The Algorithm Research Of Power Line Real-time Detection Based On Matlab

Image information can provide important information to humans. It is an important means of human access to information, with a large amount of information, fast transmission, the long-distance range and so on. Digital imaging systems are moving in the direction of automation and intelligence to continue moving forward, and a lot of image input and output devices have gradually become a common multimedia and computer peripheral equipment. For digital images, there are many ways for edge detection, such as the use of arc transform. Their principles are to measure the gray level jump in some meaningful way, and then define the existence of the edge. An ideal edge is a connected set of pixels, each pixel is on a vertical gradient step of the gray level jump. Actually, the imperfect of optical system, sampling and other image acquisition cause the vague of the edge, the degree of which depends on factors like the performance of image acquisition system, sampling rate and image lighting conditions.Image gradient is an important aspect of digital image processing, as well as an important prerequisite of image processing technology including image segmentation, feature extraction and image recognition. This paper presents a method calculating the image gradient to obtain the gradient vector histogram and achieve the initial recognition of the power line detection method. According to this method, color image processing method based on three channels and its analysis and comparison have been put forward. With the accelerative development of the grid, the low-flying helicopters would be in increasing danger of collision with power lines, so the power line detection and identification system of helicopters is necessary. Analyzing the image we obtained and identifying the information, we can get more scientific and reliable results in a shorter period of time.The algorithm provided by this paper firstly converts color image into gray image, and detects its edge. After making a removal of noise and using line detection template to calculate the gradient of each pixel value, we can find out the histogram of the gradient vector. Finally, put the data into the simulation environment of Matiab, simulate the image and gradient vector histogram of the original image, the Gaussian filtered image and image of detection based on one order differential operator respectively. The simulation results show that the method has better noise immunity and accurate positioning. The image analysis data is more complete, and the details are more significant than before. After obtaining the conclusion with this algorithm, this paper studies from the angle of three channels, and analyze and compare the image processing results of three channels respectively, to find out more accurate distinguishing method.