A noise basesd hybrid thresholding for enhanced noise reduction using double density dual tree complex wavelet transform

Nowadays de-noising plays a vital role in image processing and a lot of research has contributed in this area and many algorithms were developed but still it requires further improvement. Present de-noising techniques use threshold with windowing techniques such as dual tree CWT(DTCWT), Double Density DWT or Double Density Dual Tree Discrete Wavelet Transform (DDDTDWT). During Image enhancement the noise also increases along with desired image information. To reduce this increased noise previous research provides two types of threshold techniques such as static and adaptive. In static threshold, a random noise model is considered where as in adaptive threshold known directive noise samples taken from input image instead of the unknown noise. In this paper, a hydride threshold technique is proposed by combining both the threshold techniques based on noise levels to improve the noise reduction and PSNR values. This research employs the DDDTDWT technique which has more directionality than DTCWT along with Soft threshold and shrinkage to reduce the noise. Here promising results are compared with proposed hybrid threshold technique and previous static and adaptive threshold techniques.