本文選題：目標(biāo)精細(xì)檢索 + 精細(xì)標(biāo)注　； 參考：《北京交通大學(xué)》2016年博士論文

【摘要】：隨著圖像采集設(shè)備的普及和移動互聯(lián)網(wǎng)的飛速發(fā)展,圖像數(shù)量呈現(xiàn)爆炸式增長,如何快速準(zhǔn)確的在海量的圖像數(shù)據(jù)中進(jìn)行目標(biāo)檢索,是近年來計算機視覺領(lǐng)域的一個研究熱點,具有十分重要的學(xué)術(shù)意義和應(yīng)用價值。而隨著用戶對于檢索要求的不斷提高,目標(biāo)精細(xì)檢索系統(tǒng)也開始進(jìn)入人們的視野。通常來說,目標(biāo)精細(xì)檢索系統(tǒng)可以從兩方面進(jìn)行定義：(1)能夠生成更加精細(xì)的圖像標(biāo)注信息。更精細(xì)的標(biāo)注包括物體區(qū)域的像素級標(biāo)注(分割信息),以及物體部位的標(biāo)注信息。這些精細(xì)的標(biāo)注信息允許檢索系統(tǒng)返回更加精細(xì)的檢索結(jié)果：(2)能夠理解用戶更加精細(xì)的檢索意圖描述。例如用戶以手繪草圖作為檢索輸入,該草圖描述著檢索目標(biāo)的形狀細(xì)節(jié)、姿態(tài)、角度等信息。對精細(xì)檢索意圖的理解允許檢索系統(tǒng)返回和用戶輸入高度匹配的目標(biāo)�？偟膩碚f,相比于傳統(tǒng)的目標(biāo)檢索系統(tǒng),目標(biāo)精細(xì)檢索系統(tǒng)能夠返回更加符合用戶需求的檢索結(jié)果,避免用戶對檢索結(jié)果進(jìn)行二次處理和篩選,滿足用戶精細(xì)化的檢索需求,大大提高目標(biāo)檢索的效率,具有非常重要的意義。本文的工作以目標(biāo)精細(xì)檢索為目標(biāo),從以上兩個方面入手進(jìn)行研究,取得了以下成果：(1)針對目標(biāo)標(biāo)注中的目標(biāo)多樣性和像素級標(biāo)注問題,本文提出了一種基于超像素(superpixel)和改進(jìn)與或圖(AND/OR Graph)模型的目標(biāo)標(biāo)注方法。目標(biāo)物體在外觀、姿態(tài)上的多樣性,會顯著降低目標(biāo)標(biāo)注的性能,增加像素級標(biāo)注的難度。針對這個問題,本文將目標(biāo)物體定義為一系列部位的組合,提出一種改進(jìn)的與或圖模型來組織部位之間的關(guān)系,以提高對于外觀和姿態(tài)變化的魯棒性,并利用基于圖模型的快速推理算法實現(xiàn)對物體部位的最優(yōu)選擇。在生成候選部位集合的過程中,考慮到像素級標(biāo)注的要求,本文以超像素區(qū)域的輪廓形狀作為特征,基于模板庫來實現(xiàn)候選物體部位集合的生成。超像素和改進(jìn)與或圖模型的結(jié)合,使得本文的方法對于目標(biāo)多樣性具有較好的魯棒性,并且能夠?qū)崿F(xiàn)目標(biāo)的像素級標(biāo)注。在多個公共數(shù)據(jù)庫上的實驗結(jié)果證明了本文的方法能夠有效的應(yīng)對目標(biāo)多樣性問題,實現(xiàn)目標(biāo)區(qū)域的精細(xì)(像素級)標(biāo)注。(2)針對目標(biāo)部位標(biāo)注中的魯棒性問題,本文提出了一種基于輪廓預(yù)測及增強的目標(biāo)部位標(biāo)注方法。相較于目標(biāo)整體,目標(biāo)部位具有形變較小的優(yōu)點,但同時也具有有效特征少,易受噪聲干擾的問題�；谶@些特點,本文通過增強物體部位的輪廓邊緣來提高目標(biāo)部位標(biāo)注對于噪聲干擾的魯棒性。本文利用學(xué)習(xí)算法從正樣本集中自動的學(xué)習(xí)一組典型的輪廓邊緣模式(edge patterns)。基于學(xué)習(xí)得到的輪廓模式,本文提出一種“輪廓預(yù)測-增強”策略對輸入圖像進(jìn)行過濾,預(yù)測圖像中可能存在的物體部位輪廓邊緣,根據(jù)預(yù)測結(jié)果在增強物體部位輪廓邊緣的同時抑制噪聲邊緣,以達(dá)到提高部位標(biāo)注魯棒性的目的。INRIA和TUD數(shù)據(jù)庫上的實驗結(jié)果表明了本文的方法的確有效的提高了目標(biāo)部位標(biāo)注的魯棒性。(3)針對手繪草圖檢索中的噪聲問題,本文提出了一種輪廓邊緣選擇算法。由于自然圖像中存在的大量噪聲,手繪草圖和自然圖像之間存在巨大的視覺差異。如何有效的降低噪聲邊緣的影響,是提高檢索系統(tǒng)性能的一個關(guān)鍵點。本文將手繪目標(biāo)圖像和邊緣圖像(自然圖像經(jīng)邊緣檢測生成)視為一系列線段的組合,提出了一個HLR (histogram of line relationship)描述子通過描述線段之間的關(guān)系來描述物體形狀。因為邊緣圖像中包含大量的噪聲邊緣,如物體細(xì)節(jié)邊緣和背景邊緣,基于HLR描述子,本文對邊緣進(jìn)行選擇,保留物體輪廓邊緣,忽略噪聲邊緣。該算法為每一個HLR描述子生成大量假設(shè),每種假設(shè)對應(yīng)一種邊緣選擇的結(jié)果,最終將邊緣選擇問題轉(zhuǎn)化為一個尋找最佳假設(shè)組合的最優(yōu)化問題。相應(yīng)的,本文提出一個快速算法來求解這個最優(yōu)化問題。實驗表明,HLR描述子和邊緣選擇算法都有效的提高了檢索性能,增強了檢索系統(tǒng)對于噪聲的魯棒性。(4)針對手繪草圖中的邊緣不穩(wěn)定問題,本文提出了一種最優(yōu)局部匹配算法。自然圖像經(jīng)過邊緣提取不僅會生成噪聲邊緣,也會造成輪廓邊緣丟失,即邊緣不穩(wěn)定問題。這個問題增加了手繪目標(biāo)圖像和自然圖像之間的匹配困難。噪聲邊緣的存在使得邊緣圖像(自然圖像經(jīng)邊緣檢測生成)成為手繪草圖的一個超集,而輪廓邊緣丟失使得邊緣圖像成為手繪草圖的一個子集。于是,本文將手繪圖像和自然圖像之間的匹配問題歸納為一個最優(yōu)局部匹配問題,提出了一個全新的SP (structure point)描述子和層次匹配算法來解決這個問題。SP描述子通過描述線段間的交點來描述物體的局部結(jié)構(gòu)信息。層次匹配算法將SP描述子層次的分解為描述子集合,通過自頂向下的匹配方式來實現(xiàn)SP之間的最優(yōu)局部匹配。在多個數(shù)據(jù)庫上的實驗結(jié)果證明了SP描述子和層次匹配算法對于邊緣不穩(wěn)定現(xiàn)象的有效性。
[Abstract]:With the popularity of image acquisition equipment and the rapid development of mobile Internet, the number of images has been explosively growing. How to quickly and accurately retrieve the target in massive image data is a research hotspot in the field of computer vision in recent years. It has very important significance and application value. The target fine retrieval system has also begun to enter the field of vision. Generally speaking, the target fine retrieval system can be defined from two aspects: (1) it can generate more detailed image annotation information. More detailed annotations include pixel level mark (segmentation information) in the object area, and the annotation information of the object parts. These fine tagging information allows the retrieval system to return to more detailed retrieval results: (2) the ability to understand the user's more detailed description of the retrieval intention. For example, the user uses a hand drawn sketch as the retrieval input, which describes the information of the shape details, posture, and angle of the retrieval target. The understanding of the fine retrieval intention allows the retrieval system. In general, compared to the traditional target retrieval system, the target precision retrieval system can return the retrieval results more consistent with the user's requirements, avoid the user's two processing and screening of the retrieval results, meet the user's fine retrieval requirements, and greatly improve the efficiency of the target retrieval. It is of great significance. In this paper, the following two aspects are studied with the goal of fine target retrieval, and the following results are obtained: (1) aiming at the problem of target diversity and pixel level annotation in target tagging, this paper proposes a target based on superpixel and AND/OR Graph model. The diversity of the object in appearance and attitude can significantly reduce the performance of the target annotation and increase the difficulty of the pixel level annotation. In this paper, the target object is defined as a combination of a series of parts, and an improved and or graph model is proposed to organize the relationship between the parts, in order to improve the appearance and posture change. In the process of generating candidate parts, taking into account the requirement of pixel level annotation in the process of generating candidate parts, this paper takes the contour shape of the super pixel area as the feature, based on the template library to produce the generation of the candidate object set. The combination of entry and graph model makes this method robust to target diversity and can implement pixel level annotation for target. The experimental results on multiple public databases show that the proposed method can effectively deal with the problem of target diversity and realize the fine (pixel level) annotation of the target area. (2) For the problem of robustness in target location, this paper proposes a method of target location based on contour prediction and enhancement. Compared with the whole target, the target position has the advantage of small deformation, but it also has less effective characteristics and easy to be disturbed by noise. Based on these characteristics, this paper strengthens the wheel of the object. The profile edge is used to improve the robustness of the target location for noise interference. This paper uses learning algorithms to automatically learn a group of typical contour edge patterns (edge patterns) from the positive sample set. Based on the learned contour pattern, a "contour prediction enhancement" strategy is proposed to filter the input image and predict the image. The contour edge of the part of the object may exist, and the noise edge is suppressed at the same time on the edge of the contour of the object in accordance with the prediction results. The experimental results on.INRIA and TUD database to improve the robustness of the annotation of the parts show that the method of this paper is indeed effective to improve the robustness of the target location. (3) for hand drawing In this paper, a contour edge selection algorithm is proposed in this paper. Because of the large number of noises in the natural image, there is a huge visual difference between the hand drawn sketch and the natural image. How to effectively reduce the influence of the noise edge is a key point to improve the ability of the retrieval system. This paper will draw the image of the hand-painted target. And edge images (natural images generated by edge detection) as a combination of line segments, a HLR (histogram of line relationship) descriptor is proposed to describe the shape of an object by describing the relationship between the line segments. The edge image contains a large number of noise edges, such as the edge of the object details and the background edge, based on the HLR description. In this paper, the edge is selected, the edge of the object is retained and the edge of the noise is ignored. This algorithm generates a large number of hypotheses for each HLR descriptor. Each hypothesis corresponds to a result of the edge selection. Finally, the edge selection problem is transformed into an optimization problem finding the best hypothesis combination. To solve this optimization problem, the experiment shows that the HLR descriptor and the edge selection algorithm both effectively improve the retrieval performance and enhance the robustness of the retrieval system to the noise. (4) an optimal local matching algorithm is proposed in this paper for the edge instability in the hand drawn sketch. The edge extraction of natural images will not only be generated. The edge of the noise also causes the loss of contour edge, that is, edge instability. This problem increases the difficulty of matching between the hand drawn target image and the natural image. The existence of the edge of the noise makes the edge image (the edge detection of the natural image) become a superset of the hand drawn sketch, and the edge loss makes the edge image become. A subset of hand drawn sketches, this paper sums up the matching problem between hand-painted and natural images as an optimal local matching problem. A new SP (structure point) descriptor and hierarchical matching algorithm are proposed to solve the problem, and the.SP descriptor describes the local structure of the object by describing the intersection point between the line segments. The hierarchical matching algorithm decomposes the SP descriptor hierarchy into the descriptor set, and realizes the optimal local matching between SP by the top-down matching method. The experimental results on multiple databases prove the validity of the SP descriptor and the hierarchical matching algorithm for the edge instability.
【學(xué)位授予單位】：北京交通大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2016
【分類號】：TP391.3

【相似文獻(xiàn)】

相關(guān)期刊論文 前10條

1 王強;孫正興;李曼舞;蔣維;;手繪草圖識別中的相關(guān)反饋方法[J];計算機科學(xué);2005年02期

2 李彬;梁爽;孫正興;;基于空間關(guān)系的手繪草圖檢索[J];計算機科學(xué);2005年12期

3 李曼舞,孫正興,顧慶東;基于輪廓的手繪草圖檢索研究[J];計算機應(yīng)用;2005年02期

4 王衛(wèi);尹建峰;孫正興;;一種手繪草圖的快速參數(shù)化方法[J];計算機科學(xué);2006年01期

5 余淼;;基于手繪草圖的協(xié)同設(shè)計研究進(jìn)展[J];徐州教育學(xué)院學(xué)報;2008年01期

6 陳東帆,王榮航;聯(lián)機手繪草圖的識別原理[J];計算機輔助設(shè)計與圖形學(xué)學(xué)報;1993年02期

7 邢啟江;智能手繪草圖系統(tǒng)的研究[J];中國煤炭經(jīng)濟學(xué)院學(xué)報;1998年03期

8 趙鵬;朱偉偉;吳獻(xiàn)文;劉慧婷;;融合熵和勢能的手繪草圖特征描述方法[J];華南理工大學(xué)學(xué)報(自然科學(xué)版);2014年05期

9 劉峰;劉志鏡;;手繪人臉識別的研究[J];現(xiàn)代電子技術(shù);2006年10期

10 高竹紅;湯進(jìn);羅斌;;基于結(jié)構(gòu)圖的手繪草圖檢索[J];計算機技術(shù)與發(fā)展;2008年03期

相關(guān)會議論文 前5條

1 孫正興;邱慶華;張莉莎;張斌;周若鴻;;在線手繪草圖識別的相關(guān)問題研究[A];第一屆中國情感計算及智能交互學(xué)術(shù)會議論文集[C];2003年

2 孫正興;邱慶華;張莉莎;張斌;周若鴻;;在線手繪草圖識別的相關(guān)問題研究[A];第一屆中國情感計算及智能交互學(xué)術(shù)會議論文集[C];2003年

3 李雪峰;周良;;基于自適應(yīng)頂點探測的手繪草圖識別[A];2008年中國高校通信類院系學(xué)術(shù)研討會論文集（上冊）[C];2009年

4 龔健;費廣正;石民勇;曹瑋;;基于手繪草圖輪廓檢索的簡筆繪圖系統(tǒng)[A];'2008系統(tǒng)仿真技術(shù)及其應(yīng)用學(xué)術(shù)會議論文集[C];2008年

5 欒鵬;;基于骨架手繪草圖的三維建模方法[A];第四屆全國幾何設(shè)計與計算學(xué)術(shù)會議論文集[C];2009年

相關(guān)重要報紙文章 前1條

1 沈于睿;像Jean Nouvel一樣無界徜徉[N];21世紀(jì)經(jīng)濟報道;2008年

相關(guān)博士學(xué)位論文 前2條

1 王樹;圖像中目標(biāo)精細(xì)檢索關(guān)鍵技術(shù)研究[D];北京交通大學(xué);2016年

2 陳雪錦;手繪現(xiàn)實：手繪草圖的真實感解釋[D];中國科學(xué)技術(shù)大學(xué);2008年

相關(guān)碩士學(xué)位論文 前10條

1 唐張鵬;基于手繪草圖的三維人體運動檢索[D];浙江大學(xué);2015年

2 章志青;基于手繪草圖的圖片合成系統(tǒng)設(shè)計與實現(xiàn)[D];南京大學(xué);2013年

3 任帥;基于自適應(yīng)加權(quán)的手繪草圖檢索算法研究與應(yīng)用[D];復(fù)旦大學(xué);2014年

4 王斐;卷積神經(jīng)網(wǎng)絡(luò)在手繪草圖識別中的應(yīng)用研究[D];安徽大學(xué);2016年

5 吳獻(xiàn)文;基于手繪草圖的圖像檢索研究與系統(tǒng)實現(xiàn)[D];安徽大學(xué);2015年

6 李欣;一種基于手繪草圖技術(shù)的建模與動畫系統(tǒng)[D];中國傳媒大學(xué);2007年

7 胡華強;基于手繪草圖的三維CAD系統(tǒng)[D];浙江大學(xué);2004年

8 劉珠峰;基于手繪草圖的三維建模研究[D];南京航空航天大學(xué);2010年

9 溫英濤;手繪草圖界面設(shè)計關(guān)鍵技術(shù)的研究[D];山東大學(xué);2006年

10 楊春雨;基于手繪草圖識別的電原理圖生成[D];東北大學(xué);2011年

，

本文編號：1995761