Based On Computer Vision Logistics Tower Location Positioning Method Research

In real life,logistics has become a part of people's life,quietly entered our lives,the development of the electricity supplier,improve the living standards of the people,are signs of logistics technology has become the most concern now social aspects.This paper puts forward that "logistics tower",the main thing is to combine logistics and automated warehouse reasonably,and cooperate with each other."Logistics tower" includes computer technology,logistics technology,automation,threedimensional warehouse and transportation and other projects,is a convenience for people's lives,but also to save time and money projects.The main work of this paper is the background of computer vision,image taken with the camera to get shelf,realize location positioning.First of all,the image is pre processed,and the amount of information contained in the color picture is large,which increases the difficulty of image processing.In order to reduce the amount of calculation and improve the efficiency of processing,the image is processed with gray level.Then,the gray image has been denoised.Directly to the camera to capture the color image processing has a great deal of calculation,reduce the efficiency of the algorithm.Secondly,the research of Canny edge detection.On the basis of the traditional Canny edge detection,this paper proposes an adaptive threshold edge detection method for fog and haze,and analyzes its performance,taking into account the shortcomings of the classical Canny algorithm.The collected image smoothing image,add Gauss calculate the gradient magnitude and direction,the gradient amplitude,non maxima suppression,double threshold and edge connection process,using the method of dark channel prior,remove the noise caused by haze.The adaptive median filter is used to solve the deviation of the image operation under the fixed threshold,better to find the noise points,and do not harm the edge information.The final image contour extraction is advantageous to preserve the edge information.To solve the above problems,the algorithm has the characteristics of noise elimination,fog removal,and better generalization ability,and can provide a good solution to the problem.Thirdly,the research on realizing shelf positioning.Even if the calibration method based on calibration of the technology in the traditional camera,this paper presents two order distortion,relationship between image coordinate and world coordinate system and camera coordinate system,the proposed algorithm is used to solve the coupling problem,the algorithm mentioned in this paper combines linear imaging distortion circle approximation method and use the area of model approximation algorithm.In order to realize the maximum noise removal for the image taken,the experimental results are better.