UDC 621 IMAGE DENOISING BASED ON WAVELET USING DIFFERENT THRESHOLDING TECHNIQUES © Pushpa Koranga, Graphic Era Hill University, Dehradun, India Garima Singh, Graphic Era Hill University, Dehradun, India Abstract. <...> When image is captured it sometimes gets degraded by noise. <...> So the main aim of image denoising is to remove those noise for better quality picture which is needed in different fields such as medical, Astrophysics, geographical location, etc. <...> Till now different techniques have been adopted such as Fourier transform, discrete cosine transform, etc. <...> Fourier Transform has its own drawback so to overcome these drawbacks Wavelet Transform is used. <...> In this paper we discuss different thresholding techniques of wavelet for image denoising. <...> The main aim of the image denoising is to recover better quality picture or original image. <...> In this wavelet threshold is compared against threshold value, if it is less than threshold then set to zero otherwise modified or kept as it is [5, 7]. <...> First method is to design a stastistical optimal wavelet threshold parameter for non linear shrinkage wavelet [4]. <...> These are formed 102 Engineering and Automation Problems, № 2 – 2017 by horizontal and vertical filter [10]. <...> Many wavelet thresholding techniques like Sureshrink and Bayes shrink provided better efficiency in image denoising [6]. <...> Image denoising using Wavelet Transform consists of following steps: i) An input noisy image Y(t) is passed such that Y (t) =X (t) +N (t), where X (t) is original input data, Y (t) is Output noisy data, N (t) = Noisy data (can be Gaussian noise, speckle noise, salt and pepper noise, poisson noise). ii) Forward wavelet transform is done Y(t) ↔ W(t) (wavelet transform). <...> SURVEY a) Bayes shrink: it is a wavelet shrinkage techniques for removing noise. <...> First of all threshold value is calculculated and compared with wavelet coefficient [1]. <...> If its value is less then threshold value <...>