计算机工程与应用
計算機工程與應用
계산궤공정여응용
COMPUTER ENGINEERING AND APPLICATIONS
2015年
10期
170-173,256
,共5页
能见度%shearlet变换%图像边缘%sift图像匹配
能見度%shearlet變換%圖像邊緣%sift圖像匹配
능견도%shearlet변환%도상변연%sift도상필배
visibility%shearlet transform%image edge%sift image matching
针对现有的基于图像边缘的能见度计算方法效果不理想的问题,提出一种改进的基于图像边缘检测求能见度的方法。相对于原有的方法,该方法首先选取目标图像并用sift算法对其进行模板匹配,然后利用离散Shearlet变换提取边缘特征,再将这些特征和前向散射式能见度仪测量的数据进行线性拟合,建立它们之间的关系式,从而求出能见度。实验结果表明,该算法相对于sobel算法和小波变换算法,相对误差减少了2.761%和1.607%,计算的准确性和有效性明显得到改善。
針對現有的基于圖像邊緣的能見度計算方法效果不理想的問題,提齣一種改進的基于圖像邊緣檢測求能見度的方法。相對于原有的方法,該方法首先選取目標圖像併用sift算法對其進行模闆匹配,然後利用離散Shearlet變換提取邊緣特徵,再將這些特徵和前嚮散射式能見度儀測量的數據進行線性擬閤,建立它們之間的關繫式,從而求齣能見度。實驗結果錶明,該算法相對于sobel算法和小波變換算法,相對誤差減少瞭2.761%和1.607%,計算的準確性和有效性明顯得到改善。
침대현유적기우도상변연적능견도계산방법효과불이상적문제,제출일충개진적기우도상변연검측구능견도적방법。상대우원유적방법,해방법수선선취목표도상병용sift산법대기진행모판필배,연후이용리산Shearlet변환제취변연특정,재장저사특정화전향산사식능견도의측량적수거진행선성의합,건립타문지간적관계식,종이구출능견도。실험결과표명,해산법상대우sobel산법화소파변환산법,상대오차감소료2.761%화1.607%,계산적준학성화유효성명현득도개선。
According to the existing method for image edge visibility, the result is not ideal, this paper presents an improved method based on image edge detection for visibility. Firstly, it selects the target image using sift algorithm for template matching, then the discrete shearlet transform is used to extract the edge features, and then these features and the results of forward-scattering visibility meter are related. After establishing the relationship, visibility can be computed by the features with the built relationship. The experimental results show that the relative error is reduced by 2.761% relative to Sobel algorithm, and by 1.607%relative to wavelet transform algorithm. The accuracy and efficiency are obviously improved.