本文選題：雙目視覺 + 立體匹配��； 參考：《浙江大學(xué)》2017年碩士論文

【摘要】：立體視覺及三維重建技術(shù)日益受到各行業(yè)的關(guān)注和重視,尤其是虛擬現(xiàn)實(shí)、增強(qiáng)現(xiàn)實(shí)及混合現(xiàn)實(shí)技術(shù)發(fā)展得如火如茶,立體匹配作為立體視差法三維重建技術(shù)的關(guān)鍵有著重要的研究價值。目前從基本的局部SAD及NCC相似性度量到全局能量的置信傳播及圖割方法衍生的立體匹配算法層出不窮,但對于遮擋問題及非子模能量函數(shù)的優(yōu)化仍是一項(xiàng)重大的挑戰(zhàn)。本文采用3D標(biāo)簽來描述各個像素的空間特征,在已有的馬爾可夫隨機(jī)場最大后驗(yàn)(MRF-MAP)能量代價模型下,采用3D標(biāo)簽的法向及切平面特征來建立二階先驗(yàn)的立體曲面。二階先驗(yàn)平滑項(xiàng)放松了前向平行平面的約束使得滿足斜向切平面先驗(yàn)時也不受到懲罰,使得水平及垂直方向鄰域的各像素可以相互關(guān)聯(lián)。針對遮擋及多重團(tuán)簇的非子模性難題,本文改造了非對稱圖模型來控制團(tuán)簇大小,通過增加可見性節(jié)點(diǎn)將所有可能的遮擋信息呈現(xiàn)在圖中,并將可見性節(jié)點(diǎn)的關(guān)聯(lián)邊整合到能量函數(shù)的數(shù)據(jù)項(xiàng)中。同時本文改進(jìn)了偽布爾函數(shù)多項(xiàng)式優(yōu)化(QPBO),為每個像素節(jié)點(diǎn)創(chuàng)立與之對立的非節(jié)點(diǎn),使得非子模的邊也可用mincut/max flow方法進(jìn)行分割,改進(jìn)的QPBOI-R方案能在不增加能量代價的情況下合理地更新不可標(biāo)記節(jié)點(diǎn),經(jīng)過多次迭代可得到近似最小代價的解。本文最主要的貢獻(xiàn)是提出了多尺度超像素候選標(biāo)簽圖(Multi-scale Super-pixel basedProposals,MSP)結(jié)構(gòu)及其更新方法,已有的候選圖更新方法有基于圖像分割的或隨機(jī)生成的。MSP通過隨機(jī)選取已修正的標(biāo)簽來更新超像素圖像的候選標(biāo)簽,主要是利用了多個尺度的超像素圖像可以在多種特征層面下更新標(biāo)簽而不影響邊緣處的視差突變。另外,將多個超像素候選與兩個棋盤格候選相結(jié)合構(gòu)成完整的候選標(biāo)簽圖,超像素候選表征著區(qū)域的特征信息而棋盤格保留了像素級別的特征信息,實(shí)驗(yàn)證明了 MSP結(jié)構(gòu)下的視差圖在均勻無紋理區(qū)域呈現(xiàn)平滑特性,而在邊緣處又呈現(xiàn)了深度不連續(xù)特性。
[Abstract]:Stereo vision and 3D reconstruction technology have attracted more and more attention from various industries, especially virtual reality, augmented reality and mixed reality technology. Stereo matching plays an important role in 3D reconstruction of parallax. At present, stereo matching algorithms derived from basic local SAD and NCC similarity measures to global energy confidence propagation and graph cutting methods are emerging in endlessly. However, it is still a major challenge for occlusion problems and optimization of nonsubmode energy functions. In this paper, 3D tags are used to describe the spatial features of each pixel. Under the existing Markov random field maximum posterior MRF-MAP-based energy cost model, the normal and tangent plane features of 3D tags are used to establish the second-order priori stereoscopic surfaces. The second-order priori smoothing term relaxes the constraint of the forward parallel plane so that the apriori of oblique tangent plane is not penalized so that the pixels in the horizontal and vertical direction neighborhood can be correlated with each other. In order to solve the non-submodule problem of occlusion and multiple clusters, the asymmetric graph model is modified to control the cluster size, and all possible occlusion information is presented in the graph by adding visibility nodes. The associated edges of the visibility nodes are integrated into the data items of the energy function. At the same time, we improve the pseudo-Boolean function polynomial optimization and create a non-node for each pixel node, so that the edges of non-submodules can also be segmented by mincut/max flow method. The improved QPBOI-R scheme can reasonably update the unlabeled nodes without increasing the energy cost. After several iterations, the approximate minimum cost solution can be obtained. The main contribution of this paper is to propose the multi-scale super-pixel based proposals MSPs structure and its updating method. The existing candidate image updating methods include. MSP, which is based on image segmentation or randomly generated, updates candidate labels of superpixel images by randomly selecting modified labels. The multi-scale super-pixel image can update the label at multiple feature levels without affecting the parallax mutation at the edge. In addition, a plurality of super-pixel candidates are combined with two checkerboard candidates to form a complete candidate tag map. The super-pixel candidate represents the feature information of the region, while the checkerboard retains the pixel level feature information. The experimental results show that the parallax image with MSP structure is smooth in the homogeneous texture-free region, and the depth discontinuity is shown at the edge of the parallax image.
【學(xué)位授予單位】：浙江大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP391.41
，

本文編號：2020668