Fuzzy Logic Method for Speckle Noise Reduction in Ultrasound Images and Its Parallel Implementation on Multi-Cores

This paper presents a parallel method aimed at reducing speckle noise in ultrasound medical images, utilizing fuzzy logic and the fuzzy peer group concept.The method is implemented on multi-core interfaces using Open Multi-Processing.The efficiency of the method is evaluated based on execution time, Mean Squared Error, and Peak Signal-to-Noise Ratio.The evaluation is conducted on medical images from the Ultrasoundcases database that have been contaminated with speckle noise.Experimental results demonstrate that the proposed method obtains good performance in terms of the aforementioned objective quality measures.Furthermore, the application of multi-core optimization strategies shows that the new filter can reduce speckle noise in real-time.