In this paper emphasis is put on the application of the wavelet technique into the system and the local segmentation problem preliminarily solved existing in the extant bone - age recognition system . . the traditional chain - length concept is dixcarded and the percentage concept of the chain is proposed . no related reports can be found in the literature . the chain is proposed 隨后,筆者重點研究了小波技術在微機骨齡診斷系統(tǒng)中的應用,初步解決了目前骨齡系統(tǒng)中存在的局部分割問題,拋棄傳統(tǒng)的鏈長觀念,提出了鏈的百分比概念,國內(nèi)外文獻未見有相關報道;針對分割結果的不封閉性,采用邊緣點生長的方法,有效地解決了區(qū)域的封閉性。
Applied in license plate segmentation problem , a new segmentation method of automobile license plate based on wavelet transform and neural network is pointed out [ 71 ] . 2 ) phase of image feature extraction : combined with the feature extraction of structural and statistical method , a method of image character feature extraction based on wavelet and moments analysis is presented [ 74j . 3 ) phase of image classificaton [ 73 ] : after investigation on intelligence recognition technology , the paper puts forward basic structure of recognition machine ' s model , and makes a primary research of basic structure and design method , then makes research of the multi - character method 并應用于車牌分割問題,提出基于小波與神經(jīng)元模式識別的車牌圖像分割方法; 2 )特征提取階段:將結構特征提取方法和統(tǒng)計特征提取方法的緊密有機結合,提出一種基于小波和矩的車牌圖像字符特征向量提取方法; 3 )分類識別階段:對智能識別技術進行研究,提出智能識別機的模型結構,對識別機的基本層次結構和設計方法進行初探;并針對多特征方法進行一定的研究;本文提出的基于模式識別的圖像處理方法對其他領域的圖像處理具有一定的參考價值。
A novel method of content - based image segmentation using deformable template matching is proposed . a two - dimensional ( 2 - d ) deformable template based on orthogonal curves is built by pre - computing extensions of the deformable template along orthogonal curves and sampling the curves uniformly . then the definitions of internal and external energy functions are given according to the image segmentation problem , and genetic algorithm is used to obtain globally optimal solutions . the proposed method uses a lower - dimensional search space than conventional methods and reduces the sensitivity of the algorithm to initial placement of the template . experiments on real - world images and in simulations at low signal - to - noise ratio show the robustness and good performance of the method 本文提出一種采用可變形模板匹配技術進行基于內(nèi)容的圖像分割算法.通過預先計算出可變形模板沿著變形的正交曲線,并對模板曲線及正交曲線進行離散抽樣,建立一基于正交曲線的二維( 2 - d )可變形模板,針對圖像分割問題定義控制可變形模板進行變形的內(nèi)、外部能量函數(shù),本文采用遺傳算法搜索能量函數(shù)最小的全局最優(yōu)解.該新算法比傳統(tǒng)的可變形模板匹配方法降低了搜索空間的維數(shù),減少了算法對模板初始位置的敏感.對實際圖像及模擬低信噪比圖像處理的結果表明,新算法具有良好的分割精度及穩(wěn)定性