This paper proposes support vector machine (SVM) based voice activity detection using FuzzyEn to improve detection performance under noisy conditions. The proposed voice activity detection (VAD) uses fuzzy entropy (FuzzyEn) as a feature extracted from noise-reduced speech signals to train an SVM model for speech/non-speech classification. The proposed VAD method was tested by conducting various experiments by adding real background noises of different signal-to-noise ratios (SNR) ranging from −10 dB to 10 dB to actual speech signals collected from the TIMIT database. The analysis proves that FuzzyEn feature shows better results in discriminating noise and corrupted noisy speech. The efficacy of the SVM classifier was validated using 10-fold cross validation. Furthermore, the results obtained by the proposed method was compared with those of previous standardized VAD algorithms as well as recently developed methods. Performance comparison suggests that the proposed method is proven to be more efficient in detecting speech under various noisy environments with an accuracy of 93.29%, and the FuzzyEn feature detects speech efficiently even at low SNR levels.
We investigated the registration of medical images based on the Normalized Tsallis entropy using mutual information measure. A prerequisite for successful registration is unambiguous maximum of mutual information. We discuss the framework of our algorithm with Normalized Tsallis entropy as the core. Further we propose a type II fuzzy based technique to select the optimal Tsallis parameter q which provides the best alignment. Consequently, specific instances of image registration involving rigid affine transformations were studied. Registration was applied to clinically acquired mammogram. The accuracy was compared with several other techniques. Our algorithm shows promising results. Further, the Need for Pre-registration in mammogram is discussed in detail. Our algorithm can be effective enough to replace Shannon and Tsallis entropy based affine registration.
This article proposes an alternative approach to Tsallis entropy by considering the nonextensive property of mammograms. The novel thresholding technique is performed by normalized Tsallis entropy characterized by another parameter q, which depends on the nonextensiveness of mammogram. In previous studies, q was calculated using the histogram distribution, which can lead to erroneous results when pectoral muscles were included. In the present study, we propose a new technique to calculate q, which is independent of mammogram grades. We use type-II fuzzy sets to find the optimal value of q. The proposed approach has been tested on various images, and the results have demonstrated that the proposed normalized Tsallis entropy approach outperforms the two-dimensional nonfuzzy approach and conventional Shannon entropy partition approach and can be equally effective as Tsallis entropy. Moreover, our technique is completely automatic, and trial and error method is avoided as in previous literatures.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.