邊緣檢測(cè)器性能的客觀定量評(píng)價(jià)
OBJECTIVE AND QUANTITATIVE EVALUATION OF EDGE DETECTORS
-
摘要: 本文提出了一種新的邊緣檢測(cè)器的客觀評(píng)價(jià)方法。該方法既能反映邊緣檢測(cè)器對(duì)圖像輪廓的檢測(cè)性能,又能對(duì)圖像細(xì)節(jié)處的邊緣檢測(cè)能力進(jìn)行正確的評(píng)價(jià)。由于該方法將整個(gè)測(cè)試圖區(qū)分成三個(gè)不同性質(zhì)的區(qū)域,不同的區(qū)域用不同的參數(shù)來反映邊緣檢測(cè)器的不同性能。因此,能準(zhǔn)確反映噪聲對(duì)邊緣檢測(cè)器的影響。同時(shí),它將整個(gè)邊緣檢測(cè)器的性能分解成邊緣提取能力和噪聲抑制能力兩部分,便于我們?nèi)?、?zhǔn)確和客觀地評(píng)價(jià)不同特性的邊緣檢測(cè)器,以適應(yīng)不同應(yīng)用場(chǎng)合的需要。
-
關(guān)鍵詞:
- 邊緣檢測(cè)器; 客觀定量評(píng)價(jià)法
Abstract: This paper presents a new method to evaluate edge detectors objectively and quantitatively. This method is capable of showing correctly the performance of detecting the contour of images, as well as of detecting the fine details of images. The effects of noises on edge detectors can be accurately reported by partitioning the whole test image into three areas having different characteristics and then, at different areas, using different parameters to match the different performance of edge detectors. At the same time, the whole performance of edge detectors is divided into two parts: the edge-extracting ability and the noise-attenuating ability. So, the edge detectors, having different characteristics, can be evaluated comprehensively, accurately and objectively, and be applied to different cases. -
Jain A K. Fundamentals of Digital Image Processing. Englewood, Cliffs, NJ: Prentice Hall, 1989.[2]Yoo J, Bouman C A, Delp, E J, et al. The nonlinear prefiltering and difference of estimates approaches to edge detection: Applications of stack filters, CVGIP, 1993, 55(2): 140-159.[3]Lee J S J, Haralick R M, Shapiro L G. Morphological edge detection, IEEE Trans. on Robotics Automat, 1987, RA-3(2): 142-156.[4]Abdou I E, Pratt W K. (quantitative design and evaluation of enhancement/thresholding edge detectors, Proc[J].IEEE.1979, 67(5):753-763[5]Fram J R, Deutsch E S. On the quantitative evaluation of edge detection schemes and their comparison with human performance, IEEE Trans. on Computers, 1975, C-24(6): 616-627.[6]Kitchen, L, Rosenfeld A. Edge evaluation using local edge coherence, IEEE Trans. on SMC, 1981, SMC-11(9): 597-605.[7]Li Xiangji(李向吉), Ding Runtao(丁潤(rùn)濤).Fuzzy morphological operators to edge enhancing of[8]images, Proc. of International Conference on Signal Processing, ICSP'98, Beijing: 1998, 1017-1020.[9]Li Xiangji(李向吉),Ding Runtao. Morphological edge enhancing filtering of images, in Inter-[10]national Symposium on Multispectral Image Processing, Proceedings of SPIE, vo1.3545, 1998, 262-265.[11]Arce G R, Foster R E. Detail-preserving ranked-order based filters for image processing, IEEE[12]Trans. on ASSP, 1989, ASSP-37(1): 83-98.[13]李向吉, 丁潤(rùn)濤. 脈沖噪聲圖像中的數(shù)學(xué)形態(tài)邊緣檢測(cè)器. 中國(guó)圖形圖像學(xué)報(bào), 1998, 3(11): 903-906. -
計(jì)量
- 文章訪問數(shù): 2123
- HTML全文瀏覽量: 99
- PDF下載量: 478
- 被引次數(shù): 0
下載:
