發(fā)布時(shí)間：2018-06-23 01:39

  本文選題：圖像拼接 + 統(tǒng)一計(jì)算設(shè)備架構(gòu)��； 參考：《長(zhǎng)春理工大學(xué)》2016年博士論文

【摘要】：無(wú)論是民用領(lǐng)域的礦產(chǎn)資源勘查、土地規(guī)劃利用、環(huán)境監(jiān)測(cè)、海洋開發(fā)、氣象預(yù)報(bào)及地理信息服務(wù)還是軍事領(lǐng)域偵察監(jiān)視、精確制導(dǎo)、超視距攻防對(duì)抗等都需要有足夠?qū)挼囊晥?chǎng)和足夠高的分辨率以完成對(duì)目標(biāo)的廣域范圍監(jiān)測(cè)、搜索和跟蹤。對(duì)基于寬視場(chǎng)高分辨成像系統(tǒng)海量圖像數(shù)據(jù)的處理、分析和利用是該類系統(tǒng)建構(gòu)的核心價(jià)值所在。其中,高分辨率傳感器拼接成像過(guò)程中將涉及到對(duì)海量數(shù)據(jù)的實(shí)時(shí)處理,計(jì)算過(guò)程有著極高的復(fù)雜度,因此,圖像拼接算法的準(zhǔn)確性和實(shí)時(shí)性將成為影響系統(tǒng)性能的主要因素之一。此外,對(duì)于寬視場(chǎng)高分辨場(chǎng)景下動(dòng)態(tài)目標(biāo)的跟蹤技術(shù)也成為后期圖像分析的研究熱點(diǎn)。同時(shí),由于這類系統(tǒng)應(yīng)用的環(huán)境自身的復(fù)雜性(背景變化、光照變化、陰影變化等)和動(dòng)態(tài)目標(biāo)固有的一些特性(非剛體、姿態(tài)多變等),使得可實(shí)用的目標(biāo)跟蹤技術(shù)仍然非常具有挑戰(zhàn)性。針對(duì)以上需求,本文圍繞寬視場(chǎng)成像系統(tǒng)的圖像拼接和目標(biāo)跟蹤問(wèn)題開展研究,主要研究工作如下:本文采用了一種將先驗(yàn)信息和統(tǒng)一計(jì)算設(shè)備架構(gòu)(CUDA)相結(jié)合的自適應(yīng)并行加速算法用于提高大視場(chǎng)全景拼接成像的實(shí)時(shí)性。在圖像拼接之前,先利用高精度標(biāo)定平臺(tái)對(duì)各成像單元的重疊區(qū)域進(jìn)行預(yù)標(biāo)定。標(biāo)定之后,利用基于CUDA的快速魯棒特征檢測(cè)方法提取參考圖像與待配準(zhǔn)圖像的候選特征點(diǎn)集,再利用基于隨機(jī)KD-Tree索引的近似最近鄰搜索算法選取參考圖像與待配準(zhǔn)圖像的初始匹配點(diǎn)對(duì),本文還采用了基本線性代數(shù)運(yùn)算子程序用于加速算法搜索速度。對(duì)于參考圖像與待配準(zhǔn)圖像誤匹配點(diǎn)對(duì)的刪除和空間變換矩陣的參數(shù)估計(jì),本文采用的是一種在傳統(tǒng)的漸近式抽樣一致性算法基礎(chǔ)上改進(jìn)的基于CUDA的并行算法。實(shí)驗(yàn)結(jié)果表明本文采用的算法極大地提高了圖像拼接速度,可以滿足圖像拼接實(shí)時(shí)性的工程應(yīng)用要求。為了對(duì)場(chǎng)景中動(dòng)態(tài)飛行目標(biāo)進(jìn)行識(shí)別,提出一種基于混沌雙種群進(jìn)化策略的圖像分割方法。利用進(jìn)化策略能從選定的初始解出發(fā),通過(guò)逐代迭代進(jìn)化逐步改進(jìn)當(dāng)前解,直至最后收斂于最優(yōu)解或滿意解的特點(diǎn)和優(yōu)勢(shì),將其用于圖像分割閾值最優(yōu)解的求解上。為了克服傳統(tǒng)基于閾值的圖像分割方法的缺點(diǎn),例如較高的復(fù)雜度和早熟問(wèn)題,本文提出了一個(gè)高效的基于進(jìn)化策略的圖像分割算法,它通過(guò)使用多種群進(jìn)化策略來(lái)計(jì)算閾值。在進(jìn)化過(guò)程中同時(shí)存在局部種群和全局種群兩個(gè)群體,進(jìn)而確保算法的全局和局部搜索能力。該算法的每一步迭代過(guò)程中,首先,基于混沌理論生成若干個(gè)初始個(gè)體,并將這些個(gè)體分別加入局部種群和全局種群,計(jì)算這些個(gè)體的適應(yīng)度函數(shù)值。然后,將選擇、重組、變異等進(jìn)化操作算子作用于局部種群和全局種群,進(jìn)行迭代進(jìn)化,進(jìn)化后的個(gè)體集合中選擇最好的若干個(gè)體放入局部種群,其余放入全局種群,直至收斂。最后,種群中的最優(yōu)個(gè)體即為所求的解。實(shí)驗(yàn)結(jié)果表明,本文提出的方法比傳統(tǒng)的遺傳算法有著更快的收斂速度。種群多樣性信息能有效指導(dǎo)進(jìn)化策略的進(jìn)化過(guò)程,因此本文又提出了改進(jìn)的混沌雙種群進(jìn)化策略算法,采用了多動(dòng)機(jī)強(qiáng)化學(xué)習(xí)算法設(shè)定初始種群和本地種群數(shù)值,動(dòng)態(tài)學(xué)習(xí)種群比例,以使進(jìn)化策略的局部搜索能力和全局搜索能力進(jìn)一步均衡化。動(dòng)機(jī)層的引入為先驗(yàn)知識(shí)和領(lǐng)域知識(shí)的引入提供了條件,由此可以加速?gòu)?qiáng)化學(xué)習(xí)的學(xué)習(xí)進(jìn)程。本文根據(jù)圖像分割問(wèn)題實(shí)際,定義了動(dòng)機(jī)集合,采用了MMQ投票(MMQ-voting)方法用于指導(dǎo)智能體動(dòng)作的選擇策略。經(jīng)過(guò)實(shí)驗(yàn)驗(yàn)證,本文采用的多動(dòng)機(jī)強(qiáng)化學(xué)習(xí)方法能使強(qiáng)化學(xué)習(xí)以較快的速度收斂于最優(yōu)動(dòng)作策略,從而使種群個(gè)體多樣性保持在一個(gè)合適的狀態(tài),有助于進(jìn)一步提高圖像最優(yōu)分割閾值的搜索效率。為了對(duì)場(chǎng)景中動(dòng)態(tài)飛行目標(biāo)進(jìn)行跟蹤,提出一種基于強(qiáng)化學(xué)習(xí)的動(dòng)態(tài)目標(biāo)跟蹤方法,將目標(biāo)跟蹤問(wèn)題建模成強(qiáng)化學(xué)習(xí)問(wèn)題,并提出了一個(gè)兩階段強(qiáng)化學(xué)習(xí)算法用于圖像中的目標(biāo)跟蹤。我們?cè)O(shè)置了多個(gè)追蹤智能體來(lái)跟蹤圖像中的目標(biāo),在算法的每一步中,首先對(duì)每個(gè)追蹤智能體進(jìn)行動(dòng)態(tài)子任務(wù)分配,即先是給每個(gè)追蹤智能體動(dòng)態(tài)分配一個(gè)子目標(biāo),之后每個(gè)追蹤智能體根據(jù)其當(dāng)前的子目標(biāo)選擇其行動(dòng)。學(xué)習(xí)算法將學(xué)習(xí)過(guò)程劃分為兩個(gè)部分,一個(gè)是學(xué)習(xí)任務(wù)分配的策略,另一個(gè)是學(xué)習(xí)動(dòng)作選擇的策略,每個(gè)追蹤智能體通過(guò)共享Q函數(shù)來(lái)共享所學(xué)知識(shí)、提高學(xué)習(xí)效率。實(shí)驗(yàn)結(jié)果驗(yàn)證了該方法的有效性。
[Abstract]:Whether it is the mineral exploration in the civil field, land planning and utilization, environmental monitoring, marine development, weather forecast and geographic information service or military field surveillance and surveillance, precision guidance, and over the horizon attack and defense confrontation need to have wide field of view and high enough resolution to complete the wide range monitoring, search and tracking of the target. The analysis and utilization of massive image data based on wide field of view high-resolution imaging system is the core value of this kind of system construction. Among them, high resolution sensor splicing will involve real-time processing of mass data and high complexity in the calculation process. Therefore, the accuracy and reality of the image mosaic algorithm will be true. Time character will be one of the main factors that affect the performance of the system. In addition, the tracking technology for dynamic targets in the wide field of view high resolution scene has also become a hot topic in the later image analysis. At the same time, due to the complexity of the environment itself (background change, illumination change, shadow change, etc.) and some inherent characteristics of the dynamic target, the application of this kind of system The practical target tracking technology is still very challenging. Aiming at the above requirements, this paper focuses on the problem of image mosaic and target tracking in the wide field of view imaging system. The main research work is as follows: a combination of prior information and unified computing device architecture (CUDA) is used in this paper. The adaptive parallel acceleration algorithm is used to improve the real-time performance of the panoramic mosaic imaging in large field. Before the image splicing, the high precision calibration platform is used to pre calibrate the overlapped regions of the imaging units. After calibration, the CUDA based fast robust feature detection method is used to extract the candidate feature points of the reference image and the image to be registered. Set, and then use the approximate nearest neighbor search algorithm based on random KD-Tree index to select the initial matching points of the reference image and the image to be registered. This paper also uses the basic linear algebra operation subroutine to speed up the search speed. It is estimated that this paper uses an improved CUDA based parallel algorithm based on the traditional asymptotic sampling consistency algorithm. The experimental results show that the algorithm used in this paper can greatly improve the speed of image stitching, and can meet the engineering application requirements of the real-time image splicing. An image segmentation method based on chaotic dual population evolution strategy is proposed. Using the evolutionary strategy, it can gradually improve the current solution by iterative evolution from the selected initial solution, and finally converge to the characteristics and advantages of the optimal solution or satisfactory solution, and apply it to the solution of the threshold optimal solution of the image segmentation. The shortcoming of threshold image segmentation methods, such as high complexity and precocious problem, presents an efficient image segmentation algorithm based on evolutionary strategy. It uses a variety of cluster evolution strategies to calculate threshold. In the process of evolution, there are two groups of local population and whole local population, which can ensure the overall situation of the algorithm. In each iteration of the algorithm, first of all, a number of initial individuals are generated based on the chaos theory, and these individuals are added to the local population and the global population to calculate the fitness function values of these individuals. Then, the evolutionary operators such as selection, reorganization and mutation are used to act on the local population and the global population. The optimal individual is placed in the local population and the rest is placed in the global population to converge. Finally, the optimal individual in the population is the solution. The experimental results show that the proposed method has a faster convergence rate than the traditional genetic algorithm. The population diversity information can be effective. In order to guide the evolutionary process of evolutionary strategy, this paper also proposes an improved chaotic dual population evolution strategy algorithm, which uses a multi motivation reinforcement learning algorithm to set the initial population and local population values and dynamically learn the population proportion, so that the local search ability and the overall search ability of the evolutionary strategy are further balanced. The introduction of motivation layer is introduced. It provides the conditions for the introduction of prior knowledge and domain knowledge, which can accelerate the learning process of intensive learning. Based on the reality of the image segmentation, this paper defines the set of motivation and uses the MMQ voting (MMQ-voting) method to guide the selection strategy of the action of the agent. The method can make the reinforcement learning converge to the optimal action strategy at a faster speed, so that the individual diversity of the population is kept in a suitable state and helps to further improve the search efficiency of the optimal segmentation threshold of the image. In order to track the dynamic flight targets in the scene, a dynamic target tracking method based on reinforcement learning is proposed. The target tracking problem is modeled as a reinforcement learning problem, and a two stage reinforcement learning algorithm is proposed for target tracking in the image. We set up multiple tracking agents to track the target in the image. In each step of the algorithm, we first assign each tracking agent into the action state sub task, that is, each tracking intelligence first is given to each tracking intelligence. The learning algorithm divides the learning process into two parts, one is the strategy of learning task allocation and the other is the strategy of learning action selection. Each tracking intelligent body shares the learned knowledge by sharing the Q function, and improves the learning process. Learning efficiency. Experimental results verify the effectiveness of the method.
【學(xué)位授予單位】：長(zhǎng)春理工大學(xué)
【學(xué)位級(jí)別】：博士
【學(xué)位授予年份】：2016
【分類號(hào)】：TP391.41

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 唐偉力;龍建忠;;一種基于降雨模型的圖像分割方法在礫巖圖像分割中的應(yīng)用[J];成都信息工程學(xué)院學(xué)報(bào);2007年02期

2 黃曉莉;曾黃麟;王秀碧;劉永春;;基于脈沖耦合神經(jīng)網(wǎng)絡(luò)的圖像分割[J];信息技術(shù);2008年09期

3 肖飛;綦星光;;圖像分割方法綜述[J];可編程控制器與工廠自動(dòng)化;2009年11期

4 汪一休;;一種交互式圖像分割的修正優(yōu)化方法[J];中國(guó)科學(xué)技術(shù)大學(xué)學(xué)報(bào);2010年02期

5 李丹;;圖像分割方法及其應(yīng)用研究[J];科技信息;2010年36期

6 龔永義;黃輝;于繼明;關(guān)履泰;;基于熵的兩區(qū)域圖像分割[J];中國(guó)圖象圖形學(xué)報(bào);2011年05期

7 張甫;李興來(lái);陳佳君;;淺談圖像分割方法的研究運(yùn)用[J];科技創(chuàng)新與應(yīng)用;2012年04期

8 汪梅;何高明;賀杰;;常見(jiàn)圖像分割的技術(shù)分析與比較[J];計(jì)算機(jī)光盤軟件與應(yīng)用;2013年06期

9 魏慶;盧照敢;邵超;;基于復(fù)雜性指數(shù)的圖像分割必要性判別技術(shù)[J];計(jì)算機(jī)工程與應(yīng)用;2013年16期

10 陳曉丹;李思明;;圖像分割研究進(jìn)展[J];現(xiàn)代計(jì)算機(jī)(專業(yè)版);2013年33期

相關(guān)會(huì)議論文 前10條

1 楊魁;趙志剛;;圖像分割技術(shù)綜述[A];2008年中國(guó)高校通信類院系學(xué)術(shù)研討會(huì)論文集（下冊(cè)）[C];2009年

2 楊暄;郭成安;李建華;;改進(jìn)的脈沖耦合神經(jīng)網(wǎng)絡(luò)及其在圖像分割中的應(yīng)用[A];第十屆全國(guó)信號(hào)處理學(xué)術(shù)年會(huì)（CCSP-2001）論文集[C];2001年

3 楊生友;;圖像分割在醫(yī)學(xué)圖像中應(yīng)用現(xiàn)狀綜述[A];2009中華醫(yī)學(xué)會(huì)影像技術(shù)分會(huì)第十七次全國(guó)學(xué)術(shù)大會(huì)論文集[C];2009年

4 閆平昆;;基于模型的圖像分割技術(shù)及其醫(yī)學(xué)應(yīng)用[A];第十五屆全國(guó)圖象圖形學(xué)學(xué)術(shù)會(huì)議論文集[C];2010年

5 高嵐;胡友為;潘峰;盧凌;;基于小生境遺傳算法的SAR圖像分割[A];可持續(xù)發(fā)展的中國(guó)交通——2005全國(guó)博士生學(xué)術(shù)論壇（交通運(yùn)輸工程學(xué)科）論文集（下冊(cè)）[C];2005年

6 孫莉;張艷寧;胡伏原;趙榮椿;;基于Gaussian-Hermite矩的SAR圖像分割[A];第十三屆全國(guó)圖象圖形學(xué)學(xué)術(shù)會(huì)議論文集[C];2006年

7 李盛;;基于協(xié)同聚類的圖像分割[A];第十四屆全國(guó)圖象圖形學(xué)學(xué)術(shù)會(huì)議論文集[C];2008年

8 張利;許家佗;;舌象圖像分割技術(shù)的研究與應(yīng)用進(jìn)展[A];中華中醫(yī)藥學(xué)會(huì)中醫(yī)診斷學(xué)分會(huì)第十次學(xué)術(shù)研討會(huì)論文集[C];2009年

9 秦昆;李振宇;李輝;李德毅;;基于云模型和格網(wǎng)劃分的圖像分割方法[A];《測(cè)繪通報(bào)》測(cè)繪科學(xué)前沿技術(shù)論壇摘要集[C];2008年

10 高惠琳;竇麗華;陳文頡;謝剛;;圖像分割技術(shù)在醫(yī)學(xué)CT中的應(yīng)用[A];中國(guó)自動(dòng)化學(xué)會(huì)控制理論專業(yè)委員會(huì)A卷[C];2011年

相關(guān)博士學(xué)位論文 前10條

1 白雪飛;基于視覺(jué)顯著性的圖像分割方法研究[D];山西大學(xué);2014年

2 黃萬(wàn)里;基于高分衛(wèi)星數(shù)據(jù)多尺度圖像分割方法的天山森林小班邊界提取研究[D];福建師范大學(xué);2015年

3 王輝;圖像分割的最優(yōu)化和水平集方法研究[D];電子科技大學(xué);2014年

4 高婧婧;腦部MR圖像分割理論研究[D];電子科技大學(xué);2014年

5 潘改;偏微分方程在圖像分割中的應(yīng)用研究[D];東北大學(xué);2013年

6 馮籍瀾;高分辨率SAR圖像分割與分類方法研究[D];電子科技大學(xué);2015年

7 李偉斌;圖像分割中的變分模型與快速算法研究[D];國(guó)防科學(xué)技術(shù)大學(xué);2014年

8 鄧曉政;基于免疫克隆選擇優(yōu)化和譜聚類的復(fù)雜圖像分割[D];西安電子科技大學(xué);2014年

9 帥永e，

本文編號(hào)：2055155