Underwater Image Edge Detection Based On The K-means Algorithm

Marine science and technology has become the focus of world with the economic and technological development. For the purpose of underwater exploration and construction, wide range and accurate underwater environmental data are needed. In order to obtain these data, a series of ocean technologies were under researched. And the underwater imaging method is researched in this article. Underwater image is important for scientific research and technology. Edge detection is the basis for digital image processing, there are many edge detection algorithms. But due to the absorptive and scattering nature of seawater, underwater images are essentially characterized by their poor visibility and color distortion. The traditional edge-detection algorithms are always ineffective to underwater images.In this paper the K-means clustering algorithm is used to obtain the accurate edges of underwater pipeline. The research contents are as followings:The image edge detection process and the international existing image edge detection approach is analyzed. We use the dark channel prior method to get the clear original image. Then, we calculate the processed image ’s gradient. Next, the K-means clustering algorithm is used to classify the endpoints, and retain the endpoints in the original edge image.The inhomogeneous illumination underwater image detection system is introduced, using it to take the images in the extraction of natural waters. Experimental results show that the effect of edge detection underwater images have been significantly improved.