, a preliminary deep learning based model was developed that includedthese filters with the aim of increasing defect detection accuracy.
展开▼
