Design of online classifier for surface defect detection and classification of cold rolled steel coil

摘要: The target to be achieved through this project was primarily aimed at detecting the surface defects belonging different classes in cold rolled steel coils. This grabbing images from camera, here line scan camera is used which grabs 20 frames per second. Carrying out defect detection on these and later classifying them. We present a method automatically detect localize occurring surface. Defect regions are segmented background using their distinguishing texture characteristics. locates candidate directly DCT (Discrete cosine transform) domain intensity variation information encoded coefficients. More precisely, employs analysis of each individual non-overlapping region image determine potentially defective blocks, further grown merged form image. In thesis computer vision based, framework for classification strips implemented. have designed online classifier automatic defects. we measured statistical textural features gray level co-occurrence matrix presented by Haralick geometrical also calculated. final decision SVM (Support Vector Machine) handles problem types. proposed voting strategy that multiple outputs given input with specific type. addition, approach improves performance. Experimental results demonstrate effectiveness classification. An viewer application decoding information.