本文關(guān)鍵詞： 夜間車(chē)輛檢測(cè) 閾值分割 Haar特征 車(chē)輛跟蹤 自動(dòng)評(píng)價(jià)　出處：《北京交通大學(xué)》2014年碩士論文　論文類(lèi)型：學(xué)位論文

【摘要】：車(chē)輛檢測(cè)和跟蹤是智能交通的一個(gè)重要內(nèi)容,也為“十二五規(guī)劃”智慧城市建設(shè)的首要目標(biāo)——平安城市提供關(guān)鍵技術(shù)。該領(lǐng)域涉及的運(yùn)動(dòng)目標(biāo)檢測(cè)和多目標(biāo)跟蹤技術(shù)一直是計(jì)算機(jī)視覺(jué)和機(jī)器學(xué)習(xí)的研究熱點(diǎn)。本文在前期調(diào)研工作的基礎(chǔ)上,基于車(chē)燈特征對(duì)夜間車(chē)輛進(jìn)行檢測(cè),并訓(xùn)練分類(lèi)器來(lái)降低虛警率,在此基礎(chǔ)上實(shí)現(xiàn)魯棒的夜間車(chē)輛跟蹤。利用自動(dòng)評(píng)價(jià)技術(shù)對(duì)夜間車(chē)輛跟蹤進(jìn)行自動(dòng)評(píng)價(jià),并根據(jù)結(jié)果提出了改進(jìn)的方案。 本文的主要工作有： 1、基于車(chē)燈特征的車(chē)輛檢測(cè)技術(shù)的研究。本文首先對(duì)比了多種車(chē)輛檢測(cè)和跟蹤的技術(shù),例如幀間差分法、背景差分法和自適應(yīng)閾值分割法。針對(duì)這些傳統(tǒng)方法只適用于白天、光照充足條件下的車(chē)輛檢測(cè)和跟蹤而不適用于夜間的問(wèn)題,提出了基于車(chē)燈的車(chē)輛檢測(cè)。在訓(xùn)練階段,采用針對(duì)車(chē)燈的Haar特征并融合多尺度的幾何、形狀特征進(jìn)行Adaboost分類(lèi)器訓(xùn)練。在測(cè)試階段,根據(jù)統(tǒng)計(jì)直方圖設(shè)定閾值,分割夜間圖像中的車(chē)燈。所有檢測(cè)到的車(chē)燈通過(guò)Adaboost分類(lèi)器分類(lèi)出正樣本,在此基礎(chǔ)上根據(jù)速度、幾何、形狀等特征的相似性建立數(shù)據(jù)關(guān)聯(lián)進(jìn)行多目標(biāo)跟蹤。 2、針對(duì)車(chē)輛跟蹤的自動(dòng)評(píng)價(jià)技術(shù)。對(duì)車(chē)輛跟蹤效果的評(píng)價(jià)是一個(gè)工作量極大的部分,為此,本文提出了一種自動(dòng)評(píng)價(jià)的方法來(lái)對(duì)車(chē)輛跟蹤的各項(xiàng)性能進(jìn)行自動(dòng)的評(píng)價(jià)。首先是介紹了一款半自動(dòng)標(biāo)注工具,來(lái)獲取真實(shí)的車(chē)輛跟蹤結(jié)果。利用真實(shí)的車(chē)輛跟蹤結(jié)果和車(chē)輛跟蹤算法生成的跟蹤結(jié)果通過(guò)自動(dòng)評(píng)價(jià)技術(shù)自動(dòng)生成各項(xiàng)評(píng)價(jià)指標(biāo)。這些評(píng)價(jià)指標(biāo)包括正確率、缺失率、誤判率和轉(zhuǎn)變率。各項(xiàng)評(píng)價(jià)指標(biāo)的數(shù)據(jù)顯示,本文提出的方法能夠?qū)?chē)輛進(jìn)行實(shí)時(shí)魯棒的跟蹤。
[Abstract]:Vehicle detection and tracking is an important part of the intelligent transportation system, but also for the primary objective of "12th Five-Year plan" wisdom City Construction -- safe city to provide key technology. Tracking technology of moving target detection in the fields and multi targets has been a research focus of computer vision and machine learning. This paper based on the previous research work. On the night, vehicle detection and classifier training based on the characteristics of light, to reduce the false alarm rate, to achieve robust on the basis of the night. On the night of the vehicle tracking vehicle tracking automatic evaluation using automatic evaluation technology, and put forward the improvement scheme according to the results.
The main work of this article is as follows:
1, study the characteristics of vehicle detection technology based on. This paper compares a variety of vehicle detection and tracking technology, such as the inter frame difference method, background difference method and adaptive threshold segmentation method. The traditional method is only suitable for daytime, adequate light under the condition of vehicle detection and tracking is not suitable for the night, the lights of the vehicle detection based on. During the training phase, the geometry for Haar lights and feature fusion of multi-scale, shape feature Adaboost classifier training. During the testing phase, according to the statistics histogram set the threshold segmentation in the image, the night lights. All the detected light by Adaboost classifier classification the positive samples, based on the speed, geometry, shape similarity and other characteristics of the establishment of Data Association for multiple target tracking.
2, according to the automatic evaluation technology of vehicle tracking. To evaluate the effect of vehicle tracking is a great part of a workload for this purpose, this paper presents a method for automatic evaluation to the evaluation of the performance of automatic vehicle tracking. At first it introduces a semi-automatic annotation tool to obtain real tracking results the use of vehicles. Vehicle tracking and generate results of real vehicle tracking algorithm based on tracking results through the automatic evaluation technology of automatic generation of evaluation indexes. These indexes include the correct rate, loss rate, error rate and transformation rate. The evaluation index data show that the proposed method can real-time robust vehicle tracking.

【學(xué)位授予單位】：北京交通大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類(lèi)號(hào)】：U495;TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前4條

1 彭春華;劉建業(yè);劉岳峰;晏磊;鄭江華;;車(chē)輛檢測(cè)傳感器綜述[J];傳感器與微系統(tǒng);2007年06期

2 安曉亞;孫群;肖強(qiáng);嚴(yán)薇;;一種形狀多級(jí)描述方法及在多尺度空間數(shù)據(jù)幾何相似性度量中的應(yīng)用[J];測(cè)繪學(xué)報(bào);2011年04期

3 李秋林;何家峰;;基于三幀差法和交叉熵閾值法的車(chē)輛檢測(cè)[J];計(jì)算機(jī)工程;2011年04期

4 李闖;丁曉青;吳佑壽;;一種改進(jìn)的AdaBoost算法——AD AdaBoost[J];計(jì)算機(jī)學(xué)報(bào);2007年01期

，

本文編號(hào)：1555347