Detection of Pits by Conjugate Lines: An Algorithm for Segmentation of Overlapping and Adhesion Targets in DE-XRT Sorting Images of Coal and Gangue

In lump coal and gangue separation based on photoelectric technology, the prerequisite of using a dual-energy X-ray to locate identify is obtain independent target area. However, with increase in input sorting system, actual collected images had adhesion overlapping targets. This paper proposes pit point detection segmentation algorithm solve problem The forms are divided into open closed-loop (OLA CLA). Then, an open- crossing (OLCA CLCA) proposed. We used conjugate lines detect judge position distance pixel relative lines. we set constraint relatively straight line complete detection. Finally, minimum search method was dividing corresponding image segmentation. experiment results demonstrate that accuracy 90.73%, acceptable 94.15%.