Abstract: Ultrasound imaging is a non-invasive and inexpensive technique for detection of kidney stones. As the ultrasound images are affected by speckle noise, the segmentation of the images remains a challenging task. The manual detection and measurement of segmented stones become cumbersome and suffers from inter-observer variability. Hence, a computer aided algorithm is required for automatic stone detection and reproducibility with robust despeckling and segmentation techniques. In this study, an algorithm is developed by using Adaptive Bilateral Filter (ABF) for reducing speckle noise and mathematical morphological operations for segmentation of stones in ultrasound kidney images. The speckle reduction performance of ABF is evaluated byPeak Signal to Noise Ratio (PSNR), Structural Similarity Index Metrics (SSIM) and Edge Preservation Index (β).The proposed stone detection algorithm is analyzed through Pratt’s Figure of Merit (FOM).