A Textile Image Separation Algorithm Based on BP Network

At present, the textile image is separated by manpower with low efficiency. In order to deal with this problem, this paper proposes a novel algorithm for textile image separation based on back propagation artificial neural network. The factors that can affect the result of image separation are the RGB values of a pixel itself, the color of its adjacent pixels, and its boundary probability. According to these factors, we construct a model of BP network and train it with a group of samples that are made up of feature areas of a textile image with noise. After trained, the BP network can be used for textile image separation. Finally we designed an experiment- use the BP network to separate three textile images. The result of the experiment shows that the algorithm is simple, feasible, and can decrease the people´s work.