Ma Lizhuang

A Modified Laplacian Smoothing Approach with Mesh Saliency

A good saliency map captures the locally sharp features effectively. So a number of tasks in graphics can benefit from a computational model of mesh saliency. Motivated by the conception of Lee’s mesh saliency [12] and its successful application to mesh simplification and viewpoint selection, we modified Laplacian smoothing operator with mesh saliency. Unlike the classical Laplacian smoothing, where every new vertex of the mesh is moved to the barycenter of its neighbors, we set every new vertex position to be the linear interpolation between its primary position and the barycenter of its neighbors. We have shown how incorporating mesh saliency with Laplacian operator can effectively preserve most sharp features while denoising the noisy model. Details of our modified Laplacian smoothing algorithm are discussed along with the test results in this paper.

Correlation between Child-Pugh Degree and the Four Examinations of Traditional Chinese Medicine (TCM) with Liver Cirrhosis

Finding the relationship between the objective factors of Western Medicine and the signs or the symptoms of Traditional Chinese Medicine (TCM) is an important topic. Because the factors of Western Medicine can be evaluated objectively and the treatment way of TCM is more effectively to some diseases. In this paper, C4.5 algorithm, one of the Decision Tree, is used to analyze the relationship between Child-Pugh degree and four examinations of TCM based on liver cirrhosis. We hope to look for a method of Integrating Traditional Chinese and Western Medicine. When using our dataset, some rules are obtained and the accuracy of the grade of Child-Pugh classification is 85.67%, 64.97%, 77.94% according to Child-Pugh A, Child-Pugh B and Child-Pugh C. To test if the method is effective, we compare with other algorithms such as: Logistic, BayesNet, RBF NN, ID3 algorithm. It is demonstrated that C4.5 outperform ID3, Logistic, BayesNet, and RBF NN.Finding the relationship between the objective factors of Western Medicine and the signs or the symptoms of Traditional Chinese Medicine (TCM) is an important topic. Because the factors of Western Medicine can be evaluated objectively and the treatment way of TCM is more effectively to some diseases. In this paper, C4.5 algorithm, one of the Decision Tree, is used to analyze the relationship between Child-Pugh degree and four examinations of TCM based on liver cirrhosis. We hope to look for a method of Integrating Traditional Chinese and Western Medicine. When using our dataset, some rules are obtained and the accuracy of the grade of Child-Pugh classification is 85.67%, 64.97%, 77.94% according to Child-Pugh A, Child-Pugh B and Child-Pugh C. To test if the method is effective, we compare with other algorithms such as: Logistic, BayesNet, RBF NN, ID3 algorithm. It is demonstrated that C4.5 outperform ID3, Logistic, BayesNet, and RBF NN.

An automatic image and video colorization algorithm based on pattern continuity

This paper proposes a novel method to colorize grayscale images from a reference image. Rather than depending on the comparisons between the grayscale image and the reference image based on the luminance value and neighborhood statistics of each pixel, our method consider pattern continuity and spatial consistency. We handle the pixels in the grayscale image from a higher level and gain the global results, which are better than the results based on the local statistical information. Another advantage of our method is we need less human intervention than the previous methods. This is very beneficial for the video colorization. Our method is focus on colorizing cartoon images and videos. With regard to the cartoon images, using texture analysis based on pattern continuity can gain a better result. For cartoon videos, our method generates scribbles automatically instead of drawing by the user. Once a representative frame has been colorized, all the other frames will be colorized sequentially, which makes a great improvement on automatic colorization.

Decision tree method to extract syndrome differentiation rules of posthepatitic cirrhosis in traditional Chinese medicine

Syndrome differentiation is an important topic in traditional Chinese medicine (TCM).Decision tree, one of the data mining algorithms developed, is a method to induce rules from data. In this paper, decision tree is applied to extract syndrome differentiation rules from 293 cases related to liver and kidney yin deficiency, damp-heat smoldering and Stasis and heat smoldering syndrome. Thus the decision tree classification model is obtained and some important factors are selected to three mainly syndromes of posthepatitic cirrhosis; corresponding syndrome differentiation rules are induced from the model. The classification accuracies are 79.86%, 80.5% and 82% respectively. The experiment results show that the decision method is likely a promising method to extract diagnostic rules from patient records of Chinese medicine and could be expected to be useful in the practice of traditional Chinese medicine.

Pattern recognition based on weighted and supervised ART2

It is crucial for TCM (traditional chinese medicine) post-hepatitis cirrhosis diagnosis to accurately identify the syndrome. Meanwhile, the selection of features which are relevant to a certain TCM post-hepatitis cirrhosis syndrome not only improves the performance of the classifiers, but also provides well measure for treatment. Therefore, in this paper, we analyze the classical ART2(adaptive resonance theory 2) neural network, such as the problem of pattern drifting and the same phase data with different amplitudes. Based on this, here, a novel network named SWART2 is proposed by taking dispersion testing and centroid computation learning, and introducing the weighted and supervised mechanism, which aims at improving ART2¿s ability of classification greatly for post-hepatitis cirrhosis diagnosis. Experimental results in this paper showed that the new SWART2 performed better than classical ART2.

Feature-preserving mesh denoising based on contextual discontinuities

Motivated by the conception of Lee et al. (2005)’s mesh saliency and Chen (2005)’s contextual discontinuities, a novel adaptive smoothing approach is proposed for noise removal and feature preservation. Mesh saliency is employed as a multiscale measure to detect contextual discontinuity for feature preserving and control of the smoothing speed. The proposed method is similar to the bilateral filter method. Comparative results demonstrate the simplicity and efficiency of the presented method, which makes it an excellent solution for smoothing 3D noisy meshes.

Reconstruction from contour lines based on bi-cubic Bézier spline surface

A novel reconstruction method from contours lines is provided. First, we use a simple method to get rid of redundant points on every contour, then we interpolate them by using cubic Bézier spline curve. For corresponding points of different contours, we interpolate them by the cubic Bézier spline curve too, so the whole surface can be reconstructed by the bi-cubic Bézier spline surface. The reconstructed surface is smooth because every Bézier surface is patched with G2 continuity, the reconstruction speed is fast because we can use the forward elimination and backward substitution method to solve the system of tridiagonal equations. We give some reconstruction examples at the end of this paper. Experiments showed that our method is applicable and effective.

Research on the Relationship Between Symptoms of TCM and Stage, Class, Child-Pugh Grade About Liver Cirrhosis

While studying the relationship between the signs, symptoms of TCM and Child-Pugh grade, stage, class based on liver cirrhosis, the number of attribute is important. We use the information gain method to assess the attributes and use 5 classifiers such as logistic, BayesNet, NaiveBayes, RBF and C4.5 to show how the number of attribute effect the classify accuracy. We can see there are some rules between them. In this paper we give the main symptoms and signs of TCM relating to the stage, class and Child-Pugh grade about liver cirrhosis.

Matting and Compositing in Radiance Space

Digital matting is one of the most important techniques in the film and video industry. The illustration model of the foreground object is extracted from photographs or videos and can be combined with new background image. Almost all of the existing digital matting and compositing algorithms are computed in RGB color space. However, because of the nonlinearity of the imaging system, the relationship between the color value, sampled by the imaging device, and the correspondent radiance value is approximately logarithmic. The matting equation described in color space leads to an unfaithful physical model of the matting problem. It makes errors in extracting and compositing the matte data of reflection or refraction surfaces. A matting and compositing method in radiance space is presented in this paper. With recovering the Hurter-Driffield curve of the imaging system, we introduce the matting problem into radiance space. We compare the results of triangulation matting and compositing in color and radiance space. It comes out that the matting data can be calculated more accurately in radiance space