International Journal of Computer Applications (0975 – 8887) International Conference on Advances in Science and Technology (ICAST-2014) 22 Study on the Reduction of Speckle Noise in Ultrasound Images Parineeta Suman M.E (Pursuing) Saraswati College of Engineering, Kharghar Deepa Parasar Assistant Professor Saraswati College of Engineering, Kharghar Vijay R. Rathod, Ph.D. IIT Roorkee St. Xavier’s Technical Institute ABSTRACT The ultrasound images, the speckle noise is inherent in medical ultrasound images and it is the cause of reduced contrast-to-noise ratio and resolution. The presence of speckle noise is not attractive because it reduces the image quality and affects the tasks of the individual interpretation and paper diagnosis. In this paper, we study various techniques to reduce speckle noise from ultrasound images Post acquisition method as the only scale spatial filtering method and multi-scale method. Keywords De-noising, Speckle Noise, isotropic diffusion , Contourlet transform 1. INTRODUCTION Ultrasound images are widely used in the medical field as it is relatively safe, economical, adaptable and transferable. Despite many advantages, ultrasound images contain noise called "speckle" that corrupts the image resolution and contrast makes the interpretation of the physical structure lying below extremely difficult for physicians. Speckle noise in ultrasound image is introduced due to the nature of the consistency of the imaging modality, the interference of backscattered signals of each sub-cell resolution and resolution on the receiver or transducer. Backscattered waves undergo constructive or destructive interference randomly spoils the image in random granular model, known as speckle. Due to the process of acquiring the ultrasound speckle noise is multiplicative in nature, which is directly proportional to the gray level in an area and is statistically independent signals. The main purpose of image denoising techniques to eliminate noise while preserving as much as possible the important information about the image. Thus, the speckle is considered the main source of noise in ultrasound imaging and should be treated without affecting the characteristics of important picture. The main objectives of size reduction of medical ultrasound imaging are: 1) To improve the human interpretation of ultrasound images - size reduction allows cleaning of the ultrasound image with clear boundaries. 2) Despeckling is a pre-treatment for many tasks of ultrasound images as the segmentation processing and recording - speckle reduction improves the speed and accuracy of segmentation and automatic recording and semi- automatic. 2. METHODS FOR SPECKLE REDUCTION Several techniques have been proposed for removing impurities in the medical ultrasound imaging. In this section, we present the theoretical overview and classification of existing techniques despeckling 2.1 Compounding Methods In the mixing method [1] - [3] a series of ultrasound images of the desired image are acquired from different directions and at different scan frequencies or transducers in different strains. Then, images are averaged to form a composite image. The method of composition may improve the detectability of the desired image, but they suffer from degraded spatial resolution and increased system complexity. 2.2 Post-Acquisition Methods This method does not require much technical system modification of image processing .The post-acquisition falls under two categories (1) scale spatial filtering methods (2) multiscale 2.2.1 Scale Spatial Filtering Methods Speckle reduction filter that modifies the degree of smoothing depending on the ratio of local variance to local mean is that developed. In smoothing method is increased in the region where homogeneous and fully developed speckle is reduced or avoided by other regions to preserve the details. In this method, we assume that the pixels that have similar gray level and connectivity are related and likely to belong to the same object or region. After all pixels are allocated to different groups, the spatial filtering is performed on the basis of local statistics adaptive regions whose sizes and shapes are determined by the information content of the image. The main difficulty in applying region growing based methods is to design appropriate similarity criteria for region growing. Various types of filters are used in the application of ultrasonic imaging despeckling. The types of filters most commonly used are: a. Lee and Kaun Filter [4] - The Lee filter [5] is based on the approach that the smoothing is performed on the area having low variance. The Lee filter form an output image by calculating a linear combination of the intensity of the center pixel in the filter window with the average intensity of the window.The Kuan filter [6] is a generalization of the Lee filter. The Kuan filter converts the multiplicative model of speckle into an additive linear form. However, the formulation is different from lee fiter.The main disadvantage of Lee filter is that it tends to ignore speckle noise in the areas closest to edges and lines. This filter is incapable of removing high frequency noise and also it cannot remove noise in high and low variance regions whereas in kaun filter, calculation of ENL, which determines the level of speckle in an image is measure problem.