Color Compensation Method of 2-D Clothing Image Based on Visual Communication from the Perspective of Internet of Things

Abstract

Abstract From the perspective of the Internet of Things, market competition is becoming increasingly fierce, and Internet of Things technology can provide more possibilities for clothing brands’ advertising and marketing strategies, further shaping brand image. However, the color of 2-D clothing images is imbalanced. Therefore, when using the current method for color compensation processing of 2-D clothing images, it is prone to interference from issues such as lighting shadows, uneven brightness, and angle changes. The structural similarity of 2-D clothing images is low, the peak signal-to-noise ratio (PSNR) is low, the standard deviation is low, and the visual effect is poor. The research goal is to solve this problem, and a color compensation method of 2-D clothing image based on visual communication from the perspective of the Internet of Things is proposed. Firstly, the 2-D clothing image is segmented by histogram of oriented gradients feature and exemplar support vector machine classifier, and then the dimension of the image is reduced by weighted subspace probabilistic clustering analysis(WSPCA) algorithm and nonlinear algorithm, and the noise in the image is eliminated by homomorphic filtering method. Finally, the preprocessed image is input into a Gaussian homomorphic filter to complete the color compensation of the 2-D clothing image. Analyzing the experimental results, it can be seen that the proposed algorithm has high structural similarity, high PSNR, high standard deviation, and good visual effect.