A Shadow Detection Method Based on SLICO Superpixel Segmentation

Abstract

This paper proposes a shadow detection algorithm based on SLICO superpixel segmentation to address the issues of single-image shadow detection. Firstly, SLICO superpixels is used to segment the shadow image to generate superpixel blocks to detect the shadow contour. Then proposes a fusion-characteristics-based SVM classifiers, the superpixel blocks are classified and merged to detect the shadow areas. Finally, the proposed algorithm is compared with Otsu threshold method and traditional SVM detection method. The experiment results verified the effectiveness of the proposed algorithm. The detection performances comparison of SSIM and PSNR indicates that the proposed algorithm obtains relative higher performances than the reference algorithms.