Rev. Roum. Sci. Techn. – Électrotechn. et Énerg., 55, 1, p. 59–68, Bucarest, 2010 DENOISING IMAGES USING A NEW TYPE OF BISHRINK FILTER ALEXANDRU ISAR 1 , SORIN MOGA 2 , DORINA ISAR 1 Key words: MAP-filter, Double tree complex wavelet transform, Sensitivity. This paper presents a new denoising method in the wavelet domain, which aims to reduce the noise, preserving the structural features (like the discontinuities) and textural information of the scene. In this paper we propose the association of the Double Tree Complex Wavelet Transform, (DT-CWT) with a Maximum A Posteriori (MAP) filter named bishrink. The corresponding denoising algorithm is simple and fast. Some simulation results and comparisons prove the performance of the new algorithm. 1. INTRODUCTION The aim of a denoising algorithm is to reduce the noise level, while preserving the image features. In the wavelet domain, the noise is uniformly spread throughout the coefficients, while most of the image information is concentrated in the few largest ones (sparsity of the wavelet representation). Donoho and Johnstone propose a three steps denoising algorithm [1]: 1. the computation of a forward WT, 2. the filtering with a non-linear filter, 3. the computation of the corresponding inverse wavelet transform (IWT). They use the Discrete Wavelet Transform (DWT) and the soft-thresholding filter. This filter puts to zero all the wavelet coefficients with the absolute value smaller then a threshold. This threshold is selected to minimize the min-max approximation error. The soft-thresholding filter was enhanced in [2–4]. A highly appealing particularity of the WTs is the inter-scale dependence. If at a given scale a coefficient is large, its correspondent at the next scale (having the same spatial coordinates) will be also large. In [2–4] the inter-scale dependencies are used to improve the denoising performance. The wavelet coefficients statistical models which exploit the dependence between coefficients give better results compared to the ones using an independent assumption [5, 6, 8, 9]. The denoising is performed in [5] and [6] with the aid of maximum a posteriori filters, (MAP). If we denote 1 University “Politehnica” of Timişoara, E-mail: alexandru.isar@etc.upt.ro 2 Telecom Bretagne, Brest, France