本文選題：BP神經(jīng)網(wǎng)絡(luò) + 區(qū)域分類器��； 參考：《大連海事大學(xué)》2017年碩士論文

【摘要】：二十世紀(jì)以來(lái),汽車的擁有量顯著增長(zhǎng),隨之而來(lái)的交通事故也發(fā)生的越來(lái)越頻繁,人們對(duì)交通安全問(wèn)題的關(guān)注與日俱增,世界各國(guó)的交通部正積極采取一些有效措施,減少交通事故的發(fā)生。因此,基于駕駛員的視線追蹤系統(tǒng)便應(yīng)運(yùn)而生。但現(xiàn)有的系統(tǒng)一般局限于簡(jiǎn)單場(chǎng)景,且必須在前期做好校準(zhǔn)工作的情況下進(jìn)行的,而對(duì)于不加約束的人、無(wú)校準(zhǔn)、光照變化等問(wèn)題仍存在很大的研究空間,其實(shí)時(shí)性、精確性和魯棒性與實(shí)際應(yīng)用之間還存在很大距離。針對(duì)這些問(wèn)題,本文提出基于BP神經(jīng)網(wǎng)絡(luò)的無(wú)校準(zhǔn)駕駛員注視區(qū)域估計(jì)方法,將重點(diǎn)從頭部姿態(tài)和瞳孔視線角度參數(shù)獲取、BP神經(jīng)網(wǎng)絡(luò)注視區(qū)域估計(jì)算法,及實(shí)驗(yàn)對(duì)比評(píng)估三個(gè)部分進(jìn)行研究,主要工作如下:首先,本文需要先獲取頭部姿態(tài)和視線方向的角度參數(shù),在此過(guò)程中,針對(duì)駕駛員身體出現(xiàn)左右晃動(dòng),或者不同駕駛員身高不同,而發(fā)生相對(duì)于攝像機(jī)的左右偏移和上下偏移問(wèn)題,提出一種基于幾何關(guān)系的頭部姿態(tài)校正算法。同時(shí),本文通過(guò)建立3D眼球模型進(jìn)行瞳孔視線方向估計(jì)。然后,構(gòu)建了一個(gè)基于BP神經(jīng)網(wǎng)絡(luò)的無(wú)校準(zhǔn)駕駛員注視區(qū)域估計(jì)系統(tǒng)。通過(guò)BP神經(jīng)網(wǎng)絡(luò)模型對(duì)駕駛員在駕駛過(guò)程中的頭部姿態(tài)及視線角度參數(shù)進(jìn)行訓(xùn)練并構(gòu)建區(qū)域分類器,并通過(guò)該網(wǎng)絡(luò)模型進(jìn)行駕駛員無(wú)校準(zhǔn)注視區(qū)域估計(jì)。最后,對(duì)本文方法進(jìn)行評(píng)估。通過(guò)對(duì)比實(shí)驗(yàn)表明,本文提出的方法不僅能滿足學(xué)術(shù)研究的要求,而且能實(shí)現(xiàn)在復(fù)雜環(huán)境下駕駛員的注視區(qū)域估計(jì),滿足了對(duì)實(shí)驗(yàn)的實(shí)時(shí)性,精確度和魯棒性的要求,并為安全駕駛的輔助系統(tǒng)打下良好的基礎(chǔ)。
[Abstract]:Since twentieth Century, the number of cars has increased significantly, and the traffic accidents are becoming more and more frequent. People pay more attention to traffic safety. The transportation department of the world is actively taking some effective measures to reduce the occurrence of traffic accidents. However, the existing systems are generally limited to simple scenes, and must be carried out in the early stage of calibration, but there is still a lot of research space for unconstrained people, no calibration, light change and other problems, and there is a great distance between the reality, the accuracy and the robustness. An uncalibrated area estimation method based on BP neural network, which focuses on the head attitude and the eye view angle parameters, the BP neural network fixation area estimation algorithm, and the experimental comparison and evaluation of three parts are studied. The main work is as follows: first, we need to obtain the angle reference of the head attitude and the direction of sight. In this process, a head attitude correction algorithm based on the geometric relationship is proposed for the driver body sloshing, or different driver's height, and the left and right offset and up and down migration of the camera. At the same time, the 3D eye model is established to estimate the eye direction of the pupil. Then, the construction of the eye direction is constructed. An uncalibrated area estimation system based on BP neural network is introduced. The BP neural network model is used to train the driver's head attitude and view angle parameters in the driving process and construct a regional classifier, and the driver's uncalibrated gaze area estimation is carried out through the network model. Finally, the method is carried out. A comparative experiment shows that the proposed method can not only meet the requirements of academic research, but also realize the estimation of the driver's gaze area under the complex environment, meet the requirements of the real-time, accuracy and robustness of the experiment, and lay a good foundation for the auxiliary system of safe driving.
【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
【分類號(hào)】：TP391.41;TP183

【參考文獻(xiàn)】

相關(guān)期刊論文 前1條

1 喬體洲;戴樹(shù)嶺;;基于回歸森林的面部姿態(tài)分析[J];計(jì)算機(jī)輔助設(shè)計(jì)與圖形學(xué)學(xué)報(bào);2014年07期

，

本文編號(hào)：2091573