D. Govind

V. Sowmya, D. Govind, K. P. Soman

A color image contains luminance and chrominance components representing the intensity and color information respectively. The objective of the work presented in this paper is to show the significance of incorporating the chrominance information for the task of scene classification. An improved color-to-grayscale image conversion algorithm by effectively incorporating the chrominance information is proposed using color-to-gay structure similarity index (C2G-SSIM) and singular value decomposition (SVD) to improve the perceptual quality of the converted grayscale images. The experimental result analysis based on the image quality assessment for image decolorization called C2G-SSIM and success rate (Cadik and COLOR250 datasets) shows that the proposed image decolorization technique performs better than 8 existing benchmark algorithms for image decolorization. In the second part of the paper, the effectiveness of incorporating the chrominance component in scene classification task is demonstrated using the deep belief network (DBN) based image classification system developed using dense scale invariant feature transform (SIFT) as features. The levels of chrominance information incorporated by the proposed image decolorization technique is confirmed by the improvement in the overall scene classification accuracy . Also, the overall scene classification performance is improved by the combination of models obtained using the proposed and the conventional decolorization methods.

D. Govind, Anju Susan Biju, Aguthu Smily

The objective of the present work is to propose a method to automatically detect polarity of the speech signals by estimating instants of significant excitation of the vocaltract and the cosine phase of the analytic signal representation. The phase changes in the analytic signal around the Hilbert envelope (HE) peaks are found to vary according to the polarity of the given speech signal. The relevant HE peaks for the Hilbert phase analysis are selected by estimating the instants of significant excitation in speech. The speech polarity identification rate obtained for the proposed method is almost equal to the state of the art residual skewness method for speech polarity detection. The proposed method also provides the same results for the polarity detection in electro-glottogram signals. Finally, the robustness of the proposed method is confirmed from the reduced detection error rates obtained in noisy environments with various signal to noise ratios (SNRs). The MATLAB codes used for implementing the proposed method are available for download from the following link: http://nlp.amrita.edu:8080/TTS/polarityprograms.zip