迷彩由於成效快、成本低、製作與施工方便等優勢，廣泛運用到軍事偽裝，而迷彩的最主要目的為降低在戰場被敵人偵測的機率以提高己方的生存率。本文主要目的乃在探討利用邊緣視覺特性對迷彩效能評估提出系統化合理評價迷彩效能的測試方法，本文並提出使用Q 數值計算法區分不同迷彩與背景融合度，並創造90%、 85%、80%、70%、60 %等五組不同融合度之迷彩放置於草地自然場景中，迷彩隨機分佈於-21°、-14°、-7°、0°、7°、14°、21°等7 個水平位置點，並利用眼動儀全程監控受測者的搜尋認知歷程，從實驗結果得出的正確率與偵測時間之數據分析，可彌補以往利用數值分析法與主觀評量法並無法得出之人眼辨識迷彩的量化數據，並從各種不同視角間迷彩效果之變化，發現較佳迷彩具有使邊緣視覺偵測視角縮減與偵測時間上昇效果。

This study presents an experimental framework for evaluating metrics at the searching and discrimination of a camouflage target from its background. Such metrics could provide a method to identify camouflage effect and underlying models of searching and discrimination.A computer aided Q index was adopted to classify and quantify the level of differences. Six student volunteers from military school participated in the study, and they were asked to view image stimuli consisting of various camouflage patterns embedded within background patterns.These psychophysical experiments provided a quantitative basis for comparison of human judgments to the computed values of target distinctness metrics. Five levels of similarities with Q index of 90 %, 85 %, 80 %, 70 % and 60 %were tested. Targets with camouflages were presented in participants’ visual filed with viewing angles of -21°, -14°, -7°, 0°, 7°, 14°, and
21°. The observers were asked to search in each image scene and identify suspected target locations. Detection time, hit rate, rates of subjective confidence, and rates of task difficulties were collected to justify the quality of the assessment system and the design of camouflages.Overall, the set of target distinctness metrics considered, a metric based on a model of Q index was the most strongly correlated with the psychophysical data. The result showed that all performance measures allowed us to distinguish good performing camouflage from poorly performing. Camouflage performance worse with increased viewing angles occurred only when the background similarities were greater than 85%, hit rates remained higher when the background similarities were 80%. Based on the result of the study, support by using both central and/or peripheral vision, good camouflage performance occurred when the background similarities greater than 85%. The present findings provide scientific evidence and quantifying
methodology that camouflage design can be based.