發(fā)布時(shí)間：2018-03-13 21:20

  本文選題：弱對比度車輛識別　切入點(diǎn)：選擇注意　出處：《北京交通大學(xué)》2014年碩士論文　論文類型：學(xué)位論文

【摘要】：由于造價(jià)低廉、易于應(yīng)用,基于圖像的車輛識別技術(shù)而成為近年來國內(nèi)外研究的熱點(diǎn)。國內(nèi)外相關(guān)學(xué)者提出了很多富有建設(shè)性的方法并取得了一定成功,但是目前仍然存在環(huán)境適應(yīng)性和魯棒性差的缺點(diǎn),且對于復(fù)雜交通場景和惡劣天氣(如霧霾、雨雪等)下的弱對比度車輛目標(biāo)難以獲得令人滿意的識別率,這已經(jīng)嚴(yán)重制約了基于圖像的車輛識別技術(shù)的發(fā)展。因此車輛識別,尤其是弱對比度車輛目標(biāo)識別已經(jīng)成為一項(xiàng)具有挑戰(zhàn)意義和重要研究價(jià)值的工作。 針對以上問題,本文借鑒人類的視覺感知原理建立了適應(yīng)性強(qiáng)、魯棒性好的車輛識別模型,并探討建立聯(lián)想機(jī)制模型用于弱對比度車輛目標(biāo)的準(zhǔn)確識別。論文的具體工作如下： 1、基于人類的視覺選擇注意機(jī)制建立了雙向驅(qū)動(dòng)融合的注意模型用于車輛識別。該模型在基于bottom-up數(shù)據(jù)驅(qū)動(dòng)的Saliency模型基礎(chǔ)上,選擇車輛目標(biāo)最魯棒的結(jié)構(gòu)和形狀特征建立兩級知識庫實(shí)現(xiàn)了top-down的任務(wù)驅(qū)動(dòng),在高層指導(dǎo)Saliency模型中的視覺選擇注意過程,實(shí)現(xiàn)了任務(wù)驅(qū)動(dòng)與數(shù)據(jù)驅(qū)動(dòng)的融合。其中,在數(shù)據(jù)驅(qū)動(dòng)過程中,利用譜分析方法和顯著度函數(shù)代替了基于高斯金字塔的多尺度顯著特征融合算法,提高了模型的實(shí)時(shí)性；在構(gòu)建形狀知識庫時(shí),利用格式塔知覺理論的相關(guān)原理建立了用于提取車輛目標(biāo)閉合邊界集的多目標(biāo)分割模型。 2、探討建立了用于弱對比度目標(biāo)識別的聯(lián)想機(jī)制模型。提出了具有優(yōu)秀聯(lián)想能力的綠色神經(jīng)元交互聯(lián)想網(wǎng)絡(luò),利用高維聯(lián)想空間映射網(wǎng)完成了對于弱對比度目標(biāo)不完整特征的模式異聯(lián)想,并通過解聯(lián)想映射網(wǎng)實(shí)現(xiàn)了目標(biāo)的自聯(lián)想功能；同時(shí)建立神經(jīng)調(diào)節(jié)函數(shù)和神經(jīng)交互函數(shù)模擬了生物神經(jīng)信號傳導(dǎo)過程中神經(jīng)元的交互作用,使聯(lián)想網(wǎng)絡(luò)具有更快的收斂速度。在此基礎(chǔ)上,本文借鑒人類視皮層中的WHAT通路將大腦皮層的聯(lián)想功能合理抽象為聯(lián)想產(chǎn)生、聯(lián)想匹配和綜合分析的層次化模型,從而構(gòu)建了能有效識別弱對比度目標(biāo)的聯(lián)想機(jī)制模型。 通過仿真實(shí)驗(yàn)得出,雙向驅(qū)動(dòng)融合的注意模型對于只有清晰車輛目標(biāo)的樣本集的識別率為90.4%,誤識別率為4.9%；對于包含弱對比度車輛目標(biāo)的測試樣本集的綜合識別率為76.4%,綜合誤識別率為4.8%；引入聯(lián)想機(jī)制模型后系統(tǒng)的綜合識別率為88.5%,綜合誤識別率為4.9%。實(shí)驗(yàn)結(jié)果表明,對于清晰車輛目標(biāo),雙向驅(qū)動(dòng)融合的注意模型具有很高的識別率,且魯棒性強(qiáng)；引入聯(lián)想機(jī)制模型能在保證系統(tǒng)魯棒性的基礎(chǔ)上顯著提高系統(tǒng)對于弱對比度目標(biāo)的識別能力。
[Abstract]:Because of its low cost and easy application, the image-based vehicle recognition technology has become a hot topic in recent years. Many constructive methods have been put forward by domestic and foreign scholars and some success has been achieved. However, there are still shortcomings of poor environmental adaptability and robustness, and it is difficult to obtain satisfactory recognition rate for vehicle targets with weak contrast in complex traffic scenarios and severe weather (such as haze, rain and snow). This has seriously restricted the development of image-based vehicle recognition technology, so vehicle recognition, especially the weak contrast vehicle target recognition, has become a challenge and important research work. In view of the above problems, this paper builds a vehicle recognition model with strong adaptability and good robustness based on the principle of human visual perception. This paper also discusses the establishment of associative mechanism model for the accurate identification of vehicle targets with weak contrast. The specific work of this paper is as follows:. 1. Based on the human visual selective attention mechanism, a bidirectional driving fusion attention model is established for vehicle recognition. The model is based on the bottom-up data-driven Saliency model. Selecting the most robust structure and shape features of vehicle targets, a two-level knowledge base is established to realize the task driven of top-down, and the visual selection attention process in the high-level Saliency model is guided by the fusion of task driving and data driving. In the data-driven process, spectral analysis method and saliency function are used to replace the multi-scale salient feature fusion algorithm based on Gao Si pyramid, which improves the real-time performance of the model. Based on the related principle of Gestalt perception theory, a multi-objective segmentation model for extracting closed boundary sets of vehicle targets is established. 2. The association mechanism model for weak contrast target recognition is established, and a green neural interactive association network with excellent association ability is proposed. Using the high-dimensional associative space mapping net, the pattern heterodyne association for the incomplete feature of the weak contrast target is completed, and the self-associative function of the target is realized through the de-associative mapping net. At the same time, the neural regulation function and the neural interaction function are established to simulate the interaction of neurons in the process of biological nerve signal transduction, which makes the associative network converge faster. Based on the WHAT pathway in human visual cortex, the association function of cerebral cortex is reasonably abstracted into a hierarchical model of association generation, association matching and comprehensive analysis, and a model of association mechanism which can effectively identify the target of weak contrast is constructed in this paper. The simulation results show that, The attention model of bidirectional driving fusion has a recognition rate of 90.4 for a sample set with only clear vehicle targets and a false recognition rate of 4.9.The synthetic recognition rate for a test sample set containing a weak contrast vehicle target is 76.4, and the comprehensive error recognition rate is 76.4. After introducing the associative mechanism model, the comprehensive recognition rate of the system is 88. 5 and the comprehensive error recognition rate is 4. 9. The experimental results show that, For clear vehicle targets, the attention model of bidirectional driving fusion has a high recognition rate and strong robustness, and the associative mechanism model can significantly improve the recognition ability of weak contrast targets on the basis of ensuring the robustness of the system.
【學(xué)位授予單位】：北京交通大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2014
【分類號】：TP391.41;U495

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 鄒琪;羅四維;鄭宇;;利用多尺度分析和編組的基于目標(biāo)的注意計(jì)算模型[J];電子學(xué)報(bào);2006年03期

2 文學(xué)志;方巍;鄭鈺輝;;一種基于類Haar特征和改進(jìn)AdaBoost分類器的車輛識別算法[J];電子學(xué)報(bào);2011年05期

3 沈\，

本文編號：1608170