Papers by Keyword: Maximum Entropy

Paper TitlePage

Authors: H.Y. Chin, C.C. Ling, S. Fung, C.D. Beling
Authors: Gang Liu, Fang Li
Abstract: To solve the difficulty of equipment testability verification experiment sample selection, a sample selection method of testability verification experiment was presented. The failure-test dependency matrix and its extended matrix, failure modes function equivalence set and failure modes test equivalence set were analyzed, on the basis, failure modes sample equivalence set was proposed. Meanwhile, importance characteristic was analyzed and maximum entropy was solved, and the sample selection method was presented, its process was obtained. Moreover, the method was comparatively analyzed with actual methods in the experiment. The results show that application the method in paper, can reduce sample number, save test expenses, and meet the test requirement of adequacy, so the method in paper is proved effective and feasible.
Authors: Hui He, Bo Chen
Abstract: Recently, sentiment analysis of text is becoming a hotspot in the study of natural language processing, which has drawn interesting attention due to its research value and extensive applications. This paper introduces a smart sentiment analysis system, which is to satisfy three aspects of sentiment analysis requirement. These are Chinese sentiment word recognition and analysis, sentiment related element extraction and text orientation analysis. Promising results and analysis are presented at the end of this paper.
Authors: Bao Shu Li, Wen Li Wei, Ke Bin Cui, Xue Tao Xu
Abstract: According to the limitations of the shooting environment, captured image exist the phenomenon of image blurring and noise. This paper proposes that the improved maximum entropy method recovery blurred image which acquire in aerial. Finally, according to the first order Markoff theory to evaluate the quality of the processed image, the results show that maximum entropy image restoration method compared to the conventional approach increase image clarity and details more better.
Authors: Arun Shukla, L. Hoffmann, Alfred A. Manuel, M. Peter
Authors: Ming Gang Du, Shan Wen Zhang
Abstract: Crop disease leaf image segmentation is a key step in crop disease recognition. In the paper, a segmentation method of crop disease leaf image is proposed to segment leaf image with non-uniform illumination based on maximum entropy and genetic algorithm (GA). The information entropy is regarded as the fitness function of GA, the maximum entropy as convergence criterion of GA. After genetic operation, the optimal threshold is obtained to segment the image of disease leaf. The experimental results of the maize disease leaf image show that the proposed method can select the threshold automatically and efficiently, and has an advantage over the other three algorithms, and also can reserve the main spot features of the original disease leaf image.
Authors: Cheng Jian Sun, Song Hao Zhu, Zhe Shi
Abstract: This paper proposes a novel multi-view learning framework which leverages the information contained in pseudo-labeled images to improve the prediction performance of image classification using multiple views of an image. In the training process, labeled images are first adopted to train view-specific classifiers independently using uncorrelated and sufficient views, and each view-specific classifier is then iteratively re-trained using initial labeled samples and additional pseudo-labeled samples based on a measure of confidence. In the classification process, the maximum entropy principle is utilized to assign appropriate category label to each unlabeled image using optimally trained view-specific classifiers. Experimental results on a general-purpose image database demonstrate the effectiveness and efficiency of the proposed multi-view semi-supervised scheme.
Authors: Stephen B. Dugdale, M.A. Alam, H.M. Fretwell, M. Biasini, D.A. Wilson
Authors: Arun Shukla, L. Hoffmann, Alfred A. Manuel, M. Peter
Showing 1 to 10 of 22 Paper Titles