本文關(guān)鍵詞： 計(jì)算機(jī)視覺(jué) 視覺(jué)跟蹤 Boosting算法 GPU加速 目標(biāo)特征提取　出處：《華東師范大學(xué)》2017年碩士論文　論文類(lèi)型：學(xué)位論文

【摘要】：視覺(jué)跟蹤是計(jì)算機(jī)視覺(jué)中的一個(gè)重要領(lǐng)域,它在視頻監(jiān)控、運(yùn)動(dòng)分析和交通監(jiān)管等方面有廣泛的應(yīng)用。盡管目前有大量的文獻(xiàn)給出視覺(jué)跟蹤的解決方案,但由于目標(biāo)姿態(tài)變換、運(yùn)動(dòng)模糊、遮擋以及場(chǎng)景中光照變化等不利因素的存在,基于視覺(jué)的目標(biāo)跟蹤仍然是具有挑戰(zhàn)性的研究課題。把跟蹤看成目標(biāo)與背景的分類(lèi)問(wèn)題是解決視覺(jué)跟蹤的常見(jiàn)方法,它不需要建立復(fù)雜模型描述目標(biāo),而是找到區(qū)分目標(biāo)和背景的分類(lèi)器。Grabner等人提出的基于Boosting的在線目標(biāo)跟蹤算法是基于分類(lèi)的經(jīng)典算法,該算法通過(guò)隨機(jī)位置的Haar-like特征在線訓(xùn)練弱分類(lèi)器用于選擇區(qū)分效果好的特征。本文嘗試使用多層次特征選擇和特征融合實(shí)現(xiàn)目標(biāo)跟蹤任務(wù),針對(duì)在線Boosting目標(biāo)跟蹤算法只對(duì)目標(biāo)區(qū)域內(nèi)位置特征作選擇的問(wèn)題,增加了濾波器類(lèi)型的選擇,提出了兩層級(jí)聯(lián)的Boosting改進(jìn)算法;在Boosting算法框架下選擇深度網(wǎng)絡(luò)中適合跟蹤的不同層次特征和不同維度特征;基于GPU的并行機(jī)制,加速兩層級(jí)聯(lián)的Boosting改進(jìn)算法。1、本文在Boosting跟蹤算法的基礎(chǔ)上提出兩層級(jí)聯(lián)的Boosting跟蹤方法。改進(jìn)方法通過(guò)諸多濾波器模板提取目標(biāo)局部特征,使用Boosting分別對(duì)目標(biāo)區(qū)域內(nèi)圖像小塊位置和它對(duì)應(yīng)的濾波器類(lèi)型進(jìn)行選擇,并且有效地融合兩種特征,提升了目標(biāo)跟蹤的準(zhǔn)確性。2、本文將深度神經(jīng)網(wǎng)絡(luò)中間各層的輸出作為特征圖譜輸入Boosting算法實(shí)現(xiàn)目標(biāo)跟蹤,目的是選擇適合跟蹤任務(wù)的高維特征。使用Boosting分別對(duì)深度神經(jīng)網(wǎng)絡(luò)中不同層次特征和不同維度特征進(jìn)行選擇,并在實(shí)驗(yàn)結(jié)果對(duì)比中找到適合目標(biāo)跟蹤的特征組合方式。3、本文針對(duì)提出的兩層級(jí)聯(lián)的Boosting跟蹤方法給出加速的方案�；贕PU的并行機(jī)制,將兩層級(jí)聯(lián)Boosting跟蹤方法中大量繁瑣的矩陣運(yùn)算進(jìn)行加速,提升跟蹤算法的速度,增大算法的可行性。
[Abstract]:Visual tracking is an important field in computer vision. It is widely used in video surveillance, motion analysis and traffic supervision. However, due to the target attitude change, motion blur, occlusion and scene changes in the light, and other adverse factors exist. Target tracking based on vision is still a challenging research topic. It is a common method to solve the problem of target and background classification, and it does not need to establish a complex model to describe the target. The online target tracking algorithm based on Boosting proposed by Grabner et al. Is a classical algorithm based on classification. This algorithm uses the Haar-like feature of random position to train the weak classifier to select the feature with good performance. This paper attempts to use multi-level feature selection and feature fusion to achieve target tracking task. Aiming at the problem that the online Boosting target tracking algorithm only selects the location characteristics in the target region, the filter type selection is added, and a two-layer cascade Boosting improved algorithm is proposed. Under the framework of Boosting algorithm, different level and dimension features suitable for tracking in depth network are selected. Based on the parallel mechanism of GPU, the improved Boosting algorithm of two-layer cascade is accelerated. Based on the Boosting tracking algorithm, a two-layer cascaded Boosting tracking method is proposed in this paper. The improved method extracts the local features of the target by a lot of filter templates. Boosting is used to select the location of the image block and the corresponding filter type in the target region, and the two features are fused effectively, which improves the accuracy of target tracking. 2. In this paper, the output of the middle layers of the depth neural network is used as the feature map input Boosting algorithm to achieve target tracking. The purpose of this paper is to select the high-dimensional features suitable for tracking tasks. Boosting is used to select the features of different levels and different dimensions in the depth neural network. And in the comparison of experimental results to find a suitable target tracking feature combination mode. 3, this paper proposes a two-layer cascaded Boosting tracking method to accelerate the scheme. Based on the parallel mechanism of GPU. A large number of complex matrix operations in the two-layer cascade Boosting tracking method are accelerated, the speed of the tracking algorithm is improved, and the feasibility of the algorithm is increased.
【學(xué)位授予單位】：華東師范大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類(lèi)號(hào)】：TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前4條

1 楊通鈺;彭國(guó)華;;基于NCC的圖像匹配快速算法[J];現(xiàn)代電子技術(shù);2010年22期

2 尹宏鵬;柴毅;匡金駿;陽(yáng)小燕;;一種基于多特征自適應(yīng)融合的運(yùn)動(dòng)目標(biāo)跟蹤算法[J];光電子.激光;2010年06期

3 李安平;敬忠良;胡士強(qiáng);;基于自適應(yīng)表面模型的概率視頻跟蹤算法[J];控制與決策;2007年01期

4 朱淼良,錢(qián)徽;自然景物中大氣退化模型的研究[J];計(jì)算機(jī)輔助設(shè)計(jì)與圖形學(xué)學(xué)報(bào);2001年09期

相關(guān)碩士學(xué)位論文 前1條

1 關(guān)勇;物聯(lián)網(wǎng)行業(yè)發(fā)展分析[D];北京郵電大學(xué);2010年

，

本文編號(hào)：1463462