Camouflage Spots Analysis Based On Background

Camouflage is an important method in modern high-tech war, it can increase the value of military hardware and preserve itself. Front vague, battlefield complex and strong sudden are the character of the modern war. Single camouflage is difficult to meet the requirements of the different contexts in disguise. Traditional camouflage design method is vulnerable to man-made factors, the degree of integration of camouflage and background is low, and the design process is cumbersome, time-consuming. The paper will focus on the low accuracy and the cumbersome process of the traditional camouflage design, In order to improve the accuracy and real-time of the camouflage design.The main contents and results of this paper are as follows:(1) This paper presents an image segmentation method by combining MS algorithm and minimal spanning tree algorithm. In this paper we use the segmentation method to analysis the camouflage background. First, use the MS algorithm to quickly produce the over-segmentation regional, and then use the over-segmentation regional as a node to constructor the corresponding graph. Through the integration of the MS algorithm and the minimum spanning tree algorithm, we can reduce the over-segmentation of the MS algorithm can reduce the nodes and edges of the minimum spanning tree algorithm. Compared to MS algorithm、K-means algorithm and edge detection algorithm, my method can not only improve the accuracy of the background analysis, but also can improve the algorithm speed. My segmentation method can improve the real-time and the accuracy of the camouflage design.(2) Camouflage design not only requires camouflage patches should be similar to background patches, it also requires camouflage color and background color have a good match. In order to improve the similarity of the background color and camouflage color. In this paper, the color characteristic of background image is described by HSV color space with the visual characteristics of the human eye, the camouflage colors are selected from dominant colors of background, which are extracted by using specially quantized color histogram and applying threshold value. We analyzed the validity of the algorithm through experiments. The experimental results show that the quantitative color histogram can objectively evaluation of the target and its background color characteristics and the algorithm speed is significantly improved.(3) By analyzing intersection of the color characteristics and using of Manhattan distance we evaluated the camouflage effectiveness of the target color characteristics and texture similarity. The evaluation of this chapter can also be used to guide camouflage design.