Environmental Target Extraction From UAV Image And Visualization Method Research

Corresponding uav remote sensing application technology at home and abroad are analyzed in the aspect of image information of the application, and analyzes the information extraction using remote sensing images in the environmental protection application status. Second, starting from the environmental protection application, on the basis of existing introduced geological information mapping method based on the uav image synthetically, in order to meet the special map in the integrated geographical elements.This article selects the unmanned aerial vehicle(uav) uav image as the research to like. Based uav image to image classification problems, applying the idea of engineering, from the basic theory of basic mathematical principle, method and implementation of computer technology(MATLAB, ENVI, eCongnition) method and so on has carried on the system research, design, the key technology of uav image recognition framework, improve the automatic interpretation method of uav remote sensing images; Image to image recognition method for unmanned aerial vehicle several key technical problems in put forward the feasible technologies and solutions, and a simulation was carried out.Comparing a variety of software, the majority of image processing in various applications such as image correction, image stitching is not very spirit, lack of images for a valid application of mathematics and the theory of object-oriented, image recognition classification applications, limits the use of unmanned aerial vehicle(uav) high resolution image resources efficiency. To the theoretical system of uav image recognition classification as the supplement and extension, surrounding buildings, roads, water, such as identification method and technology to complete the effective exploration and experiment. In accordance with the reservation process, the uav image correction, splicing, feature extraction, feature matching, such as pretreatment, and validation in the MATLAB implementation.Study sums up the main algorithm of uav image mosaicing, validation of SIFT algorithm has made the detailed theory and the algorithm is realized in MATLAB. Using SIFT algorithm to achieve the fast image stitching. HSV model is studied, and by using computer vision and human cognitive psychology theory to analyze the user perception of one color model.Theory is applied to realize the image of different scales, multiple segmentation, after extracting data, classification. Got to like, roads, water, building the data. Make the external symbol has a corresponding digital data. Implements the unmanned aerial vehicle(uav) the original image to the performance of the final figure. Fully express the meaning of the symbol.