Fast edge extraction algorithm based on H-LBP and its application in nuclear imaging

Nuclear imaging technology is one of the most widely used methods in nondestructive testing and extracting useful information from the images is one of the important tasks in nuclear imaging processing. H-LBP is an effective algorithm for edge extraction, however, the computational complexity and slowly computation speed restricts its feasibility in real applications. To cope with the above problems, a fast edge extraction algorithm based on H-LBP, namely FH-LBP, is proposed by optimizing the computing framework of H-LBP and adopting 1-normed sparse evaluation strategy to realize the global parameters adaptive selection. Experimental results show the new method possesses excellent performance and high robustness, which make it feasible and practical in real applications. (authors)