《電子技術(shù)應(yīng)用》
您所在的位置:首頁 > 其他 > 設(shè)計應(yīng)用 > 基于優(yōu)化多視角圖像采集的點云分類
基于優(yōu)化多視角圖像采集的點云分類
2021年電子技術(shù)應(yīng)用第10期
何瑞函1,2,蔡 勇1,2,張建生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.制造過程測試技術(shù)省部共建教育部重點實驗室,四川 綿陽621010
摘要: 基于二維多視角3D識別方法中,可使用多個2D投影圖表示三維模型特征信息,但不同視角投影圖像的特征不同,神經(jīng)網(wǎng)絡(luò)對其學(xué)習效率也有所差異。卷積神經(jīng)網(wǎng)絡(luò)能夠映射圖像的特征,可用此方法分析這個問題?;旌弦暯菙?shù)據(jù)集分析不同視角投影特征在卷積神經(jīng)網(wǎng)絡(luò)中的重要性,根據(jù)重要性的不同優(yōu)化混合視角數(shù)據(jù)集的采集密度。最終實驗結(jié)果表明,不同視角產(chǎn)生的二維圖像分類準確率不一樣,其中俯視角度投影的分類準確率最差,優(yōu)化后的數(shù)據(jù)集在相同神經(jīng)網(wǎng)絡(luò)模型下達到了最優(yōu)分類準確率。
中圖分類號: TP391.7
文獻標識碼: A
DOI:10.16157/j.issn.0258-7998.200892
中文引用格式: 何瑞函,蔡勇,張建生. 基于優(yōu)化多視角圖像采集的點云分類[J].電子技術(shù)應(yīng)用,2021,47(10):82-85.
英文引用格式: He Ruihan,Cai Yong,Zhang Jiansheng. Point cloud classification based on optimized multi-view image acquisition[J]. Application of Electronic Technique,2021,47(10):82-85.
Point cloud classification based on optimized multi-view image acquisition
He Ruihan1,2,Cai Yong1,2,Zhang Jiansheng1,2
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,Mianyang 621010,China
Abstract: In the 3D recognition method based on 2D multi-perspective, multiple 2D projected images can be used to represent the feature information of 3D model. However, the features of projected images from different perspectives are different, and the learning efficiency of the neural network is also different. The convolutional neural network can map the features of images, and this method can be used to analyze this problem. The importance of projection features of different perspectives in the convolutional neural network was analyzed in the mix-view data set, and the collection density of the mix-view data set was optimized according to the different importance. The final experimental results show that the classification accuracy of 2D images generated from different perspectives is different, among which the classification accuracy of overhead projection is the worst, and the optimized data set achieves the optimal classification accuracy in the same neural networks model.
Key words : 3D point cloud;multi-view image;convolution neural network;image classification

0 引言

    隨著激光掃描技術(shù)的發(fā)展,點云作為最能表現(xiàn)物體三維特征的數(shù)據(jù)深受研究者熱愛。深度學(xué)習、卷積神經(jīng)網(wǎng)絡(luò)(Convolutional Neural Network,CNN)[1-2]近幾年引領(lǐng)計算機視覺領(lǐng)域的研究趨勢,并且CNN網(wǎng)絡(luò)在二維圖像分類與識別上顯得高效。點云在空間中的無序性、旋轉(zhuǎn)不變性、非結(jié)構(gòu)化數(shù)據(jù),導(dǎo)致其不能直接作為CNN網(wǎng)絡(luò)的輸入[3]。因此,使用深度學(xué)習對點云進行研究的方法有3種:多視圖[4]、體素法[5]、直接對點云[6-7]

    基于二維多視角3D識別的方法,本文通過優(yōu)化采集數(shù)據(jù)集的方式提升CNN神經(jīng)網(wǎng)絡(luò)[8]的分類效果。本文對點云模型進行不同視角投影,得到多組2D圖像數(shù)據(jù)集。首先用多個VGG16卷積模型[9-10]提取單獨視角數(shù)據(jù)集,得到多個映射了圖像特征的卷積神經(jīng)網(wǎng)絡(luò)模型;然后將多個包含特征的VGG16模塊加上自定義層后“并聯(lián)”連接分類層作為分析網(wǎng)絡(luò)模型,混合視角圖像數(shù)據(jù)集作為網(wǎng)絡(luò)輸入;最后通過分析多個特征提取模塊的權(quán)重,優(yōu)化多視角圖像的采集密度,提升二維多視角3D識別效率,即分類效果。




本文詳細內(nèi)容請下載:http://theprogrammingfactory.com/resource/share/2000003788




作者信息:

何瑞函1,2,蔡  勇1,2,張建生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.制造過程測試技術(shù)省部共建教育部重點實驗室,四川 綿陽621010)




wd.jpg

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