A Comparative Analysis of Signal Denoising Schemes for Cricket DRS

To obtain noiseless signals from the noisy signals is one of the challenging task. A lot of noise filtering techniques have been employed for noise removal from an audio signal. Wavelet denoising technique is one of the technique that using threshold algorithm for noise removal in audio signals. Double-density dual-tree discrete wavelet transform (DDDTWT) using a level dependent threshold algorithm to eradicate noise from signals and also maintain the signal quality. Audio signal contaminated with Additive White Gaussian Noise is chosen for the implementation. The results in terms of signal to noise ratio (SNR) and root mean square error (RMSE) are compared with the values of dual-tree discrete wavelet transform (DTDWT) and double-density discrete wavelet transform (DDDWT) methods and also with global thresholding method. In this paper, audio denoising techniques PCA blind signal separation, Gaussian low pass filter, Wiener noise reduction and Noise deconvolution for noise reduction are used to increase classification and accuracy of cricket DRS. The results of MATLAB simulations show that the proposed method is more effective and gives better performance for denoising audio signals in terms of both SNR and RMSE. All the denoised snick signals; were passed through PCA blind signal separation, Gaussian low pass filter, Wiener noise reduction and Noise deconvolution when tested by cricket DRS a 98 percentage of classification rate were achieved.