Orientated texture segmentation for detecting defects

Quality control activities form an essential information feedback loop for any business, with potential influence on the design, process planning, logistics functions and manufacturing. Visual inspection is an important part of quality control in the industries like textile and wood. In automated visual inspection (AVI), all feature vectors are calculated using standard algorithms such as Gabor filter, wavelet transform, morphological filter etc. In applications like defect detection, where directionality is the main concern, it is very necessary to concentrate on the dominant orientation of the regions. Orientation as a feature vector increases the computational speed, reliability as well as reduces the memory requirement hence feasible for real-time factory implementations. Hence the proposed algorithm is powerful for defect detection as compared to Gabor algorithm. Step and window size has to be properly tuned while calculating the orientation image. Number of clusters used for classification plays a vital role in final segmentation results.