Gradient Vector Flow Based Technique For Kidney Restoration In MR Images

The biomedical image analysis methods are the effective and important tool for screening, detection, and diagnosis. It is nearly inevitable tool and works as the helping hand for physicians. One of the major issues related with the biomedical images is that most of the images (of different modalities) are suffered from noise and other different quality related problems like poor contrast, blurring, and difficulties in extracting suitable information. Therefore, Accurate Kidney segmentation is challenging in the field of bio-medical image analysis especially for computer aided diagnosis and treatment planning of kidney diseases due to intensity inhomogeneity caused by imperfections of image acquisition process. During Kidney segmentation, there are chances of missing pixels, verify the kidney shape is important. In this paper, approach proposed for restoration poor quality MRI kidney images. an effective method designed for restoring missing kidney edges from poor quality MRI images based on gradient vector flow (GVF) and neighbor’s component grouping. The GVF of edge pixels in the Canny edge map of the input image is explored to identify the dominant edge pixels which represent kidney components. And this done by implementing three conditions, these conditions depend on “arrows” (gradient directions singe), angle between GVF two vectors, and magnitude value. The proposed method evaluated on collected dataset from hospital the dataset is effective because it includes images affected by different cause and collected images are real. Experimental results show that the proposed model outperforms the existing methods.