Application of a new image segmentation method to detection of defects in castings

X-ray-based inspection technique is well applied to identification and evaluation of internal defects in castings, such as cracks, porosities, and foreign inclusions. Combining X-ray inspection with digital image processing and automatic image assessment is now the preferred approach for the continuous inspection of castings. However, in practical application, the quality of the X-ray image is poor. Under the circumstances, many classical thresholding methods usually cannot obtain ideal segmentation results. In this paper, we propose an effective segmentation method for the detection of typical internal defects in castings derived for an X-ray inspection system. The proposed method takes advantage of the fuzzy set theory and bound histogram and presents fuzzy exponential entropy for object and background according to the fuzzy sets and gray-level distribution of the image. The ideal threshold is obtained by maximizing the fuzzy exponential entropy associated with the distribution of the object and background classes in the bound histogram. Experimental results indicate that the proposed method is a powerful method to analyze images derived from X-ray inspection for automatically detecting typical internal defects in castings.