Research On Edge Detection Technology Of Color Image

Image is the most direct and main form of getting information from outside. And edge is one of the most important features of the image. Image edge detection holds a very important position in the field of digital image processing. It is the basis of many kinds of image processing technology such as image restoration, data hiding, pattern recognition and so on, which has been widely applied in social life such as national defense, military, industry, agriculture, medicine and so on. As people’s increasing demand of color image and the progress of the color technology, color image processing gradually becomes one of the most important tasks in image processing. And the research of color image edge detection technology has been becoming very active.The paper is closely around the subject of color image edge detection, and gradually carries on the theories and experiment discussion. Firstly, from the point of view of the application, it introduces several common color models and transformational relations from each other, such as the most widely used RGB model at present、CMY(CMYK) model used in the color printing industry、YUV model and YIQ model in the color TV system,which are facing the hardware devices. And HSI model、HSV model、L*a*b*model which are facing visual perception. Then it sets forth the technology of color image edge detection on the basis of gray image edge detection technology, introduces the component output synthesis method in detail and carries on contrastive analysis on the detection performance of the two kinds of traditional color image edge detection methods through experimental simulation. At last, the paper proposes an improved edge detection algorithm based on the wavelet decomposition, combined with the wavelet analysis theory, and extend it to the field of color image edge detection. By using the powerful local analysis ability of wavelet decomposition, the algorithm analyses and processes the image more deeply and improves its performance.Through the experiment simulation in RGB color model, it shows that the nature of the algorithm is better than that of the classical edge detection algorithm. The edges detected by the algorithm are whole and exquisite, and it has great noise restraining ability.. However, the results of color image edge detection also depend on the choices of the color model, so the algorithm needs further optimization to enhance its generality and improve its detection performance.Due to the not long development period of color image edge detection, with the addition of the difficulty and depth of the subject itself, the topic of the color image edge detection needs further study.