《電子技術(shù)應(yīng)用》
您所在的位置:首頁 > 其他 > 設(shè)計(jì)應(yīng)用 > 運(yùn)動(dòng)恢復(fù)結(jié)構(gòu)生成點(diǎn)云的密度調(diào)控方法
運(yùn)動(dòng)恢復(fù)結(jié)構(gòu)生成點(diǎn)云的密度調(diào)控方法
2020年電子技術(shù)應(yīng)用第9期
蒙 浩1,,2,李自勝1,2
1.西南科技大學(xué) 制造科學(xué)與工程學(xué)院,,四川 綿陽621010,; 2.西南科技大學(xué) 制造過程測(cè)試技術(shù)省部共建教育部重點(diǎn)實(shí)驗(yàn)室,,四川 綿陽621010
摘要: 針對(duì)序列圖像生成點(diǎn)云與激光掃描點(diǎn)云密度不一致問題,,提出了一種基于運(yùn)動(dòng)恢復(fù)結(jié)構(gòu)生成點(diǎn)云的密度調(diào)控方法。該方法首先比較序列圖像生成的點(diǎn)云與目標(biāo)點(diǎn)云的密度以確定密度增大減小方向;其次提取圖像特征點(diǎn),,根據(jù)圖像特征點(diǎn)與對(duì)應(yīng)生成點(diǎn)云的空間映射關(guān)系設(shè)定單元格大小,,對(duì)圖像特征點(diǎn)區(qū)域均勻網(wǎng)格劃分,以距各單元格中心距離最近的特征點(diǎn)來表示其所屬單元格中的所有點(diǎn),,實(shí)現(xiàn)密度減小,,最小二乘擬合插值各單元格中的特征點(diǎn)實(shí)現(xiàn)密度增大;然后采用KLT特征點(diǎn)跟蹤算法得到匹配點(diǎn),;最后根據(jù)匹配點(diǎn)對(duì)生成圖像點(diǎn)云,。實(shí)驗(yàn)結(jié)果表明,該方法通過調(diào)整單元格大小或插值步長來控制圖像生成點(diǎn)云密度,,并取得了較好的調(diào)控效果。
中圖分類號(hào): TN249,;TP391.7
文獻(xiàn)標(biāo)識(shí)碼: A
DOI:10.16157/j.issn.0258-7998.191282
中文引用格式: 蒙浩,,李自勝. 運(yùn)動(dòng)恢復(fù)結(jié)構(gòu)生成點(diǎn)云的密度調(diào)控方法[J].電子技術(shù)應(yīng)用,2020,,46(9):88-93,,97.
英文引用格式: Meng Hao,Li Zisheng. Density control method of point cloud generated by structure from motion[J]. Application of Electronic Technique,,2020,,46(9):88-93,97.
Density control method of point cloud generated by structure from motion
Meng Hao1,,2,,Li Zisheng1,2
1.School of Manufacturing Science and Engineering,,Southwest University of Science and Technology,,Mianyang 621010,China,; 2.Key Laboratory of Testing Technology for Manufacturing Process,,Southwest University of Science and Technology, Mianyang 621010,,China
Abstract: Aiming at the inconsistency between the density of point cloud generated by sequence images and that generated by laser scanning,a density control method based on structure from motion is proposed.The method firstly compares the density of the point cloud generated by the sequence image with the density of the target point cloud to determine whether to increase or decrease the density of the point cloud generated by the sequence image. Secondly,the image feature points are extracted,and the cell size is set according to the spatial mapping relationship between the image feature points and the corresponding generated point cloud. The feature points closest to the center of each cell are used to represent all points in the cell to which they belong,so that the density is reduced,and the feature points in each cell are interpolated by least squares fitting to increase the density. Then,,KLT feature point tracking algorithm is used to get matching points. Finally,the image point cloud is generated based on the matching point pairs.The experimental results show that this method can control the density of point cloud generated by the image by adjusting the cell size or interpolation step size,achieving a good control effect.
Key words : 3D reconstruction of sequence images,;density regulation,;structure from motion method;feature point area meshing

0 引言

    點(diǎn)云密度匹配是不同點(diǎn)云源數(shù)據(jù)融合的基礎(chǔ),,在點(diǎn)云數(shù)據(jù)配準(zhǔn)及孔洞修復(fù)等方面有著重要作用?,F(xiàn)有點(diǎn)云密度調(diào)控方法主要是通過精簡方法來達(dá)到減小點(diǎn)云密度的目的。李仁忠[1]等通過輸入點(diǎn)云數(shù)據(jù)創(chuàng)建一個(gè)三維體素柵格,用每個(gè)體素柵格的重心來近似顯示整個(gè)體素柵格中的所有點(diǎn),,從而達(dá)到點(diǎn)云精簡的目的,,該方法在充分保留點(diǎn)云幾何特征的前提下,精簡結(jié)果比較均勻,。吳祿慎[2]等提出了一種改進(jìn)的重采樣算法,,將點(diǎn)云數(shù)據(jù)劃分為特征區(qū)域和平坦區(qū)域,對(duì)特征區(qū)域進(jìn)行曲率采樣,,對(duì)平坦區(qū)域進(jìn)行均勻采樣,,該方法既保持了曲率采樣保留模型細(xì)節(jié)的特點(diǎn),又保留了均勻采樣避免出現(xiàn)空白區(qū)域的優(yōu)點(diǎn),。袁小翠[3]等提出一種特征保持的點(diǎn)云數(shù)據(jù)精簡方法,,首先對(duì)點(diǎn)云K均值聚類,然后估計(jì)點(diǎn)云法矢和候選特征點(diǎn),,將包含特征點(diǎn)的聚類細(xì)分為多個(gè)子類,,最后基于自適應(yīng)均值漂移法對(duì)數(shù)據(jù)進(jìn)行分類,各聚類中心的集合即為精簡結(jié)果,,該方法能夠較好地保留原始曲面的幾何特征,。SHI B Q[4]等提出了自適應(yīng)K均值精簡法,對(duì)點(diǎn)云K均值聚類后根據(jù)類內(nèi)法矢偏差是否大于給定閾值,,將聚類分為兩類,,迭代判斷各類內(nèi)法矢偏差直至小于給定閾值,最后保留平坦區(qū)域的聚類中心以及非平坦區(qū)域法矢偏差最大的點(diǎn),,該方法對(duì)于噪聲過于敏感,。LI H[5]等提出一種基于法向量標(biāo)準(zhǔn)差的點(diǎn)云精簡算法,對(duì)點(diǎn)云數(shù)據(jù)采樣后進(jìn)行正態(tài)分布計(jì)算,,通過相鄰點(diǎn)之間的法向角計(jì)算特征點(diǎn)之間的分離閾值,,在特征點(diǎn)之間逐步向下采樣,實(shí)現(xiàn)點(diǎn)云的自適應(yīng)精簡,,該方法可以較好地保留原模型的特點(diǎn)和形狀,。CHEN Y[6]等通過將點(diǎn)云數(shù)據(jù)劃分為特征區(qū)域與非特征區(qū)域,針對(duì)不同區(qū)域采用不同的精簡方法可以較好地保留細(xì)節(jié)特征,。在點(diǎn)云密度調(diào)控方法中,,鮮有增大點(diǎn)云密度方面的研究。

    本文基于運(yùn)動(dòng)恢復(fù)結(jié)構(gòu)法[7](Structure From Motion,,SFM),,從二維特征點(diǎn)的密度調(diào)控出發(fā),對(duì)兩幅序列圖像生成點(diǎn)云的密度進(jìn)行調(diào)控,。密度調(diào)控既可以減小點(diǎn)云密度,,精簡點(diǎn)云數(shù)據(jù),,又可以增大點(diǎn)云密度,使細(xì)節(jié)信息得到加強(qiáng),。在本文中,,二維特征點(diǎn)的密度用平面特征點(diǎn)的平均間距來度量,點(diǎn)云密度用空間點(diǎn)云的平均間距來度量,。




本文詳細(xì)內(nèi)容請(qǐng)下載:http://forexkbc.com/resource/share/2000002989




作者信息:

蒙  浩1,,2,李自勝1,,2

(1.西南科技大學(xué) 制造科學(xué)與工程學(xué)院,,四川 綿陽621010;

2.西南科技大學(xué) 制造過程測(cè)試技術(shù)省部共建教育部重點(diǎn)實(shí)驗(yàn)室,,四川 綿陽621010)

此內(nèi)容為AET網(wǎng)站原創(chuàng),,未經(jīng)授權(quán)禁止轉(zhuǎn)載。