RETRACTED: Camouflage texture design based on its camouflage performance evaluation

Abstract

Along with the development of the technique in camouflage design, camouflage design and evaluation methods have attracted much attention by researchers. Most of traditional methods of camouflage texture effect evaluation are subjective method. It is very tedious and inconvenient to evaluate the camouflage pattern using the artificial method. In this paper, a systemic and rational method for camouflage texture design based on camouflage evaluation feedback is proposed. In our model, two different types of features was selected to make a comprehensive evaluation on a given digital camouflage. First, we applied visual saliency technique to evaluate camouflage performance. Digital camouflage itself has some special properties, such as camouflage histogram feature, rectangular feature and mosaic feature, which can well describe the differences between the camouflaged targets and the background. Then, the two kinds of features are fused to evaluate the camouflage performance using a nonlinear fusion function. Experimental results show that our method accords with the human visual perception mechanism, and is helpful for evaluation and design of the camouflage texture. The proposed approach can not only calculate the camouflage performance, but also optimize the process of digital camouflage design as a feedback mechanism.