本文選題：行人檢測 + 雙目立體視覺��； 參考：《儀器儀表學(xué)報》2017年11期

【摘要】：針對單目視覺行人檢測無法獲得深度信息從而導(dǎo)致冗余信息較多、檢測效率和準(zhǔn)確度存在局限性的問題,首先,在圖像的預(yù)處理階段提出了一種利用雙目立體視覺產(chǎn)生的視差信息優(yōu)化分析來簡化復(fù)雜場景的動態(tài)規(guī)劃棒狀像素場景(stixel-world)表達方式;然后,在行人目標(biāo)檢測階段,對傳統(tǒng)HOG特征中block尺度進行分析、降維,采用Fisher準(zhǔn)則篩選得到了適用于道路環(huán)境下的多尺度HOG(multi-HOG)特征,將Multi-HOG特征與LUV顏色通道特征進行融合,最后采用交叉核支持向量機(hikSVM)分類器對行人目標(biāo)分類。實驗結(jié)果表明,采用改進過后的Stixel-world算法用于圖像預(yù)處理極大地減少了計算時間。縮小了行人檢測的候選區(qū)域,基于特征融合和hik-SVM的目標(biāo)檢測算法在保證檢測準(zhǔn)確度的前提下,具有較好的實時性和魯棒性。
[Abstract]:In view of the problem that the pedestrian detection of monocular vision can not obtain depth information, which leads to more redundant information, the detection efficiency and accuracy are limited. In the stage of image preprocessing, an optimized analysis of parallax information generated by binocular stereo vision is proposed to simplify the expression of dynamic programming rod-like pixel scene in complex scene, and then, in the stage of pedestrian target detection, Based on the analysis of block scale in traditional HOG features and dimension reduction, the multi-scale hog multi-hog features suitable for road environment are selected by Fisher criterion, and the Multi-HOG features and LUV color channel features are fused. Finally, cross kernel support vector machine (SVM) classifier is used to classify pedestrian targets. The experimental results show that the improved Stixel-world algorithm can greatly reduce the computing time for image preprocessing. The candidate area of pedestrian detection is reduced and the target detection algorithm based on feature fusion and hik-SVM has better real-time and robustness on the premise of ensuring detection accuracy.
【作者單位】： 湖南大學(xué)汽車車身先進設(shè)計制造國家重點實驗室;
【基金】：國家自然科學(xué)基金(51475153) 深圳市科技計劃(JCYJ20160530193357681)項目資助
【分類號】：TP391.41;U463.6

【相似文獻】

相關(guān)碩士學(xué)位論文 前2條

1 孫維毅;基于雙目立體視覺的自主資源勘探車輛環(huán)境識別技術(shù)研究[D];吉林大學(xué);2014年

2 侯能干;基于特征融合和多核學(xué)習(xí)的行人檢測方法研究[D];合肥工業(yè)大學(xué);2014年

，

本文編號：1905075