Research On Edge Detection Method Of Natural Image

From a long time ago,the image is the main source for people to understand and knowledge the world.In the past,people mainly to the image transmission,and now the image processing,extraction of information is becoming more and more important.Edge is the most basic feature of the image,which contains the main information of the image.As one of the key steps in image processing,edge detection plays a very important role in the following image processing,such as the more in-depth description,recognition and matching.Firstly,the research background and significance of the natural image edge detection methods are introduced.And the research status of the related technologies and algorithms of edge detection are analyzed and summarized.Then this paper introduces the basic edge detection algorithm,including the basic principle and characteristics of the algorithm,and it analyzes and compares these algorithms based on the results of the natural images.Secondly,this paper introduces the principle of the classical ant colony algorithm and the application model based on TSP.And the principle of edge detection algorithm based on ant colony algorithm is analyzed and expounded.Then this paper introduces the edge detection algorithm based on the basic ant colony,and introduces the adaptive distributed threshold strategy.In this strategy,each ant carries a local threshold and dynamically updates it in the image.Then,according to the random initialization of ants will make each experiment to get different results and there exist many ants in the non edge area of the problem,proposed the ant distribution strategy based on gradient,and let these ants uniform distribution.This strategy can make the ants more distributed in the edge region when initializing the distribution.Finally,the experimental results are compared and analyzed.Finally,according to the road image in natural image,the edge detection method is used to detect and identify the lane.Firstly,the image is pre-processed,and then the edge detection is performed by using a variety of edge detection algorithms.And the experimental results are analyzed and compared.Then,according to the background of the image,the improvement of the interference background is proposed by analyzing the characteristics of Lane,and it is based on contour map.This improvement makes the accuracy and efficiency of the detection of the lane line based on Hough transform has been improved.On the basis of the previous work,this paper introduces the identification of the physical properties of lane lines.This has contributed to the identification of lane type and the driver's safe driving.