The document image segmentation is very useful for printing, faxing and data processing. An algorithm is developed for segmenting and classifying document image. Feature used for classification is based on the histogr...The document image segmentation is very useful for printing, faxing and data processing. An algorithm is developed for segmenting and classifying document image. Feature used for classification is based on the histogram distribution pattern of different image classes. The important attribute of the algorithm is using wavelet correlation image to enhance raw image's pattern, so the classification accuracy is improved. In this paper document image is divided into four types; background, photo, text and graph. Firstly, the document image background has been distingusished easily by former normally method;secondly, three image types will be distinguished by their typical histograms, in order to make histograms feature clearer, each resolution's HH wavelet subimage is used to add to the raw image at their resolution. At last, the photo, text and praph have been devided according to how the feature fit to the Laplacian distrbution by 2 and L . Simulations show that classification accuracy is significantly improved. The comparison with related shows that our algorithm provides both lower classification error rates and better visual results.展开更多
Segmenting a document image into text and picture areas is very important for compressing efficiently document images. This paper introduces an algorithm of multiscale image segmentation for document image compression...Segmenting a document image into text and picture areas is very important for compressing efficiently document images. This paper introduces an algorithm of multiscale image segmentation for document image compression,which uses of wavelet-domain hidden Markov tree model in order to directly calculate the parameter of model based on original image to be segmented,and to obtain multiscale classification and segmentation of image. The idea of the method is to combine several new technologies such as multilevel wavelet transform,multiscale decision,across-scale dependencies and joint probability density function. The paper describes in detail the concept of the dyadic block,the correspondency between wavelets and dyadic blocks based on quad-tree,the hidden Markov model and multiscale likelihood computation.展开更多
研究了一种基于 B 样条小波变换的自动阈值分割算法。提出基于灰度统计特性的直方图移动平均法,从而有效地消除噪声对图像的影响,使图像直方图更加光滑。运用 B 样条小波变换快速算法,大大减少了计算量。基于小波变换多分辨分析的策略,...研究了一种基于 B 样条小波变换的自动阈值分割算法。提出基于灰度统计特性的直方图移动平均法,从而有效地消除噪声对图像的影响,使图像直方图更加光滑。运用 B 样条小波变换快速算法,大大减少了计算量。基于小波变换多分辨分析的策略,提出一种多尺度小波变换分割方法,进一步提高了分割精度。通过对算法仿真研究,验证了本算法的可行性及有效性。展开更多
文摘The document image segmentation is very useful for printing, faxing and data processing. An algorithm is developed for segmenting and classifying document image. Feature used for classification is based on the histogram distribution pattern of different image classes. The important attribute of the algorithm is using wavelet correlation image to enhance raw image's pattern, so the classification accuracy is improved. In this paper document image is divided into four types; background, photo, text and graph. Firstly, the document image background has been distingusished easily by former normally method;secondly, three image types will be distinguished by their typical histograms, in order to make histograms feature clearer, each resolution's HH wavelet subimage is used to add to the raw image at their resolution. At last, the photo, text and praph have been devided according to how the feature fit to the Laplacian distrbution by 2 and L . Simulations show that classification accuracy is significantly improved. The comparison with related shows that our algorithm provides both lower classification error rates and better visual results.
文摘Segmenting a document image into text and picture areas is very important for compressing efficiently document images. This paper introduces an algorithm of multiscale image segmentation for document image compression,which uses of wavelet-domain hidden Markov tree model in order to directly calculate the parameter of model based on original image to be segmented,and to obtain multiscale classification and segmentation of image. The idea of the method is to combine several new technologies such as multilevel wavelet transform,multiscale decision,across-scale dependencies and joint probability density function. The paper describes in detail the concept of the dyadic block,the correspondency between wavelets and dyadic blocks based on quad-tree,the hidden Markov model and multiscale likelihood computation.
文摘研究了一种基于 B 样条小波变换的自动阈值分割算法。提出基于灰度统计特性的直方图移动平均法,从而有效地消除噪声对图像的影响,使图像直方图更加光滑。运用 B 样条小波变换快速算法,大大减少了计算量。基于小波变换多分辨分析的策略,提出一种多尺度小波变换分割方法,进一步提高了分割精度。通过对算法仿真研究,验证了本算法的可行性及有效性。
