發(fā)布時(shí)間：2018-05-09 18:57

  本文選題：多視圖三維重建 + 種子點(diǎn)提取��； 參考：《華中師范大學(xué)》2017年博士論文

【摘要】：多視圖三維重建直接從多幅二維圖像中恢復(fù)場(chǎng)景的三維結(jié)構(gòu),是計(jì)算機(jī)視覺一個(gè)研究熱點(diǎn),在工業(yè)檢測(cè)、逆向工程、城市規(guī)劃、文物與遺跡保護(hù)和展示等眾多領(lǐng)域有重要的應(yīng)用價(jià)值。隨著智能手機(jī)和高分辨、低成本圖像傳感器的大規(guī)模普及,表現(xiàn)出廣闊的應(yīng)用前景。近二十多年來出現(xiàn)了許多基于多視圖的三維重建算法,這些算法可大致分為三類:基于體積的算法、基于深度圖的算法和基于特征點(diǎn)生長(zhǎng)的算法。要完美處理現(xiàn)實(shí)中各種復(fù)雜情況,如表面的快速起伏變化、細(xì)小結(jié)構(gòu)、微弱紋理特征、遮擋效應(yīng)等,以達(dá)到更高的重建精度和重建完整度,同時(shí)保證高的重建效率,現(xiàn)有方法仍需改進(jìn)提高。本文提出了一種新的基于空間一致生長(zhǎng)的多視圖三維重建算法,它基于特征點(diǎn)生長(zhǎng),但對(duì)傳統(tǒng)基于特征點(diǎn)生長(zhǎng)的多視圖三維重建算法的整體框架進(jìn)行了拓展,在現(xiàn)有三個(gè)環(huán)節(jié)(稀疏種子點(diǎn)提取、生長(zhǎng)和濾波)基礎(chǔ)上,增加了一個(gè)新的環(huán)節(jié):利用已生長(zhǎng)完畢的點(diǎn)進(jìn)行有條件的初始值矯正,同時(shí)還對(duì)現(xiàn)有三個(gè)環(huán)節(jié)進(jìn)行了更新改進(jìn),取得了明顯效果。全文工作和創(chuàng)新點(diǎn)總結(jié)如下。在稀疏種子點(diǎn)提取環(huán)節(jié),提出了一種新的基于DAISY描述符的稀疏種子點(diǎn)提取方法。傳統(tǒng)SFM方法通常提取每幅圖像的SIFT特征點(diǎn),通過特征點(diǎn)匹配提取稀疏種子點(diǎn),往往由于匹配錯(cuò)誤或失敗導(dǎo)致種子點(diǎn)質(zhì)量降低或數(shù)目減少。本文對(duì)每個(gè)特征點(diǎn)采用高性能DAISY描述符進(jìn)行描述,然后沿對(duì)極線搜索與其DAISY特征最相似的點(diǎn),提高了稀疏種子點(diǎn)的數(shù)量和精度,改變了傳統(tǒng)的在有限的特征點(diǎn)之間直接進(jìn)行匹配的方式。為了保證上述方法順利實(shí)施,本算法提出了一系列配套措施,如通過最佳選圖和少數(shù)特征點(diǎn)的匹配,采用隨機(jī)抽樣一致性方法計(jì)算圖像對(duì)之間的基礎(chǔ)矩陣。再如,在多幅圖中采用DAISY特征描述符沿對(duì)極線進(jìn)行搜索,利用奇異值分解的方法求解在所有圖中均匹配成功的點(diǎn)的對(duì)應(yīng)空間位置,避免了單幅圖搜索可能存在的不確定性,同時(shí)利用重投影誤差濾除大誤差的點(diǎn)。最后,根據(jù)有條件的雙重二次曲面擬合來近似真實(shí)物體表面,求取種子點(diǎn)初始方向。以上種子點(diǎn)的提取和后續(xù)空間一致生長(zhǎng)都是在多層圖像金字塔上進(jìn)行的,進(jìn)一步提高了算法效率和成功率。在生長(zhǎng)環(huán)節(jié),提出了一種空間一致生長(zhǎng)策略。傳統(tǒng)方法需要依賴參考圖來尋找下一個(gè)生長(zhǎng)點(diǎn),即生長(zhǎng)點(diǎn)的初始位置和方向是在一個(gè)局部坐標(biāo)系確定的,隨著參考圖的更換,局部坐標(biāo)系也隨之發(fā)生改變。另外為了避免從質(zhì)量不高的種子點(diǎn)生長(zhǎng)出更低精度乃至錯(cuò)誤的點(diǎn),傳統(tǒng)算法對(duì)種子點(diǎn)進(jìn)行排序,優(yōu)先從最優(yōu)種子點(diǎn)進(jìn)行生長(zhǎng),這種串行算法限制了其計(jì)算效率。本文提出的算法從所有種子點(diǎn)出發(fā)無差別地同步向外擴(kuò)張生長(zhǎng),非常適用于并行計(jì)算;同時(shí)在一個(gè)固定的世界坐標(biāo)系內(nèi),直接從每個(gè)種子點(diǎn)現(xiàn)有空間位置出發(fā),沿其切平面確定生長(zhǎng)點(diǎn)的空間初始位置,不需要間接依靠圖像尋找生長(zhǎng)點(diǎn),也不需要在每幅圖像上時(shí)刻記錄哪些點(diǎn)已經(jīng)生長(zhǎng)完畢,有利于節(jié)省存儲(chǔ)空間。進(jìn)一步通過后續(xù)初始值矯正與錯(cuò)誤點(diǎn)濾除等措施,本算法有效防止了從質(zhì)量不高的種子點(diǎn)生長(zhǎng)出更多低精度乃至錯(cuò)誤的點(diǎn),保證了每個(gè)空間點(diǎn)生長(zhǎng)的相對(duì)獨(dú)立性。另外在生長(zhǎng)優(yōu)化之前,從眾多視圖中挑選最佳主、副圖參與優(yōu)化,在優(yōu)化過程中及時(shí)更新最佳主、副圖以提高重建效率、精度和完整度,而傳統(tǒng)的方法在生長(zhǎng)點(diǎn)優(yōu)化過程中通常不更換主圖。同時(shí),根據(jù)物體表面的紋理強(qiáng)弱自適應(yīng)地調(diào)整各種生長(zhǎng)參數(shù),如窗口大小和圖像金字塔層次等,以提高重建完整度。在新增加的初值矯正環(huán)節(jié),提出了有條件的雙重二次曲面擬合方法,根據(jù)已經(jīng)生長(zhǎng)完畢的點(diǎn)來擬合真實(shí)物體表面,進(jìn)而對(duì)初始值進(jìn)行矯正。從理論上來說,每個(gè)點(diǎn)的生長(zhǎng)都是獨(dú)立,即與其他區(qū)域是否已經(jīng)重建完畢無關(guān)。但在實(shí)際中,生長(zhǎng)點(diǎn)的初值來源于鄰近種子點(diǎn),如果初始值離真實(shí)表面較遠(yuǎn),可能導(dǎo)致無法收斂或收斂到一個(gè)局部極小。通過初始值矯正可以有效地提高收斂速度和精度,為此首先判斷鄰域點(diǎn)是否足夠稠密且以生長(zhǎng)點(diǎn)位中心,如果條件滿足,則進(jìn)行第一次擬合,刪除大誤差點(diǎn)后進(jìn)行第二次擬合,從而保證所擬合的曲面充分接近真實(shí)表面。接下來將生長(zhǎng)點(diǎn)投影到擬合曲面,即可實(shí)現(xiàn)初始值矯正。在濾波環(huán)節(jié),設(shè)計(jì)了三個(gè)自適應(yīng)濾波器,分別根據(jù)光滑一致性、深度一致性和方向一致性原理,對(duì)誤差點(diǎn)進(jìn)行檢測(cè)濾波。一方面保證了誤差點(diǎn)的有效濾除,另一方面又避免了正確點(diǎn)的無辜刪除。在濾波過程中,為了排除局部曲率半徑、鄰域點(diǎn)密度、遮擋等因素的影響,首先進(jìn)行條件判斷,決定是否濾波;如果進(jìn)行濾波,則自適應(yīng)地調(diào)整濾波參數(shù)。本文對(duì)來源于Middlebury標(biāo)準(zhǔn)數(shù)據(jù)庫(kù)、DTU標(biāo)準(zhǔn)數(shù)據(jù)庫(kù)、VGG多視圖三維重建數(shù)據(jù)庫(kù)和我們自己拍攝的不同類型實(shí)際場(chǎng)景進(jìn)行了三維重建,均取得了較好的重建結(jié)果,證明本算法具有較好的穩(wěn)定性。與其他多視圖三維重建算法相比,本算法重建結(jié)果局部瑕疵與缺陷明顯減少。定量評(píng)估結(jié)果表明,無論重建精度與重建完整度,本算法都位列前茅,特別是明顯優(yōu)于同類基于特征點(diǎn)生長(zhǎng)的多視圖三維重建算法,證明本文對(duì)基于特征點(diǎn)生長(zhǎng)的多視圖三維重建算法整體框架所進(jìn)行的拓展,和對(duì)現(xiàn)有三個(gè)環(huán)節(jié)的改進(jìn),成效十分明顯。
[Abstract]:The 3D reconstruction of multi view 3D reconstruction directly from multiple two-dimensional images is a hot topic in computer vision. It has important application value in many fields, such as industrial detection, reverse engineering, city planning, cultural relics and relics protection and display. In the past more than 20 years, there have been many 3D reconstruction algorithms based on multi view. These algorithms can be roughly divided into three types: Based on volume based algorithms, algorithms based on depth map and algorithm based on feature point growth. Structure, weak texture feature, occlusion effect and so on, in order to achieve higher reconstruction precision and reconstruction integrity, while ensuring high reconstruction efficiency, the existing methods still need to be improved. In this paper, a new multi view 3D reconstruction algorithm based on spatial uniform growth is proposed. It is based on the feature point growth, but the traditional growth based on feature points is much more. The overall frame of the 3D reconstruction algorithm is expanded. On the basis of three existing links (sparse seed extraction, growth and filtering), a new link is added: using the finished points to correct the conditional initial values, and the existing three rings are updated and improved, and the full text work has been achieved. In sparse seed point extraction, a new sparse seed point extraction method based on DAISY descriptors is proposed. The traditional SFM method usually extracts the SIFT feature points of each image, and extracts sparse seed points through feature point matching, often resulting in the quality reduction or number of seed points due to matching errors or failures. In this paper, a high performance DAISY descriptor is used for each feature point to be described, and then the number and accuracy of the sparse seed point are improved along the polar line search and the most similar point of its DAISY feature, and the traditional matching method between the finite feature points is changed. A series of matching measures, such as the matching of the best selection map and a few feature points, are used to calculate the base matrix between the image pairs by random sampling consistency method. Again, the DAISY feature descriptors are used to search the polar lines in the multiple images, and the method of singular value decomposition is used to solve the corresponding points that have been matched in all graphs. Space position avoids the possible uncertainty of single map search, and uses the re projection error to filter the point of large error. Finally, the initial direction of the seed point is obtained by using the conditional double two order surface fitting to approximate the real object surface. The extraction of the above seed points and the consistent growth of the following space are all in the pyramid of multi-layer images. On the tower, the efficiency and success rate of the algorithm are further improved. In the growth link, a spatial uniform growth strategy is proposed. The traditional method needs to rely on the reference graph to find the next growth point, that is, the initial position and direction of the growth point are determined in a local coordinate system. With the change of the reference map, the local coordinate system is also followed. In addition, in order to avoid the points of lower precision and error from the poor quality seed points, the traditional algorithms sort the seed points and give priority to the growth of the best seed points. This serial algorithm restricts the efficiency of calculation. The algorithm proposed in this paper extends from all kinds of subpoints to expand and grow synchronously from all kinds of subpoints. It is very suitable for parallel computing; at the same time, in a fixed world coordinate system, starting from the existing space position of each seed point, the initial position of the growth point is determined along its tangent plane. It does not need to rely on the image to find the growth point indirectly, and it does not need to record which points have been grown on each image. It is beneficial to the festival. This algorithm effectively prevents more low precision and error points from poor quality seed points, and ensures the relative independence of each space point growth. In addition, before optimization of growth, select the best owner from many views and participate in the sub graph. Optimization, in the process of optimization, the best master is updated in time to improve the efficiency, accuracy and integrity of the reconstruction, while the traditional method usually does not replace the main graph in the process of growth point optimization. At the same time, according to the texture strength of the surface, the growth parameters, such as the size of the window and the Pyramid level of the image, are adjusted to improve the reconstruction. In the newly added initial value correction link, a conditional double two surface fitting method is proposed to fit the surface of the real object based on the points that have been grown, and then correct the initial value. In theory, the growth of each point is independent, that is, it has nothing to do with the reconstruction of other regions. But in practice, The initial value of the growth point is derived from the adjacent seed point. If the initial value is far away from the real surface, it may lead to the failure to converge or converge to a local minimum. The convergence speed and accuracy can be effectively improved by the initial value correction. The first fitting, after the deletion of the large error points, second times of fitting, so as to ensure that the fitted surface is fully close to the true surface. Then the growth point is projected to the fitting surface, and the initial value can be corrected. In the filtering link, three adaptive filters are designed, which are based on smooth consistency, depth consistency and direction consistency, respectively. On the one hand, the error point is effectively filtered and the innocent deletion of the correct point is avoided. In the filtering process, in order to exclude the influence of the local curvature radius, the neighborhood point density, the occlusion and so on, the filter is determined first, and the filter is adjusted adaptively if the filtering is carried out. In this paper, we reconstruct the 3D reconstruction from the Middlebury standard database, the DTU standard database, the VGG multi view 3D reconstruction database and the different types of actual scenes we have taken. The results show that the algorithm has good stability. Compared with other multi view 3D reconstruction algorithms, The results of quantitative evaluation show that both the reconstruction accuracy and the reconstruction integrity are among the best, especially the multi view 3D reconstruction algorithm based on the feature point growth of the same kind, which proves the overall framework of the multi view 3D reconstruction algorithm based on the characteristic point growth. The expansion and the improvement of the existing three links are very effective.

【學(xué)位授予單位】：華中師范大學(xué)
【學(xué)位級(jí)別】：博士
【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5條

1 郭復(fù)勝;高偉;;基于輔助信息的無人機(jī)圖像批處理三維重建方法[J];自動(dòng)化學(xué)報(bào);2013年06期

2 王俊;朱利;;基于圖像匹配-點(diǎn)云融合的建筑物立面三維重建[J];計(jì)算機(jī)學(xué)報(bào);2012年10期

3 史利民;郭復(fù)勝;胡占義;;利用空間幾何信息的改進(jìn)PMVS算法[J];自動(dòng)化學(xué)報(bào);2011年05期

4 張峰;史利民;孫鳳梅;胡占義;;一種基于圖像的室內(nèi)大場(chǎng)景自動(dòng)三維重建系統(tǒng)[J];自動(dòng)化學(xué)報(bào);2010年05期

5 龍甫薈,鄭南寧;計(jì)算機(jī)視覺模型的研究與發(fā)展[J];信息與控制;1997年02期

相關(guān)博士學(xué)位論文 前1條

1 李兆歆;多視角立體三維重建方法研究[D];哈爾濱工業(yè)大學(xué);2016年

，

本文編號(hào)：1867030