汽車制動(dòng)系設(shè)計(jì)【含CATIA三維圖紙、說明書、開題報(bào)告】
畢 業(yè) 設(shè) 計(jì)(論 文)任 務(wù) 書 設(shè)計(jì)(論文)題目:汽車制動(dòng)系設(shè)計(jì) 學(xué)生姓名:任務(wù)書填寫要求1畢業(yè)設(shè)計(jì)(論文)任務(wù)書由指導(dǎo)教師根據(jù)各課題的具體情況填寫,經(jīng)學(xué)生所在專業(yè)的負(fù)責(zé)人審查、系(院)領(lǐng)導(dǎo)簽字后生效。此任務(wù)書應(yīng)在畢業(yè)設(shè)計(jì)(論文)開始前一周內(nèi)填好并發(fā)給學(xué)生。2任務(wù)書內(nèi)容必須用黑墨水筆工整書寫,不得涂改或潦草書寫;或者按教務(wù)處統(tǒng)一設(shè)計(jì)的電子文檔標(biāo)準(zhǔn)格式(可從教務(wù)處網(wǎng)頁上下載)打印,要求正文小4號(hào)宋體,1.5倍行距,禁止打印在其它紙上剪貼。3任務(wù)書內(nèi)填寫的內(nèi)容,必須和學(xué)生畢業(yè)設(shè)計(jì)(論文)完 的情況 一 , ,應(yīng) 經(jīng) 所在專業(yè) 系(院) 領(lǐng)導(dǎo)審 后 可 填寫。4任務(wù)書內(nèi) 學(xué)院 、 專業(yè) 名 的填寫,應(yīng)寫 文 ,不 寫 字 。學(xué)生的 學(xué)號(hào) 要寫 號(hào),不 寫 后2 或1 字。 5任務(wù)書內(nèi) 要 文 的填寫,應(yīng)按學(xué)院畢業(yè)設(shè)計(jì)(論文)currency1寫“的要求書寫。 6 fi fifl的填寫,應(yīng) 按 標(biāo)GB/T 740894 據(jù)和格式、 、fifl和”的要求,一用 字書寫。 200242fi 或 2002-04-02”。畢 業(yè) 設(shè) 計(jì)(論 文)任 務(wù) 書1畢業(yè)設(shè)計(jì)(論文)課題應(yīng)的目的: 制動(dòng)系統(tǒng)汽車用( 要)在汽車 ( 要車)一的, 其 行一 的 制制動(dòng)的一系 專 。其 用 行 的汽車按 的要求 行 制 車, 的汽車在各 下(在 ) 車, 下 行 的汽車 。 設(shè)計(jì)的 汽車制動(dòng)系, 制動(dòng)系 式制動(dòng) 制動(dòng) 系統(tǒng) 行設(shè)計(jì)和 。2畢業(yè)設(shè)計(jì)(論文)課題任務(wù)的內(nèi)容和要求(始 據(jù)、要求、工 要求 ): 設(shè)計(jì)的 汽車制動(dòng)系, 制動(dòng)系 式制動(dòng) 制動(dòng) 系統(tǒng) 行設(shè)計(jì)和 。 (1)制動(dòng)系統(tǒng) 計(jì) 制動(dòng) 設(shè)計(jì)(2)制動(dòng) 缸計(jì) 與 設(shè)計(jì)(3)制動(dòng) 布 設(shè)計(jì)(4)制動(dòng) 配計(jì) (5)用應(yīng)用軟 行建模 畢 業(yè) 設(shè) 計(jì)(論 文)任 務(wù) 書3 畢業(yè)設(shè)計(jì)(論文)課題 果的要求圖、實(shí)物 硬要求:1、畢業(yè)論文1萬字左右(并附 的 據(jù))2、文 資料譯文(附文)3000字4 要 文 : 1 許兆棠、劉永臣.汽車 造(下冊(cè))M.北京: 防工業(yè)出版社,2012.2 王望予.汽車設(shè)計(jì)第四版M.北京:機(jī)械工業(yè)出版社,2004.3 汽車工 手冊(cè)編輯委 會(huì).汽車工 手冊(cè).設(shè)計(jì)篇M.北京:人民通出版社,2001.4 汽車工 手冊(cè)編輯委 會(huì).汽車工 手冊(cè).制造篇M.北京:人民通出版社,2001.5 余志生.汽車?yán)碚摰?版M.北京:機(jī)械工業(yè)出版社,2009.6 泳龍.汽車制動(dòng)理論與設(shè)計(jì)M.北京: 防工業(yè)出版社,2005.7 劉惟.汽車制動(dòng)系統(tǒng)的 設(shè)計(jì)計(jì) M.北京:清華大學(xué)出版社,2004.8 許文娟.汽車制動(dòng)性 測(cè)試 系統(tǒng)的 與開發(fā)D.大連:大連理工大學(xué),2005.9 張才.余卓平、熊璐.制動(dòng)系統(tǒng)發(fā)展現(xiàn)狀 趨勢(shì)J.汽車 與開發(fā),2005.10 機(jī)械設(shè)計(jì)大典編委會(huì). 機(jī)械設(shè)計(jì)大典M.南昌:江西學(xué)出版社,2002.11 陳煥江.汽車檢測(cè)與診斷(上冊(cè))M.北京:機(jī)械工業(yè)出版社,2006.12 王維、劉建農(nóng)、何光里.汽車制動(dòng)性檢測(cè)M.北京:人民通出版社,2005.13 王韋、馬明星.汽車設(shè)計(jì)課 指導(dǎo)書M.北京: 電出版社,2009.14 劉惟.汽車設(shè)計(jì)M.北京:清華大學(xué)出版社,2001.15 張 .車輛制動(dòng)系統(tǒng)的發(fā)展現(xiàn)狀 趨勢(shì)淺J.農(nóng)業(yè)與,2009.16 務(wù)院發(fā)展 心產(chǎn)業(yè)經(jīng)濟(jì) , 汽車工 學(xué)會(huì),大眾汽車集團(tuán)( ), 汽車產(chǎn)業(yè)發(fā)展報(bào)告J.北京:社會(huì)學(xué)文 出版社,2009.畢 業(yè) 設(shè) 計(jì)(論 文)任 務(wù) 書5畢業(yè)設(shè)計(jì)(論文)課題工 計(jì)劃:2015.12.05-2016.01.15確選題,填寫審題;指導(dǎo)教師下發(fā)任務(wù)書,學(xué)生查閱課題 文 、資料,currency1寫開題報(bào)告。2016.01.16-2016.02.25提開題報(bào)告、文 資料 譯文、畢業(yè)設(shè)計(jì)(論文)大綱;開始畢業(yè)設(shè)計(jì)(論文)。2016.02.26-2016.04.15具體設(shè)計(jì)或 案實(shí),提畢業(yè)設(shè)計(jì)(論文)草稿,填寫 fl檢查。2016.04.16-2016.05.05完 論文或設(shè)計(jì)說明書、圖紙 材料,提畢業(yè)設(shè)計(jì)(論文)稿,指導(dǎo)老師審核。2016.05.06-2016.05.13提畢業(yè)設(shè)計(jì)紙質(zhì)文檔,學(xué)生準(zhǔn)備答辯;評(píng)閱教師評(píng)閱學(xué)生畢業(yè)設(shè)計(jì)(論文)。2016.05.13-2016.05.26根據(jù)學(xué)院統(tǒng)一安排, 行畢業(yè)設(shè)計(jì)(論文)答辯。所在專業(yè)審查意見: 通 負(fù)責(zé)人: 2016 1 22 fi畢 業(yè) 設(shè) 計(jì)(論 文)開 題 報(bào) 告 設(shè)計(jì)(論文)題目:汽車制動(dòng)系設(shè)計(jì) 學(xué)生姓名:開題報(bào)告填寫要求 1開題報(bào)告(含“文獻(xiàn)綜述”)作為畢業(yè)設(shè)計(jì)(論文)答辯委員會(huì)對(duì)學(xué)生答辯資格審查的依據(jù)材料之一。此報(bào)告應(yīng)在指導(dǎo)教師指導(dǎo)下,由學(xué)生在畢業(yè)設(shè)計(jì)(論文)工作前期內(nèi)完成,經(jīng)指導(dǎo)教師簽署意見及所在專業(yè)審查后生效;2開題報(bào)告內(nèi)容必須用黑墨水筆工整書寫或按教務(wù)處統(tǒng)一設(shè)計(jì)的電子文檔標(biāo)準(zhǔn)格式打印,禁止打印在其它紙上后剪貼,完成后應(yīng)及時(shí)交給指導(dǎo)教師簽署意見;3“文獻(xiàn)綜述”應(yīng)按論文的框架成文,并直接書寫(或打?。┰诒鹃_題報(bào)告第一欄目?jī)?nèi),學(xué)生寫文獻(xiàn)綜述的參考文獻(xiàn)應(yīng)不 15 (不 );4 期的填寫,應(yīng) 按 標(biāo)GB/T 740894 據(jù) 交 格式 交 期 時(shí) 的要求,一用 書寫?!?004 4 26 ”或“2004-04-26”。5 開題報(bào)告(文獻(xiàn)綜述)currency1按 “書寫, fi1.5fl。 畢 業(yè) 設(shè) 計(jì)(論文) 開 題 報(bào) 告 1 畢業(yè)設(shè)計(jì)(論文)題, 據(jù)所查的文獻(xiàn)資料,寫不 1000”的文獻(xiàn)綜述: 汽車生 ,汽車制動(dòng)系統(tǒng)在車 的 要的 。汽車車 及車 的 ,汽車制動(dòng)系統(tǒng)的工作 要求 。 汽車的 開,汽車的動(dòng) 系統(tǒng)生 的 。動(dòng) 系統(tǒng)的 要求汽車制動(dòng)系統(tǒng) 式 式生應(yīng)的 。汽車制動(dòng)系 汽車 上的一要系統(tǒng),它 制 汽車 動(dòng)的。車 在的 要 進(jìn)制動(dòng)操作,由 制動(dòng) 的好壞直接 系到交通,因此制動(dòng) 車 非常要的 之一。汽車制動(dòng)系統(tǒng)種類 多,傳統(tǒng)的制動(dòng)系統(tǒng) 式主要 機(jī)械式 氣動(dòng)式 液壓式 氣-液混式。它們的工作原理差不多,都 利用制動(dòng)在工作時(shí)產(chǎn)生的摩擦熱逐漸消耗車 所具 的動(dòng) , 達(dá)到車 制動(dòng)減,或者停車的目的。為 保證汽車,揮 的 ,制動(dòng)系統(tǒng)必須滿足:良好的制動(dòng) ;制動(dòng)效 的恒 ;制動(dòng)時(shí)的向穩(wěn) 好;制動(dòng)平順 好;操縱輕便 舒適;工作 ;環(huán)保;散熱 好, 天候使用。汽車制動(dòng)系統(tǒng) 由產(chǎn)生制動(dòng)作用的制動(dòng)器 制動(dòng)操縱系統(tǒng)組成。制動(dòng)器 制動(dòng)系 最主要的一部件, 汽車制動(dòng)系統(tǒng) 用 產(chǎn)生阻礙車 的 動(dòng)或 動(dòng)趨勢(shì)的 (制動(dòng) )的部件,其 輔助制動(dòng)系統(tǒng) 的緩。汽車上常用的制動(dòng)器都 利用不旋轉(zhuǎn) 件與旋轉(zhuǎn) 件工作 的摩擦而產(chǎn)生制動(dòng) 矩,稱為摩擦制動(dòng)器。 代汽車普遍采用的摩擦式制動(dòng)器的實(shí)際工作 整制動(dòng)系 最復(fù)雜 最不穩(wěn) 的部件,因此 進(jìn)制動(dòng)器機(jī) 解決制 其 的突問題具 非常要的意義。目前,汽車所用的制動(dòng)器按 摩擦副 旋轉(zhuǎn) 件的不同, 分為鼓式制動(dòng)器 式制動(dòng)器兩 類。前者摩擦副 的旋轉(zhuǎn) 件為制動(dòng)鼓,其工作 為圓柱;后者的旋轉(zhuǎn) 件則為圓 狀的制動(dòng) , 其端為工作 。由 鼓式制動(dòng)器造價(jià)便宜, 在 多企業(yè)都采用它。制動(dòng)操縱系統(tǒng) 產(chǎn)生制動(dòng)動(dòng)作 控制制動(dòng)效果并將制動(dòng) 量傳輸?shù)礁髦苿?dòng)器,按 操縱 不同 分為 液壓制動(dòng)操縱系統(tǒng) 液壓伺服式制動(dòng)操縱系統(tǒng) 動(dòng) 式液壓制動(dòng)操縱系統(tǒng)。題首先 汽車的基本參 為依據(jù)對(duì)制動(dòng)系統(tǒng)進(jìn) 設(shè)計(jì);計(jì)算制動(dòng)主缸并對(duì)其 設(shè)計(jì);為 使汽車的制動(dòng)管路的布局更理,對(duì)其布進(jìn)設(shè)計(jì);制動(dòng) 的分配計(jì)算;利用應(yīng)用軟件進(jìn)建模分 。本設(shè)計(jì) 用基 理論 專業(yè) ,對(duì)汽車制動(dòng)系統(tǒng) 進(jìn)分 , 制動(dòng)器 制動(dòng)操縱系統(tǒng)的 式,并進(jìn)理的設(shè)計(jì)計(jì)算 設(shè)計(jì),而設(shè)計(jì)具 足 制動(dòng)效 保證汽車的 。而 通 設(shè)計(jì), 理論 實(shí)際 ,更要的 的 與學(xué) , 的軟件 。 畢 業(yè) 設(shè) 計(jì)(論文) 開 題 報(bào) 告 2本題要 或解決的問題 采用的 ( ): 一 內(nèi)容(1)制動(dòng)系統(tǒng)參 計(jì)算及制動(dòng)器 設(shè)計(jì);(2)制動(dòng)主缸計(jì)算與 設(shè)計(jì);(3)制動(dòng)管路布設(shè)計(jì);(4)制動(dòng) 分配計(jì)算;(5)利用應(yīng)用軟件進(jìn)建模分 。 (1) 好理論基 的準(zhǔn) ,汽車?yán)碚?汽車設(shè)計(jì) 。(2)查汽車專業(yè) 書 資料, 解汽車制動(dòng)系統(tǒng) 。(3)查 量 論文,學(xué) 題 的 。(4)制 目, 目標(biāo), 并實(shí)。(5)計(jì)算機(jī) 軟件并 用。(6)與指導(dǎo)currency1師 專業(yè)的交,“解決到的問題。 畢 業(yè) 設(shè) 計(jì)(論文) 開 題 報(bào) 告 指導(dǎo)教師意見:1對(duì)“文獻(xiàn)綜述”的fi: fl對(duì)題所 及的問題文獻(xiàn),并 對(duì)題 的 狀 動(dòng) 前 進(jìn)綜分 述,文獻(xiàn)綜述要求。 2對(duì)本題的 及工作量的意見 對(duì)設(shè)計(jì)(論文) 果的 :對(duì)題所 及的 內(nèi)容在 專業(yè) 的基 上,進(jìn)一學(xué) 軟件,應(yīng) 期完成本 畢業(yè)設(shè)計(jì)。 3. ”同意開題: 同意 不同意 指導(dǎo)教師: 2016 03 09 所在專業(yè)審查意見:同意 : 2016 04 07 畢 業(yè) 設(shè) 計(jì)(論 文)外 文 參 考 資 料 及 譯 文譯文題目: VEHICLE DETECTION AND TRACKING 車輛檢測(cè)與跟蹤 學(xué)生姓名:專業(yè):所在學(xué)院:指導(dǎo)教師:職稱: Chapter 5Vehicle Detection and Tracking5.1 IntroductionStatistics shows that about 60% of the rear-end crash accidents can be avoided if the driver has additional warning time. According to the Ministry of Public Safety of P.R. China, there were 567,753 reported road traffic accidents in 2004, among those about 80% of the severe police-reported traffic accidents were vehiclevehicle crashes. Almost two-fifths of these crashes resulted in an injury, with over 2% of the total crashes resulting in a death. Clearly, vehicle detection is an important research area of intelligent transportation systems 2, 11, 20. It is being used in, among others, adaptive cruise control (ACC), driver assistance systems, automated visual traffic surveillance (AVTS), and self-guided vehicles. However, robust vehicle detection in real world traffic scenes is challenging.Currently, IDASW systems based on radars have a higher cost than those based on machine vision, while having narrow field of view and bad lateral resolution. In Adaptive Cruise Control (ACC) systems, a camera can detect the cut-in and over- taking vehicle from the adjacent lane earlier than a radar. Due to these reasons, it is more difficult to apply such radar-based systems into practical IDASW systems. Consequently, robust and real time vehicle detection in video attracts more attention of scholars all over the world 2, 4, 14.To detect on-road vehicle in time, this chapter introduces a multi-resolution hypothesis-validation structure. Inspired by A. Broggio2, we extract three ROIs: a near one, one in the middle, and a far one, from a 640 480 image. His approach uses fixed regions at the cost of flexibility, we remove this limitation and build a simple and efficient hypothesis-validation structure which consists of the three steps described below:1.ROI determination: We generate ROI candidates using a vanishing point of the road in the original image.2.Vehicle hypothesis generation for each ROI using horizontal and vertical edge detection: We create a multi-resolution vehicle hypothesis based on the preceding candidate regions. From the analysis of edge histograms, we generate hypotheses for each ROI and combine them into a single list.3.Hypothesis validation using Gabor features and SVM classifiers: We conduct vehicle validation using the boosted Gabor features of 9 sub-windows and the SVM classifiers. According to the judging of the classifiers, we determine whether hypotheses represent a vehicle or a non-vehicle.5.2 Related WorkHypotheses are generated using some simple features, such as color, horizontal and/or vertical edges, symmetry 2, 5, motion, and stereo visual cue. Zehang Sun proposed a multi-scale hypothesis method in which the original image was down- sampled to 320 240, 160 120, and 80 60. His vehicle hypotheses were generated by combining the horizontal and vertical edges of these three levels, and this multi-scale method greatly reduced random noise. This approach can generate multiple-hypothesis objects, but a near vehicle may prevent a far vehicle from being detected. As a result, the method fails to generate the corresponding hypothesis of the far vehicle, reducing the vehicle detection rate.B. Leibe et al. seated a video-based 3D dynamic scene analysis system from a moving vehicle 9 which integrated scene geometry estimation, 2D vehicle and pedestrian detection, 3D localization and trajectory estimation. Impressively, this paper presented a multi-view/multi-category object detection approach in a real world traffic scene. Furthermore, 2D vehicle pedestrians detection is converted into 3D observation.Vehicle symmetry is an important cue in vehicle detection and tracking. Inspired by the voting of Hough Transform, Yue Du et al. proposed a vehicle following approach by finding the symmetry axis of a vehicle 5; however, their approach has several limitations, such as large computing burden, and it only generates one object hypothesis using the best symmetry. Alberto Broggi introduced a multi-resolution vehicle detection approach, and proposed dividing the image into three fixed ROIs: one near the host car, one far from the host car, and one in the middle 2. This approach overcomes the limit of only being able to detect a single vehicle in the predefined region of the image, but it needs to compute the symmetry axis, making it not real-time.D. Gabor first proposed the 1D Gabor function in 1946 and J.G. Daugman ex- tended it to 2D later. In fact, a Gabor filter is a local bandpass filter that can reach the theoretical limit for the spatial domain and the frequency domain simultaneously. Consequently, Gabor filters have been successfully applied for object representation in various computer vision applications, such as texture segmentation and recognition 18, face recognition 19, scene recognition, and vehicle detection 14.The basic issue of a Gabor filter is how to select the parameters of a filter that responds mainly to an interesting object, such as a vehicle or a pedestrian. Accurate detection only occurs if the parameters defining Gabor filters are well selected. Three main approaches have been proposed in the literature for selecting Gabor filters for object representation: manual selection, Gabor filter bank design (including filter design) 18, and a learning approach 13, 14, 16, 19. In 1, Ilkka Autioproposed an approach for manual selection: An initial set of Gabor filters were experimentally selected from a larger set and then manually tuned. In general, a Gabor filter bank design defines a small filter pool, and determines the parameters of its filters independent of the application domain; moreover, the bandwidth of those Gabor filter design approaches cannot be determined autonomously. In image browsing and retrieval, a strategy is used to ensure that the half-peak magnitude support of the filter responses in the frequency domain touch each other by using a filter bank with 6 directions and 4 scales to compute the features of a texture 12. Due to independence of the filter bank and the application domain, such an approach can be used for object classification, detection and tracking. The main problems of this filter design approach are small filter pool sizes, no prior knowledge, and poor performance. Learning-based Gabor filter design approaches select the Gabor filters according to its application domain. Du-Ming Tsai proposed an optimization algorithm for Gabor filters using a simulated annealing approach to obtain the best Gabor filter in texture segmentation 16. A face recognition application using a strong classifier cascaded by weak classifiers was proposed by S.Z. Li; in his approach, weak classifiers were constructed based on both the magnitude and phase features from Gabor filters 19. In terms of vehicle detection, Alberto Broggi introduced a multi-resolution vehicle detection approach, and proposed dividing the image into three fixed ROIs 2. His approach allows detecting multiple vehicles in a predefined region. How- ever, it uses a symmetry axis for detecting vehicles that is not only time-consuming to compute but symmetry features are somewhat problematic. In 14, Zehang Sun proposed an Evolutionary Gabor Filter Optimization (EGFO) approach for vehicle detection, and used the statistical features of the response of selected Gabor filters to classify the test image using a trained SVM classifier. Although good performance has been reported, EGFO has large computational cost for the selection of a Gabor filter. Moreover, each Gabor filter is optimized for a complete image, but it is applied to each sub-window of a test image, which reduces the quality of the representation. The requirements of Vehicle Active Safety Systems (VASS) are strict with respect to the time performance for pedestrian detection and vehicle detection. Accordingly, in our approach we detect vehicles only in ROIs, allowing us to make a real-time implementation. The ROI approach largely prevents a near car from hiding a far car. All the hypotheses are generated in these regions. The positions of vehicles are validated by SVM classifiers and Gabor features.5.3 Generating Candidate ROIsInspired by A. Broggio 2, we extract three ROIs: a near one, one in the middle, and a far one from a 640 480 image. But his approach uses fixed regions at the cost of flexibility. In our approach, ROIs are extracted using lane markings. In a structured lane, we detect the vanishing point using the lane edges. For the consideration of real-time processing, we use a simple vanishing point detector rather than a complex one. Discontinuity and noise related problems can be solved by combining, for instance, 10 subsequent images (see Fig. 5.1(a). Edge detection is done on combined imagesconsisting of 10 overlapping subsequent images, and the equations of two lanes are deduced from a voting procedures like HT by analyzing horizontal and vertical edges. Four random points Pdi, d = l or r ; i= 0,., 3, are selected on each lane line, and each tangent direction of two points (shown in (5.1) between the closest 3 points; or ; ; (5.1)is obtained byThe tangent directions of two lane lines are calculated using the average value of the above tangent angles and are described by , or l. (5.2)Combining the average coordinates of 4 interesting points with the average tangent angles d , we can get the equations of two lane lines. The intersection point of the two lines is an approximation of the vanishing point; see Fig. 5.2. Next we consider how to extract ROIs from the original image. For the consideration of vehicle height and the camera parameters, the top boundaries of all the ROIs are 10 pixels higher than the vertical coordinates of the vanishing point. From the analysis of the camera parameters and image resolution, the heights of the near, middle, and far ROIs are 160, 60, and 30 pixels, respectively. The left and right boundaries of the near ROI are those of the image. The distance between the left boundary of the middle ROI and that of the image is just one-third of the distance between the vanishing point and the left boundary of the image, and the right one of middle ROI is determined similarly. The distance between the left boundary of the far ROI and that of image is two-thirds of the distance between the vanishing point and the left boundary of image, as well as the distance between the right boundary of the far ROI and that of the image. Figure 5.2(b) shows the results of each ROI.Fig. 5.2 Vanishing point and ROI generation第5章車輛檢測(cè)和跟蹤5.1簡(jiǎn)介據(jù)統(tǒng)計(jì)表明,如果駕駛員有預(yù)防危險(xiǎn)發(fā)生的能力,就能夠避免約60的追尾事故。根據(jù)中國(guó)大陸的交通部門報(bào)道,2004年共有567753道路交通事故,在那些報(bào)告中大約80的交通事故是汽車引起的。這些追尾事故所造成傷害的數(shù)據(jù)是所有事故數(shù)據(jù)的五分之二,導(dǎo)致死亡的超過2。顯然,車輛檢測(cè)是智能交通系統(tǒng)的一個(gè)重要的研究領(lǐng)域2,11,20。除此之外,車輛檢測(cè)被用于自適應(yīng)巡航控制系統(tǒng)(ACC),駕駛輔助系統(tǒng),自動(dòng)可視化交通監(jiān)控(AVTS),以及自導(dǎo)系統(tǒng)的車輛。然而,在檢測(cè)現(xiàn)實(shí)世界中的交通場(chǎng)景的車輛是具有挑戰(zhàn)性的。目前,雷達(dá)(IDASW)系統(tǒng)的機(jī)器成本較高,而且有橫向分辨的能力。在巡航控制(ACC)系統(tǒng)中,攝像機(jī)可以從相鄰車道的車輛檢測(cè)更早切入并使用畫面來檢測(cè)。由于這些原因,此系統(tǒng)就更加難以應(yīng)用。因此,現(xiàn)實(shí)中的車輛檢測(cè)視頻受到了世界各地的研究者們的更多關(guān)注 2,4,14。為了能夠及時(shí)檢測(cè)到道路上的車輛,本章引入了一種多分辨率結(jié)構(gòu)。由布羅基2的啟發(fā),我們提出了3個(gè)樣區(qū)點(diǎn):從一個(gè)640480的圖像中在近處的一個(gè)地方選取一個(gè)點(diǎn),中間的一個(gè)點(diǎn),和遠(yuǎn)處的一個(gè)點(diǎn)。他的方法是使用固定的地區(qū)為代價(jià)的靈活性,我們?nèi)サ暨@個(gè)限制,并建立一個(gè)簡(jiǎn)單而高效的結(jié)構(gòu)方案,其中包括下面描述的三個(gè)步驟:1. 投資回報(bào)率判定:我們生成的投資回報(bào)率所使用圖像是樣區(qū)點(diǎn)的原始圖像。2.車輛因?yàn)槊總€(gè)樣區(qū)點(diǎn)所使用水平和垂直方向的邊緣檢測(cè)情況不同而產(chǎn)生錯(cuò)誤觀點(diǎn):我們基于前述車輛候選區(qū)域分辨率創(chuàng)建假說,從邊緣直方圖的分析,我們將所生成的投資回報(bào)率合并之后得到一個(gè)單獨(dú)的列表。3.使用伽柏特性假設(shè)驗(yàn)證功能并用支持向量機(jī)分類:我們用9分窗口和支持向量機(jī)對(duì)車輛進(jìn)行驗(yàn)證。根據(jù)該向量機(jī)的判定,我們假設(shè)被測(cè)物體是表示車輛或非車輛。5.2相關(guān)工作使用一些簡(jiǎn)單的特性,如顏色,水平或垂直邊緣,對(duì)稱性2,5,運(yùn)動(dòng)和立體聲視覺線索生成假設(shè)。孫澤行提出了一種多重假說的方法,其中原始圖像以320240,160120和8060的格式取樣。假設(shè)他的車輛通過這三個(gè)層次的水平和垂直邊緣的組合時(shí),產(chǎn)生的這種多級(jí)方法分別使汽車大大降低了噪聲。那這種方法可以產(chǎn)生多個(gè)假設(shè)對(duì)象,而且是一個(gè)鄰近車輛可以防止被一個(gè)遠(yuǎn)車輛檢測(cè)。但其結(jié)果還是失敗,原因是遠(yuǎn)處的車輛檢出率特別低。雷本等人在坐一個(gè)具有3D動(dòng)態(tài)場(chǎng)景視頻系統(tǒng)9 的移動(dòng)車輛上觀察到,該車具有集成場(chǎng)景幾何估計(jì)的功能,它不僅可以檢測(cè)到車輛和行人,而且可以進(jìn)行3D定位和軌跡的估算。令人印象深刻的是,他的論文提出了一種用多視角/多類目標(biāo)檢測(cè)的方法檢測(cè)交通場(chǎng)景。此外,2D車輛行人畫面都可以轉(zhuǎn)換成三維觀察的模式來被檢測(cè)。 車輛的對(duì)稱性是在車輛檢測(cè)與跟蹤的一個(gè)重要線索。通過霍夫、杜悅等人提出,通過找到一個(gè)車輛5的對(duì)稱軸線提出以下很多問題;然而,他們的方法有一定限制,比如計(jì)算負(fù)擔(dān)重大,而且只生成一個(gè)對(duì)象來假設(shè)。阿爾貝托布羅吉特引入了多分辨率的車輛檢測(cè)方法,并提出了將圖像中取三個(gè)參考點(diǎn):一個(gè)是靠近主機(jī)的位置點(diǎn),一是遠(yuǎn)離主機(jī)的位置點(diǎn)和在中間的一個(gè)點(diǎn)2。此方法克服了僅能夠檢測(cè)單一區(qū)域車輛的局限性,但它還是需要計(jì)算對(duì)稱軸,這使得它不是實(shí)用的。 D.伽柏在1946年與道格交換版本權(quán)之后首次提出了一維伽柏函數(shù)。事實(shí)上,伽柏濾波器是一個(gè)當(dāng)?shù)氐臑V波器,而且它的頻率即將達(dá)到理論極限。因此,伽柏濾波器已成功地應(yīng)用于各種計(jì)算機(jī)系統(tǒng)當(dāng)中,如紋理分割和識(shí)別方面18,人臉識(shí)別19,場(chǎng)景識(shí)別和車輛檢測(cè)14中都能夠用到。伽柏濾波器的基本問題是如何將車輛或行人經(jīng)濾波器響應(yīng)來產(chǎn)生主要的。精確的檢測(cè)性能只有在確定伽柏濾波器的參數(shù)是否符合實(shí)際參數(shù)才能體現(xiàn)。伽柏已經(jīng)在文獻(xiàn)中提出了三種主要的方法來表示:手動(dòng)選擇,伽柏濾波器組設(shè)計(jì)(包括荷蘭國(guó)際集團(tuán)濾波器設(shè)計(jì))18,以及學(xué)習(xí)方法13,14,16,19。在1中,伊爾卡里斯蒂建議使用手動(dòng)選擇的方法:首先使用最初的一個(gè)伽柏濾波器進(jìn)行實(shí)驗(yàn)選定,然后手動(dòng)調(diào)整。在一般情況下,每一個(gè)伽柏濾波器組設(shè)計(jì)了一個(gè)小過濾池,并確定它的過濾器獨(dú)立應(yīng)用程序的參數(shù);此外,這些伽柏濾波器設(shè)計(jì)方法不能自主決定。在圖像瀏覽和檢索中,圖像是用來確保半峰級(jí)濾波器在通過使用一個(gè)具有6個(gè)向量的濾波器組來體現(xiàn)紋理的特征,使得濾波相互接觸12 。由于濾波器組和應(yīng)用領(lǐng)域的獨(dú)立性可用于濾波的檢測(cè)和跟蹤方法。該過濾器設(shè)計(jì)方法的主要問題在于小型過濾池的大小的設(shè)定,事先沒有了解,所以使用情況不好。根據(jù)伽柏濾波器的應(yīng)用領(lǐng)域來學(xué)習(xí)伽柏濾波器設(shè)計(jì)的方法。杜銘仔提出的優(yōu)化算法的伽柏濾波器是采用模擬退火方法來獲得紋理分割16的最佳伽柏濾波器。在他的方法中弱分類器是基于兩個(gè)從伽柏濾波器的幅度和相位19構(gòu)成的特點(diǎn)來使用的。在車輛檢測(cè)方面,阿爾貝托布羅吉特引入多分辨率的車輛檢測(cè)方法,并提出了將圖像劃分為三個(gè)固定的樣區(qū)點(diǎn)2。他的方法允許在預(yù)定區(qū)域檢測(cè)多輛車。然而,它使用了用于檢測(cè)一個(gè)車輛對(duì)稱軸的方法,這樣不僅計(jì)算費(fèi)時(shí),而且對(duì)稱方面也有些問題。在孫澤行提出在車輛檢測(cè)中用一個(gè)進(jìn)化伽柏濾波器進(jìn)行優(yōu)化(EGFO)的方法,和使用所選的伽柏濾波器響應(yīng)的統(tǒng)計(jì)特性用以系統(tǒng)虛擬機(jī)測(cè)試圖像來進(jìn)行分類。最后良好的性能得以體現(xiàn),EGFO有大量的計(jì)算成本的伽柏濾波器來選擇。此外,每個(gè)伽柏濾波器為每一個(gè)完整的圖像進(jìn)行了優(yōu)化,并將它應(yīng)用到測(cè)試圖像當(dāng)中,從而減少了原有的缺點(diǎn)。汽車主動(dòng)安全系統(tǒng)(VASS)的性能要求是保證行人檢測(cè)和車輛檢測(cè)的時(shí)間充足。 AC-科丁在我們的方法中發(fā)現(xiàn),我們只在固定的區(qū)域中檢測(cè)車輛,并做一個(gè)實(shí)驗(yàn)來確定。他的方法很大程度阻止自己的車與附近或者遠(yuǎn)處隱藏的車相撞。都是在這些區(qū)域中生成所有的假設(shè)。車輛的位置驗(yàn)證了支持向量機(jī)分類器的功用和特性。5.3生成候選的投資回報(bào)率由A.的布羅根2的啟發(fā),我們提取出3個(gè)樣區(qū)點(diǎn):在640480的圖像中一個(gè)鄰近的位置點(diǎn),一個(gè)在中間,和一個(gè)來自遠(yuǎn)處的位置點(diǎn)。但是,他的方法是使用固定區(qū)域的靈活性為代價(jià)的。在我們的方法中,投資回報(bào)率是使用車道標(biāo)記位置點(diǎn)提取的。在結(jié)構(gòu)化車道中,我們使用的是車道邊緣檢測(cè)的消失點(diǎn)。對(duì)于實(shí)時(shí)處理的考慮,我們用一個(gè)簡(jiǎn)單的消失點(diǎn)檢測(cè),而不是復(fù)雜的一個(gè)點(diǎn)??梢酝ㄟ^各方法來解決不連續(xù)性和噪聲有關(guān)的問題,(a)使單幀車道邊緣 (b)重疊的車道例如,在10個(gè)后續(xù)圖像中(見圖5.1(a)。結(jié)合邊緣檢測(cè)圖上的10個(gè)消失點(diǎn),以及從一個(gè)表決程序圖,通過分析水平和垂直邊緣推導(dǎo)出的兩種車道方程。提供出四個(gè)隨機(jī)點(diǎn),D = L或R; I = 0,.,3,是在每個(gè)車道線上任意選擇的,并且在每個(gè)切線方向上兩個(gè)點(diǎn)之間接近3點(diǎn)(在(5.1示出); or l; ; (5.1)是此獲得平均值計(jì)算雙車道線的切線方向所使用的上述切線的角度,得 , or l. (5.2)我們可以結(jié)合平均值坐標(biāo)得出兩個(gè)車道線的方程為。在十字路口的這兩條線的交點(diǎn)是消失點(diǎn);參照?qǐng)D5.2。下一步,我們所要考慮的是如何從原始圖像中提取出投資回報(bào)率。用于車輛的高度和攝像機(jī)的參數(shù),所有的感應(yīng)區(qū)域的頂部邊界比消失點(diǎn)的垂直坐標(biāo)高10像素。從照相機(jī)參數(shù)和圖像分辨率分析,近、中和遠(yuǎn)的樣區(qū)160,60和30的像素的區(qū)別。得到左和右邊界附近的樣區(qū)圖像。中間樣區(qū)的左邊界和該圖像之間的距離僅是消失點(diǎn)和圖像的左側(cè)邊界之間的距離的三分之一,而中間樣區(qū)被類似地確定為標(biāo)準(zhǔn)點(diǎn)。遠(yuǎn)處樣區(qū)的左邊界和該圖像之間的距離是消失點(diǎn)和圖像左邊界之間的距離的三分之二,以及遠(yuǎn)處樣區(qū)的右邊界和左邊界之間的距離圖片也是這樣。圖5.2(b)表示每個(gè)投資回報(bào)率的結(jié)果。(a)兩條線之間的焦點(diǎn)(b)部門的投資回報(bào)率(c)車道一帶的投資回報(bào)率圖 5.2 消失點(diǎn)和投資回報(bào)率的產(chǎn)生
收藏