Fractional Edge Detection Techniques for Radiographic Images based on Fuzzy Systems Abstract: - Medical images are a diagnostic technique that facilitates the doctor's job the doctor to early diagnose the patient. So, fractional edge detection algorithms have gained focus of many researchers with the medical images. The fractional edge detection could be helped in the diagnosis of early stages of diseases like Alzheimer and fracture bone, but it can lose some edge details that can facilitate the doctor's job the doctor in the diagnosis. In this paper, different fractional edge detection algorithms based on fuzzy logic have been used to enhance the performance of the edges as there is no loss in the fuzzy-based methods. A performance comparison has been done between different fractional edge detection algorithms with and without fuzzy logic. Evaluation of the noise performance upon addition of salt and pepper noise is measured through peak signal to noise ratio (PSNR) and bit error rate (BER) simulated by using MATLAB. Key-Words: - Edge Detection, Fractional Systems, Soft Computing Techniques, Radiographic, Fuzzy Systems, Medical Images 1 Introduction Medical Imaging has gained focus of many researchers as it played an important role in helping the doctors to early diagnosis a lot of diseases [1]. The medical images are mostly used as radiographic techniques to help in early diagnosis and curing [2]. Nowadays, the quality of the digital image is ameliorated by using the image enhancement techniques for additional processing [3]. Image segmentation is the key challenge of object recognition and computer vision. It has the goal to extract the information, the first step in image analysis [4]. It is the method of partitioning to extract interest parts in a simple and easy analyzed way. Edge detection can be deemed as one of the most common techniques in many applications in the field of image processing such as biomedical, radiographic images. Image edges contain enormously valuable information, so it is the foundational step for high-level information acquisition [5]. Its goal to distinguish and locate the sharp changes in brightness of an image [6]. The most commonly used discontinuity based edge detection techniques are Robert edge detection [7], Sobel Edge Detection [7], Prewitt edge detection [7], Kirsch edge detection [8], LOG (Laplacian of Gaussian) edge detection [9], [10] and Canny Edge Detection [11]. Edge detection can be implemented by either the Gradient or Laplacian methods. To detect the edges, the Gradient method is seeking for the minimum and maximum in the 1 st derivative of the image while the Laplacian method is seeking for the zero crossings in the 2 nd derivative of the image [12]. The edge detection implements by using the WESSAM S. ElARABY * , AHMED H. MADIAN *, *** , MAHMOUD A. ASHOUR * , IBRAHIM FARAG ** , MOHAMMAD NASSEF ** *Radiation Engineering Department, NCRRT, Egyptian Atomic Energy Authority, CAIRO, EGYPT **Department of Computer Science, Faculty of computing and Information, Cairo University, CAIRO, EGYPT ***Nanoelectronics Integrated System Center (NISC), Nile University, CAIRO, EGYPT Email: eng.wessamsayed@yahoo.com , amadian@nu.edu.eg , ma_ashour53@eaea.org.eg , i.farag@fci-cu.edu.eg , m.nassef@fci-cu.edu.eg WSEAS TRANSACTIONS on INFORMATION SCIENCE and APPLICATIONS Wessam S. Elaraby, Ahmed H. Madian, Mahmoud A. Ashour, Ibrahim Farag, Mohammad Nassef E-ISSN: 2224-3402 155 Volume 15, 2018