Research On Image Segmentation Of Multiphoton Microscopic Cell Image Based On SLIC Superpixels Method

Multiphoton microscopy (MPM) imaging is a technique which based on two-photon excited fluorescence (TPEF) and second harmonic generation (SHG). MPM has been widely utilized for imaging structure in biology and medical because of its advantages of enhanced imaging penetration depth, reduced photo bleaching and photo toxicity as well as high resolution. However, due to the low signal-to-noise ratio (SNR), blur boundaries and uneven background of MPM images, none of the existing methods can achieve exact segmentation results for diagnosis of histiocyte. Thus the automatic segmentation based on MPM images is still a challenging issue.In this paper, the problems of cells MPM images segmentation were discussed, and a novel multiphoton microscopy images segmentation method based on superpixels and watershed (MSW) is presented. The proposed method uses simple linear iterative clustering (SLIC) superpixels instead of pixels to analyze MPM images for the first time. The superpixels segmentation based on a new distance metric combined with spatial, CIE Lab color space and phase congruency features, divides the images into patches which keep the details of the cell boundaries. Then the superpixels are used to reconstruct new images by defining an average value of superpixels as image pixels intensity level. Finally, the marker-controlled watershed is utilized to segment the cell boundaries from the reconstructed images. Experimental results show that cellular boundaries can be extracted from MPM images by MSW with higher accuracy and robustness.