SLIC superpixel target tracking method based on uniform random

In order to improve the superpixel generation rate and boundary adhesion to achieve an effect superpixel target tracking, this paper proposed an improved SLIC superpixel algorithm based on uniform random for target tracking. Improved SLIC superpixel target tracking algorithm used uniform random method to extract SLIC superpixel characteristics, so that the redundant cluster center and algorithm complexity were reduced. Moreover, the improved algorithm achieved a more effective tracking with variable window, and it could deal with fuzzy images during tracking effectively. The experiments proved that the improved algorithm had a good performance in most well-known video sequences.