Research On Some Key Issues Of Landuse Rapid Inspection Based On Uav Images

In the high-speed development of the information age, how to obtain the ground space and time data fast and accurately has become the research focus issue, and for the growing requirements of high resolution remote sensing image and geographic space information, as well as to its present situation, real-time and accuracy demand more and more. At present, collection the ground time and space information technology are divided into two kinds: one kind is the ground measurement methods, such as:large-scale digital mapping by using Total Station technology, another kind is the earth observation from space methods, such as: GPS measurement, satellite remote sensing observation, airborne remote sensing technology. The conventional relatively measuring method, such as using satellite, big planes as the platform for aerospace photography measurement technology have been extensively applied, but these methods affected greatly by the weather, it is difficult to real-time obtain good quality high resolution images in the cloud area. And UAV remote sensing technology is a new developed fast geographic space information acquisition technology. As using UAV for the platform can fly low altitude in any complex topographical conditions and under the cloud, and can obtain high resolution images. At the same time, the platform has advantages of low price, convenience and flexible, simple operation, and not be the limiting glorious etc, and can obtain different resolution images by low altitude flight based on different demand.Firstly, how to obtain high resolution image data by UAV remote sensing platform was disscussed.The problem of UAV image processing efficiency is not high was researched and analysed, and automatic Extraction of image control points for UAV image based on POS data. Automatic mosaic in UAV images was completed based on extracting and matching feature points. Aiming at lack of general segmentation evaluation model, a evaluation model suitable for high resolution image segmentation scale choice was established.and the segmentation results were evaluated, and the object-oriented classification method was used for UAV image classification. Based on the above research, a technological system for landuse rapid inspection based on UAV images was integrated.The main research contents and achivements of this dissertation are summarized concretely as follows:1. According to the principle of correct UAV POS data, POS data correction model was established and POS data error correction parameter were acquired by layout a small amount of the control points in regional network, and the corrected POS data were used in extraction of UAV images control points..2. The steady SITT algorithm was introduced into the automatic mosaic of UAV images, and the time consumption which various stages of the algorithm was analyzed. Considering the feature of UAV to make improvement to the algorithm, which can reduce the range for searching feature points through estimating the overlap of neighboring images, and the most superior size of Gaussian kernel adapted in the UAV images was obtained through the experiment when detecting the criterion space extreme points, it can overcome the traditional SIFT algorithm’s flaw by using the fixed size method, not only reduce the time consumption, but also obtain the characteristic points as far as possible.3. Compared with the traditional aerophotogrammetry, the UAV images have the problems of bigger and more complex geometric distortion and because of the large photographic scale, the texture of images changes little in some features. These factors lead more difficult to matching UAV images, and increase the number of mismatch cognominal points. A method for detecting and eliminate the mismatch points was proposed based on the polygon matching. The method considers the reliability of cognominal points matching from distance, angle and weight.4. The image segmentation is the top issue of the object-oriented image analysis technique, how to choose the right image segmentation scale has become the key of object-oriented analysis technology. In order to improve the accuracy of remote sensing image segmentation scale, a segmentation quality function was established through analysis of images. Segmentation experiments according to different resolution multi-source remote sensing images, and the optimal segmentation scale was obtained. Finally a segmentation matching index which considering area and spectrum was proposed to evaluate the optimal segmentation scale.