Applying Principal Component Analysis to Optimize Visual Effects and Information Communication in Visualization Designs

Visual communication design requires that feeling information and exchange of information must be conveyed efficiently and accurately.In this paper, we design a robust principal component sub-analysis visual enhancement algorithm based on improved Retinex.The algorithm transforms the image to the logarithmic domain so that it satisfies the decomposition condition of RPCA.After the RPCA decomposition model to get the low-rank component and sparse component, and will use adaptive gamma correction algorithm for the low-rank component for contrast enhancement, the two components are combined and then inverse transformed in the logarithmic domain to get the enhancement results.To avoid color distortion, the input image is converted to HSV color space to separate illumination information from noise.The model uses the inexact augmented Lagrange multiplier method (IALM) to solve the optimization problem, which leads to a significant improvement in the decomposition speed.The performance of the designed algorithm is verified on the dataset, and it is found that after the color equalization process for overexposed images, the gray value distribution is more uniform, and the image shows a better sense of brightness and visual effect after the contrast is increased.The algorithm scores 0.4648 and 0.7577 in UCIQE and UIQM respectively, which are ranked first among all algorithms and have better visual effect and information communication efficiency.