小波變換邊緣檢測特性分析
THE FEATURE ANALYSIS OF IMAGE EDGE DETECTION WITH WAVELETS
-
摘要: 本文一改以往的以仿真的感性效果作為信號邊緣檢測質(zhì)量的效果評價方法,提出小波變換邊緣檢測定位精度和抗噪聲能力量化分析方法?；谛〔ㄗ儞Q的邊緣檢測算法,物理意義上是一個先平滑,再進行邊緣檢測的過程,其邊緣檢測特性與邊緣類型和尺度大小有關(guān)。隨尺度增大,定位偏差增大,反映了小波變換局部化特征強弱對邊緣檢測特性的影響。本文給出了不含噪聲和含有噪聲情況下,典型邊緣定位精度的量化表述。
-
關(guān)鍵詞:
- 小波變換; 邊緣檢測; 定位精度; 定位偏差
Abstract: A quantitative analysis on the local precision and the ability against noise for image edge detection with wavelets (IEDW) has been done in this paper. With an appropriate wavelet function, IEDW algorithm is equal with a process that consists of two child processes: denoising and edge detecting. The feature is relative to the type of edges and the scales. The influence of the scale is a morror of the local feature of wavelet transform. -
秦前清,揚中凱.實用小波分析.西安:西安電子科技大學(xué)出版社,1994, 80-92.[2]吳立德.計算機視覺.上海:上海復(fù)旦大學(xué)出版社,1992, 20-33.[3]Mallat S. A theory for multiresolution signal decomposition: The wavelet representation. IEEE Trans. on Pattern and Machine Intelligence, 1989, PAMI-11(7): 674-692.[4]Mallat S, Zhong S. Characterizaton of signals from multiscale edges. IEEE Trans. on Pattern and Machine Intelligence, 1992, PAMI-14(7): 710-731. -
計量
- 文章訪問數(shù): 2210
- HTML全文瀏覽量: 156
- PDF下載量: 466
- 被引次數(shù): 0
下載:
