Digital Medical Image Edge Detection Algorithm And Implementation

Medical images almost become all of the information is not visible to the naked eye to be visible, which are complex images that reflect the human biological tissues and organs. Modern medicine can not be separated from the information provided by the medical image increasingly. Because Medical images can reflect the patient’s condition visually, and provide the maximum patient information to doctors, which can play a decisive role in disease diagnosis, staging and choosing treatment methods. Doctors depend on the clinical medical images increasingly, and medical images occupy an increasingly important position in modern medicine.The medical image processing technology is in accordance with the clinical requirements, and using computer to process medical images. Because computer processing objects can only be digital images, so this paper’s research objects are digital medical images. Edge is the special gray change mode in the image, which contains the plentiful information. Digital image edge detection is positioning, orienting, and metrics of the gray change point with a specific pattern in the digital image. Which is a key step in digital image processing and analysis, and makes a significant impact on the description of the characteristics matching and recognition of the follow-up in the high-level. However, due to the complexity of the edge itself is coupled with the interference of external conditions, edge detection is a very complex issue. This paper studies the edge detection method of the digital medical image, and proposes two improved edge detection algorithms based on the traditional canny operator and the wavelet transform respectively. To a certain extent, these methods improve the accuracy of edge location, and suppress the interference of the noise.Main contributions of the dissertation include,(1) Introduce several classical edge detection algorithms, and discuss the detection principle of these algorithms. Make the comparative analysis of these algorithms through experimental simulation, and point out each of their advantages and disadvantages.(2) Propose an improved canny edge detection algorithm based on the traditional canny operator. Using iterative algorithms to find the best segmentation thresholds, and using mathematical morphology method to refine the post-test images. This algorithm inherits the advantages of traditional canny operator as accuracy of the original positioning, single-edge response and high ratio of signal-to-noise.(3) Propose an improved image fusion algorithm based on wavelet transform. Firstly, wavelet denoising and smoothing filter on the original image. Secondly, making edge detection with wavelet transform modulus maxima edge detection and improved canny edge detection get each of edge detection images. Thirdly, making wavelet fusion follow certain fusion rules. Finally, making the inverse wavelet transform reconstruct the fused image. The experimental results show that the fused image combines the advantages of both edge detection methods, which is an effective image edge detection method. At last, this paper analysis and outlook the research prospects of the digital image edge detection.