本文關(guān)鍵詞： 幾何校正 精度分析 不同分辨率　出處：《中國地質(zhì)大學(xué)（北京）》2013年碩士論文　論文類型：學(xué)位論文

【摘要】：遙感圖像的成像與獲取的內(nèi)在規(guī)律決定了所獲取的遙感圖像并不能被直接應(yīng)用，因而為提高影像的與實際地形的匹配度，需要對影像進(jìn)行幾何校正處理，因而從理論上探索遙感影像的變形規(guī)律，研究高精度、便于實施的遙感影像校正方法且誤差最小的方法已經(jīng)成為遙感應(yīng)用領(lǐng)域亟待解決的熱點問題。幾何校正是遙感影像處理過程的關(guān)鍵環(huán)節(jié)，為此，需對遙感影像的幾何質(zhì)量和適用性進(jìn)行必要的評價。本文以中分辨率的RapidEye遙感影像數(shù)據(jù)和高分辨率的WorldView-2衛(wèi)星影像數(shù)據(jù)為基礎(chǔ)，開展了遙感影像幾何校正及校正后精度的研究工作。在對數(shù)據(jù)處理的基礎(chǔ)上系統(tǒng)的研究并評價了幾何校正模型、控制點來源、控制點數(shù)量、遙感影像分辨率對遙感影像的幾何校正精度的影響及兩種分辨率影像能滿足的制圖比例尺。通過比較分析，取得了以下成果： 1.遙感影像幾何校正精度，，與校正數(shù)學(xué)模型的選取、與校正用控制點的數(shù)量和位置、與控制點的來源和精度、與遙感影像的空間分辨率等因素有關(guān)。 2.遙感影像的幾何校正模型的選擇影響幾何校正的精度，對地勢起伏較大區(qū)域應(yīng)先進(jìn)行正射校正后再進(jìn)行三次多項式法幾何校正可以有效提高幾何校正精度，對地勢起伏小的區(qū)域可以直接進(jìn)行二次多項式校正，校正后的精度可以達(dá)到規(guī)范要求。 3.進(jìn)行多項式法幾何校正時，幾何校正控制點的數(shù)量及定位精度是遙感影像幾何校正精度的保證。分別以5米DEM數(shù)據(jù)和30米DEM數(shù)據(jù)為控制，通過應(yīng)用不同精度的基礎(chǔ)數(shù)據(jù)對遙感影像進(jìn)行幾何校正后并對校正后的精度進(jìn)行了評定，基礎(chǔ)數(shù)據(jù)精度越高，校正后影像的幾何精度也越高。但對于中分辨率的遙感影像來說，基礎(chǔ)數(shù)據(jù)精度的提高對校正后影像幾何精度的提高并不明顯。 4.在相同校正的基礎(chǔ)條件時，對不同分辨率的兩種影像的幾何校正精度進(jìn)行了對比，分辨率高的遙感影像其幾何校正精度以及遙感制圖的精度高，但是為了避免資源浪費，應(yīng)該依據(jù)制圖比例尺選擇合適的遙感影像進(jìn)行制圖。
[Abstract]:The inherent law of remote sensing image imaging and acquisition determines that the obtained remote sensing image can not be directly applied, so in order to improve the matching degree between the image and the actual terrain, the geometric correction of the image is needed. Therefore, the deformation law of remote sensing image is explored theoretically and the high precision is studied. The method of remote sensing image correction which is easy to implement and the method with minimum error has become a hot problem to be solved urgently in the field of remote sensing application. Geometric correction is the key link in the process of remote sensing image processing. It is necessary to evaluate the geometric quality and applicability of remote sensing image. This paper bases on RapidEye remote sensing image data with medium resolution and WorldView-2 satellite image data with high resolution. Base. On the basis of data processing, the geometric correction model, the source of control points and the number of control points are systematically studied and evaluated. The effect of the resolution of remote sensing image on the geometric correction accuracy of remote sensing image and the mapping scale which can be satisfied by the two kinds of resolution images. Through comparison and analysis, the following results have been obtained: 1. The geometric correction accuracy of remote sensing image is related to the selection of calibration mathematical model, the number and position of control points for correction, the source and accuracy of control points, and the spatial resolution of remote sensing images. 2. The choice of geometric correction model of remote sensing image affects the accuracy of geometric correction. The geometric correction with cubic polynomial method can effectively improve the accuracy of geometric correction in areas with large terrain fluctuation, and can be directly corrected by quadratic polynomials for areas with small terrain fluctuation. The corrected precision can meet the specification requirements. 3. In polynomial geometric correction, the number of geometric correction control points and the positioning accuracy are the guarantee of geometric correction accuracy of remote sensing image, which are controlled by 5 m DEM data and 30 m DEM data, respectively. The geometric correction of remote sensing image is carried out by using different precision basic data and the accuracy after correction is evaluated. The higher the accuracy of the basic data is. The geometric accuracy of the corrected image is higher, but for the middle resolution remote sensing image, the improvement of the basic data precision is not obvious. 4. In the same basic condition of correction, the geometric correction accuracy of two kinds of images with different resolution is compared. The geometric correction accuracy of the high resolution remote sensing image and the precision of remote sensing mapping are high. But in order to avoid the waste of resources, we should choose the appropriate remote sensing image according to the scale of cartography.
【學(xué)位授予單位】：中國地質(zhì)大學(xué)（北京）
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2013
【分類號】：P237;TP751

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 曲喬新;王素敏;劉媛媛;張帆;;WorldView-1衛(wèi)星影像處理方法初探[J];北京測繪;2010年01期

2 吳露露;羅旭;張雁;汪波;馮仲科;;中低分辨率遙感影像幾何精校正[J];北京林業(yè)大學(xué)學(xué)報;2008年S1期

3 龔明R

本文編號：1447216