【摘要】：無(wú)人機(jī)航拍是獲取空間數(shù)據(jù)的重要途徑,被廣泛應(yīng)用于軍事和民用領(lǐng)域中.其中,基于無(wú)人機(jī)航拍視頻影像的三維重建技術(shù),在城市規(guī)劃、變化檢測(cè)、災(zāi)害評(píng)估等應(yīng)用中發(fā)揮著重要作用.在視頻影像的三維重建流程中,特征匹配是基礎(chǔ)步驟,為相機(jī)方位及參數(shù)估計(jì)提供可靠輸入信息;相機(jī)方位估計(jì)是三維重建的關(guān)鍵環(huán)節(jié),其估計(jì)精度與三維重建的效果息息相關(guān).因此,如何提升特征匹配的速度和相機(jī)方位的估計(jì)精度是當(dāng)前圖像處理、三維重建等領(lǐng)域的研究熱點(diǎn).為此,本文針對(duì)無(wú)人機(jī)航拍視頻影像的特點(diǎn),重點(diǎn)研究了快速特征匹配和相機(jī)方位估計(jì)問(wèn)題,主要研究工作如下:1、針對(duì)航拍視頻影像的特征點(diǎn)提取與匹配速度問(wèn)題,提出了無(wú)跡卡爾曼濾波和KLT匹配算法相結(jié)合的特征點(diǎn)跟蹤算法,實(shí)現(xiàn)了相鄰幀特征點(diǎn)的方位預(yù)測(cè)與快速匹配.該算法首先針對(duì)航拍視頻特點(diǎn),利用無(wú)跡卡爾曼濾波在相鄰幀中進(jìn)行特征點(diǎn)預(yù)測(cè),以確定匹配范圍;其次,根據(jù)KLT匹配算法對(duì)特征點(diǎn)進(jìn)行跟蹤,得到的匹配結(jié)果作為觀測(cè)值;最后,通過(guò)卡爾曼增益修正得到特征點(diǎn)的準(zhǔn)確位置.對(duì)比實(shí)驗(yàn)證明了該算法不僅高效,而且在匹配精度方面優(yōu)于KLT算法.2、針對(duì)航拍視頻幀視差變化小的特點(diǎn),提出了關(guān)鍵幀篩選算法,以減少頻繁相機(jī)方位估計(jì)帶來(lái)的累計(jì)誤差問(wèn)題.該算法以特征點(diǎn)對(duì)數(shù)量和運(yùn)動(dòng)大小作為衡量標(biāo)準(zhǔn),首先利用特征點(diǎn)對(duì)計(jì)算出圖像幀之間的平移量和旋轉(zhuǎn)量,然后進(jìn)行加權(quán)綜合求出圖像間的相異度,設(shè)置閾值篩選關(guān)鍵幀.對(duì)比實(shí)驗(yàn)證明了通過(guò)本文算法得到的關(guān)鍵幀的相機(jī)方位比ORB-SLAM得到的關(guān)鍵幀的初始相機(jī)方位精度高.本文首先提出無(wú)跡卡爾曼濾波和KLT光流結(jié)合算法實(shí)現(xiàn)了特征點(diǎn)的快速匹配,為相機(jī)方位估計(jì)提供了可靠的數(shù)據(jù).然后提出關(guān)鍵幀篩選算法得到了關(guān)鍵幀的相機(jī)方位的精確估計(jì)值,相關(guān)結(jié)果可為快速三維重建方法提供基礎(chǔ)理論.
[Abstract]:UAV aerial photography is an important way to obtain spatial data and is widely used in military and civilian fields. Among them, 3D reconstruction technology based on UAV aerial video image plays an important role in urban planning, change detection, disaster assessment and other applications. In the three-dimensional reconstruction process of video image, feature matching is the basic step, which provides reliable input information for camera azimuth and parameter estimation. Camera azimuth estimation is the key link of three-dimensional reconstruction, and its estimation precision is closely related to the effect of three-dimensional reconstruction. Therefore, how to improve the speed of feature matching and the accuracy of camera azimuth estimation is a hot topic in the field of image processing, 3D reconstruction and so on. Therefore, aiming at the characteristics of UAV aerial video image, this paper focuses on the fast feature matching and camera azimuth estimation. The main research work is as follows: 1. Aiming at the feature point extraction and matching speed of aerial photograph video image, the main research work is as follows: 1. A feature point tracking algorithm based on unscented Kalman filter and KLT matching algorithm is proposed to realize azimuth prediction and fast matching of feature points in adjacent frames. Firstly, the algorithm uses unscented Kalman filter to predict the feature points in adjacent frames to determine the matching range. Secondly, according to the KLT matching algorithm, the feature points are tracked and the matching results are used as observation values. Finally, the exact position of the feature points is obtained by Kalman gain correction. The experimental results show that the proposed algorithm is not only efficient, but also superior to KLT algorithm in matching accuracy. 2. In view of the small variation of disparity between aerial video frames, a key frame filtering algorithm is proposed. In order to reduce the cumulative error caused by frequent camera azimuth estimation. In this algorithm, the number of pairs of feature points and the size of motion are used as the criterion. Firstly, the translation and rotation between frames are calculated by using the pair of feature points, then the dissimilarity between images is calculated by weighted synthesis, and the threshold value is set to filter the key frames. The experimental results show that the camera azimuth accuracy of the key frame obtained by this algorithm is higher than that of the key frame obtained by ORB-SLAM. In this paper, a combination algorithm of unscented Kalman filter and KLT optical flow is proposed to realize the fast matching of feature points, which provides reliable data for azimuth estimation of camera. Then the key frame filtering algorithm is proposed to obtain the accurate estimation of the camera orientation of the Keyframe. The related results can provide the basic theory for the fast three-dimensional reconstruction method.
【學(xué)位授予單位】：集美大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 劉華詠;郝會(huì)芬;李濤;;基于視頻聚類的關(guān)鍵幀提取算法[J];物聯(lián)網(wǎng)技術(shù);2014年08期

2 陳智;;淺議無(wú)人機(jī)航拍的應(yīng)用[J];時(shí)代金融;2014年15期

3 吳偉交;王敏;黃心漢;毛尚勤;;基于向量夾角的SIFT特征點(diǎn)匹配算法[J];模式識(shí)別與人工智能;2013年01期

4 肖晗;原魁;何文浩;柴曉杰;;基于FPGA的SIFT特征點(diǎn)檢測(cè)[J];高技術(shù)通訊;2012年04期

5 冷大煒;孫衛(wèi)東;;一種高數(shù)值精度的P3P問(wèn)題半閉式解法[J];計(jì)算機(jī)應(yīng)用研究;2012年01期

6 楊現(xiàn)輝;王惠南;;ICP算法在3D點(diǎn)云配準(zhǔn)中的應(yīng)用研究[J];計(jì)算機(jī)仿真;2010年08期

7 劉曉光;陳曦;陳政偉;孫靜波;;基于圖像灰度的SSDA匹配算法[J];航空計(jì)算技術(shù);2010年01期

8 張曼;沈旭昆;;一種基于尺度空間的三維點(diǎn)云數(shù)據(jù)配準(zhǔn)算法[J];系統(tǒng)仿真學(xué)報(bào);2009年S1期

9 許海霞;王耀南;袁小芳;朱江;周維;;基于矢量差分的未標(biāo)定攝像機(jī)P5P問(wèn)題的求解[J];自動(dòng)化學(xué)報(bào);2009年08期

10 湯建良;;一個(gè)關(guān)于P5P問(wèn)題的求解算法[J];廣西師范大學(xué)學(xué)報(bào)(自然科學(xué)版);2009年01期

相關(guān)碩士學(xué)位論文 前5條

1 張一;無(wú)人機(jī)遙感影像點(diǎn)特征匹配算法研究[D];解放軍信息工程大學(xué);2015年

2 吳昊;基礎(chǔ)矩陣估計(jì)方法研究[D];蘭州大學(xué);2015年

3 謝榛;基于無(wú)人機(jī)航拍圖像的室外場(chǎng)景三維重建技術(shù)研究[D];浙江工業(yè)大學(xué);2014年

4 曹長(zhǎng)青;基于內(nèi)容的視頻檢索中關(guān)鍵幀提取算法研究[D];太原理工大學(xué);2013年

5 明巍;基于內(nèi)容的視頻檢索中關(guān)鍵幀提取算法研究[D];武漢工業(yè)學(xué)院;2010年

，

本文編號(hào)：2441085