本文選題：車(chē)輛檢測(cè) + Faster�。� 參考：《大連海事大學(xué)》2017年碩士論文

【摘要】：近些年來(lái),在視頻監(jiān)控聯(lián)網(wǎng)、高清化技術(shù)的推動(dòng)下,交通行業(yè)視頻監(jiān)控業(yè)務(wù)的數(shù)據(jù)量快速增長(zhǎng)。利用計(jì)算機(jī)視覺(jué)技術(shù)處理交通視頻得到有效的信息已經(jīng)逐漸被重視,并且根據(jù)監(jiān)控視頻內(nèi)容自動(dòng)計(jì)算車(chē)輛運(yùn)行軌跡已經(jīng)可行。本文針對(duì)攝像頭網(wǎng)絡(luò)中車(chē)輛檢測(cè)和識(shí)別問(wèn)題進(jìn)行了研究。對(duì)于攝像頭網(wǎng)絡(luò),既需要對(duì)單攝像頭下視頻處理,又需要建立多個(gè)攝像頭間的聯(lián)系,從而完成在攝像頭網(wǎng)絡(luò)中對(duì)車(chē)輛進(jìn)行連續(xù)的追蹤。本文主要完成了以下工作:使用 Faster Region with Convolutional Neural Network feature 網(wǎng)絡(luò),在原有檢測(cè)模型的基礎(chǔ)上,重新標(biāo)記數(shù)據(jù)集,對(duì)網(wǎng)絡(luò)進(jìn)行調(diào)優(yōu),重新訓(xùn)練車(chē)輛檢測(cè)模型。在單攝像頭下,提出了基于重疊面積率的車(chē)輛追蹤算法,該算法利用相鄰兩幀視頻幀中目標(biāo)車(chē)輛的重疊面積率判定是否屬于同一軌跡,對(duì)于車(chē)輛被遮擋以及Faster R-CNN檢測(cè)失敗或漏檢的情況,加入卡爾曼濾波算法。利用卡爾曼濾波的預(yù)測(cè)機(jī)制,在進(jìn)行重疊面積率計(jì)算后,若有未匹配成功的追蹤器或車(chē)輛,則使用預(yù)測(cè)值進(jìn)行匹配,這樣可以盡量避免因檢測(cè)失敗或車(chē)輛被遮擋而導(dǎo)致的追蹤車(chē)輛失敗的情況。除此之外,為更直觀(guān)的表達(dá)追蹤結(jié)果,使用單應(yīng)矩陣對(duì)追蹤結(jié)果進(jìn)行了可視化。在多攝像頭下,針對(duì)不同攝像頭下光照、拍攝角度等不同使得車(chē)輛再識(shí)別難度加大這一問(wèn)題,本文根據(jù)攝像頭間的時(shí)空關(guān)系、車(chē)型屬性以及車(chē)輛Convolutional Neural Network特征建立了車(chē)輛再識(shí)別模型。其中利用攝像頭的地理位置可以得到攝像頭的空間信息,對(duì)視頻統(tǒng)計(jì)可以得到車(chē)輛在攝像頭間的轉(zhuǎn)移時(shí)間概率密度分布,通過(guò)GoogLeNet網(wǎng)絡(luò)重新訓(xùn)練車(chē)型檢測(cè)模型,并結(jié)合VggNet網(wǎng)絡(luò)提取到的車(chē)輛CNN特征,從而在不同攝像頭下的監(jiān)控視頻中對(duì)目標(biāo)車(chē)輛進(jìn)行了再識(shí)別。
[Abstract]:In recent years, with the promotion of video surveillance network and high-definition technology, the traffic industry video surveillance business data volume is growing rapidly. Using computer vision technology to process traffic video to get effective information has been paid more and more attention, and it is feasible to automatically calculate the vehicle track according to the content of surveillance video. In this paper, the problem of vehicle detection and recognition in camera network is studied. For the webcam network, it is necessary to process the video under the single camera and to establish the connection between several cameras, so that the vehicle can be tracked continuously in the webcam network. The main work of this paper is as follows: using the Faster region with Convolutional Neural Network feature network, on the basis of the original detection model, the data set is re-marked, the network is optimized, and the vehicle detection model is retrained. A vehicle tracking algorithm based on overlapped area rate is proposed under a single camera. The algorithm uses the overlap area rate of the target vehicle in two adjacent frames to determine whether it belongs to the same track or not. In the case of vehicle occlusion and Faster R-CNN detection failure or miss detection, Kalman filter algorithm is added. The prediction mechanism of Kalman filter is used to calculate the overlap area rate. If there is a tracker or vehicle that has not been matched successfully, the prediction value is used to match. This can avoid tracking failure due to detection failure or vehicle occlusion. In addition, in order to express the tracking results more intuitively, the monoclinic matrix is used to visualize the tracking results. Under the multi-camera, aiming at the problem that different illumination and shooting angle under different cameras make it more difficult to recognize the vehicle again, according to the space-time relationship between the cameras, Vehicle rerecognition model is established by vehicle attributes and vehicle volume neural network features. The spatial information of the camera can be obtained by using the location of the camera, the probability density distribution of the transfer time between the cameras can be obtained by the video statistics, and the model of vehicle detection can be retrained through Google LeNet network. Combined with the features of vehicle CNN extracted from VggNet network, the target vehicles are rerecognized in the surveillance video under different cameras.
【學(xué)位授予單位】：大連海事大學(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)博士學(xué)位論文 前1條

1 董文會(huì);多攝像機(jī)監(jiān)控網(wǎng)絡(luò)中的目標(biāo)連續(xù)跟蹤方法研究[D];山東大學(xué);2015年

相關(guān)碩士學(xué)位論文 前1條

1 肖暢;非重疊域多攝像機(jī)網(wǎng)絡(luò)車(chē)輛跟蹤研究[D];華中科技大學(xué);2013年

，

本文編號(hào)：2032753