計算機理論論文基于特征點的圖像匹配技術研究

1、基于特征點的圖像匹配技術研究 基于特征點的圖像匹配技術研究 the research on image matching based on feature point【中文摘要】 圖像配準是對取自不同時間、不同傳感器或者不同視角的同一場景的兩幅圖像或者多幅圖像進行匹配的過程,隨著計算機視覺技術的發(fā)展,圖像匹配技術被廣泛地應用在遙感圖像、醫(yī)學圖像、三維重構、機器人視覺等諸多領域中。圖像配準的方法大致分為兩類:基于像素的圖像配準方法和基于特征的圖像配準方法。而基于特征的圖像配準方法是目前圖像配準最常用的方法,其最大的優(yōu)點在于能夠將對整個圖像進行的各種分析轉化為對圖像特征(特征點、特征曲線等)的分析

2、,從而大大減小了圖像處理過程的運算量。經(jīng)過多年的研究,基于特征的圖像配準技術已經(jīng)取得了一定的研究成果,其主要步驟包括:圖像采集、特征提取、特征匹配、圖像變換。本文內(nèi)容以圖像配準的四個步驟為主線,首先概括介紹圖像配準技術的研究現(xiàn)狀、特點及應用領域,從而展示了圖像配準技術廣闊的應用前景。接著根據(jù)不同的硬件采集設備介紹了圖像采集的兩種方式以及圖像配準常用的數(shù)學模型。論文分析和總結了現(xiàn)有圖像匹配的主要方法、性能以及存在的問題,主要研究了兩個方面的內(nèi)容:(1)本文通過研究邊緣特征和點特征,著重介紹了sift(scale invariant feature transform)點的形成過程以及涉及的邊緣提

3、取技術,為了改善信噪比低的圖像的邊緣檢測效果,提出了一種基于多尺度多結構元素的數(shù)學形態(tài)學與canny算子相結合的邊緣檢測算法,得到的邊緣比較清晰,圖像的細節(jié)表現(xiàn)力和抗噪能力得到加強,為后續(xù)的特征提取、目標識別打下了良好的基礎;(2)在經(jīng)典的sift特征點的基礎上,針對不同的視角拍攝、不同的光照和明暗變換、尤其包含考驗匹配算法的相似結構多的情形的圖片,提出了一種基于網(wǎng)絡流的特征優(yōu)化匹配算法-最小費用k流算法(mkp)。該算法綜合利用了sift特征點的方向信息和梯度信息,采用本文改進的匹配度量函數(shù)衡量特征點之間的相似性來作為流網(wǎng)絡上的費用,借助于最小費用最大流的算法求出特征點之間的最優(yōu)匹配點對,根

4、據(jù)弧上的匹配度量比值和雙向檢查約束的方法來剔除偽匹配對,得到兩幅圖像的優(yōu)化匹配點對,匹配準確率比一些經(jīng)典算法提高了14%,使得變換系數(shù)更加精確,圖像配準效果增強。實驗證明,mkp算法準確率高,魯棒性強,具有較高的使用價值?！居⑽恼?image registration, is a process to match two or more images of the same scene taken at different times, from different viewpoint, or by different sensors, with the development of th

5、e computer vision, which is widely required in remote sensing, in medicine, in 3d reconstruction,and in computer vision (target location, automatic quality control), and so on. the methods of image registration can be classified into two categories: the intensity-based matching approach and the feat

6、ure-based matching approach. however, as one of the most common methods, the largest advantage of the feature-based approach is its ability of translating the analysis of the whole image into its features that contains the feature point, the feature curve, etc. and as a result, it speeds up the imag

7、e process. after many years of research, the feature-based image registration technology has made some achievements, the majority of the registration methods consist of four steps: image acquisition, feature detection, feature matching and image resampling and transformation.this paper is arranged a

8、ccording to the above four steps. firstly, recent research, characteristics and application areas of image registration are discribed, thereby demonstrating its wide application prospect. then image acquisition in two ways by adopting different hardware equipment is discussed, as well as common imag

9、e transform model. finally, there are two aspects which are we focused on: (1) the process of creating sift (scale invariant feature transform) descriptors and edge detection technology involved are emphatically given by studying edge and point features, then, in order to improve edge detection resu

10、lt on the low snr (signal to noise ratio) image, a mathematical morphology of multi-structure and multi-scale element and canny algorithm is introduced. so the performance of the details and anti-noise ability are strengthened, the convenience for the following steps such as feature extract and obje

11、ct recognition is provided; (2) based on the sift feature extract, an algorithm using network flow to obtain optimal feature matching-min cost k flow problem is put forward, which utilizes the direction and gradient information of sift and the improved matching cost function for measuring the simila

12、rity on the network flow, with minimum cost maximum flow algorithm to derive the global optimal matching, under the arc ratio of the matching measure and the bi-directional check constraints to remove pseudo matching. eventually, optimal matching will be achieved. as a result, about 14 percent impro

13、vement can be obtained. experments on the test image sets demonstrate high accuracy, robustness, and more application. 【中文關鍵詞】 特征點檢測; 圖像配準; sift; mkp算法; 邊緣檢測 【英文關鍵詞】 feature detect; image registration; sift; mkp algorithm; edge detection 【論文目錄】摘要 4-5 abstract 5-6 1 引言 9-12 1.1 論文的研究意義 9 1.2 國內(nèi)外研究現(xiàn)狀

14、9-10 1.3 論文的主要工作 10-11 1.4 論文的組織結構 11-12 2 圖像配準的基礎知識 12-16 2.1 圖像配準的定義 12 2.2 圖像獲取 12-14 2.2.1 照相機拍攝 13 2.2.2 采集卡獲取圖像 13-14 2.3 圖像配準的數(shù)學模型 14-15 2.4 本章小結 15-16 3 圖像特征提取 16-31 3.1 邊緣特征檢測算法 16-22 3.1.1 roberts 邊緣檢測算子 16-17 3.1.2 sobel 邊緣檢測算子 17 3.1.3 prewitt 邊緣檢測算子 17-18 3.1.4 canny 邊緣檢測算子 18-20 3.1.5 基于多尺度多結構元素的數(shù)學形態(tài)學與canny 算子相結合的邊緣檢測算法 20-21 3.1.6 改進的邊緣檢測算法驗證 21-22 3.2 點特征檢測算法 22-29 3.2.1 sift 特征點 23-29 3.3 本章小結 29-31 4 基于sift 特征點的匹配算法 31-39 4.1 特征匹配算法 31-36