圖像內(nèi)容顯著性檢測的理論和方法研究
發(fā)布時間：2018-11-12 10:31
【摘要】：人類的視覺系統(tǒng)可以在廣大的、復(fù)雜的動態(tài)及靜態(tài)場景中快速定位出最吸引注意的內(nèi)容,這種能力被稱為顯著性檢測。吸引注意的內(nèi)容因此被叫做顯著性(內(nèi)容),一般來說顯著性的某種特征與周圍有很大差異,例如某些危險警示標(biāo)志。視覺系統(tǒng)的顯著性檢測能力可以使我們更聚焦在視覺場景中的某一局部,而對場景中的其它背景部分視而不見,從而當(dāng)我們面對外界的刺激時可以優(yōu)先處理部分信號并更快速的做出反應(yīng)。隨著圖像視頻獲取捕捉設(shè)備的發(fā)展,數(shù)據(jù)規(guī)模越來越大、內(nèi)容越來越復(fù)雜,因此早期的計算機(jī)視覺算法逐漸不能勝任現(xiàn)如今的任務(wù)。所以,人們試圖設(shè)計算法模擬人類視覺系統(tǒng)的顯著性檢測能力,來找出圖像中的部分重要內(nèi)容進(jìn)行后續(xù)分析處理并忽略掉冗余信息,從而加速整個任務(wù)的執(zhí)行。顯著性檢測的結(jié)果可以應(yīng)用于很多計算機(jī)視覺任務(wù),例如物體檢測和識別[9][52][80][81]、圖像分割[84][85][86]、圖像和視頻壓縮[82][83]、圖像重定向[53][79][100]、視頻摘要[88][89][90]、視覺跟蹤[94][95][96][97][98][99]、基于內(nèi)容的圖像檢索[87]、圖像編輯[91][92][93]等。因為顯著性檢測的重要性,它越來越受到重視并被大量研究,很多顯著性檢測模型相繼被提出。在計算機(jī)視覺的顯著性檢測領(lǐng)域,根據(jù)任務(wù)的不同,顯著性檢測有兩個分支：通用的顯著性檢測和特定的顯著性檢測。而根據(jù)結(jié)果的特點,每個分支還可以進(jìn)一步分為視覺顯著性檢測和顯著性物體檢測兩類。通用的顯著性檢測任務(wù),目的是查找自然圖像中吸引人注意的區(qū)域或物體,這些區(qū)域或物體沒有明確的類別。而特定的顯著性檢測,一般是根據(jù)不同任務(wù)在圖像中查找某種類型的區(qū)域或物體,比如照片中的人臉、監(jiān)控中的汽車、醫(yī)學(xué)圖像中的腫瘤。本文全面研究了圖像內(nèi)容顯著性檢測的理論與方法,從多個角度、多個方面分析了現(xiàn)有的顯著性檢測模型,并提出了新的特征、模型和評價方法,對顯著性檢測領(lǐng)域做出了較大貢獻(xiàn)。主要創(chuàng)新點包括：總結(jié)了視覺顯著性檢測模型和顯著性物體檢測模型兩類模型的發(fā)展,并且匯總了兩類模型的顯著性特征、評價數(shù)據(jù)庫和評價方法,發(fā)現(xiàn)這兩類模型有很多相似點。進(jìn)一步分析,兩類模型都有三個主要組成部分：特征對比方式、顯著性提取方向以及線索結(jié)合方法,再次說明了兩類模型的緊密關(guān)系。提出了一種通用的顯著性物體檢測模型UFO[10],在該模型中提出了聚焦度(focusness)和物體度(objectness)兩種顯著性特征,其中聚焦度可以通過尺度空間分析進(jìn)行估計而物體度則由修改的物體檢測算法計算。最后,非線性結(jié)合廣泛使用的獨特度(uniqueness)得到UFO模型。該模型在當(dāng)時國際上最大和最難的顯著性物體檢測數(shù)據(jù)庫MSRA1000和BSD300上,及統(tǒng)一的評價體系下,取得了領(lǐng)先的結(jié)果。提出了一種對基于擴(kuò)散的顯著性物體檢測模型的改進(jìn)方法。通過分析現(xiàn)有的基于擴(kuò)散的顯著性物體檢測模型,我們對該類模型的工作機(jī)制有了一種全新的解釋,我們發(fā)現(xiàn)基于擴(kuò)散的顯著性物體檢測模型的性能與擴(kuò)散矩陣和種子向量都有關(guān),并且性能上限由擴(kuò)散矩陣決定。因此,我們提出了一種通過重新合成擴(kuò)散矩陣和構(gòu)建種子向量來提高模型準(zhǔn)確性和效率的方法。之前大多數(shù)基于擴(kuò)散的顯著性物體檢測模型只關(guān)注于種子向量的生成,但是我們通過大量實驗,包括視覺顯著性提升能力、及我們提出的受限最優(yōu)的種子點效率(COSE),充分證明了我們重新合成的擴(kuò)散矩陣有更強的擴(kuò)散能力,可以使種子向量的顯著性信息更精確的傳播到整個顯著性物體。同時,視覺顯著性提升能力的實驗為改造視覺顯著性檢測模型來檢測顯著性物體提供了一個途徑。最后,我們結(jié)合重新合成的擴(kuò)散矩陣及構(gòu)建的種子向量得到GP模型[11]。我們在當(dāng)時最大的兩個數(shù)據(jù)庫MSRA10K和ECSSD上進(jìn)行顯著性物體檢測的實驗,GP在大多數(shù)評價方法下都取得領(lǐng)先水平。提出了一種特定的顯著性物體檢測模型。具體來說,該模型實現(xiàn)了一個算法,來自動的檢測乳腺超聲圖像中的腫瘤位置并勾繪出腫瘤輪廓。該模型首先利用AdaBoost分類器找出所有潛在的腫瘤區(qū)域,再利用SVM分類器進(jìn)一步把真實腫瘤區(qū)域篩選出來。最后將檢測出的腫瘤區(qū)域及非腫瘤區(qū)域的中心作為前/背景種子點,利用Random Walks分割算法得到腫瘤輪廓。實驗證明,該模型可以準(zhǔn)確定位腫瘤位置并精確勾繪腫瘤輪廓,此外該算法也可以應(yīng)對包含多個腫瘤的超聲圖像。
[Abstract]:The human vision system can quickly position the most attractive content in a large, complex dynamic and static scene, which is called the significance detection. The content of the attraction is therefore called significance (content), and there is a significant difference in some of the characteristics of the general significance, such as some dangerous warning signs. The significance detection capability of the vision system can make us more focused on a local part of the visual scene, and turn a blind eye to other background parts in the scene, so that part of the signal can be preferentially processed and the reaction can be made more quickly when we face the external stimulus. With the development of image video acquisition and capture device, the scale of the data is becoming more and more complex, so the early computer vision algorithm can't be qualified for the current task. Therefore, people try to design the algorithm to simulate the significance detection ability of the human vision system, to find out some important content in the image for subsequent analysis and to ignore the redundant information, so as to accelerate the execution of the whole task. The results of the significance detection can be applied to many computer visual tasks, such as object detection and identification[9][52][80][81], image segmentation[84][85][86], image and video compression[82][83], image redirection[53][79][100],[88][89][90], visual tracking[94][95][96][97][98][99], content-based image retrieval[87], image editing[91][92][93], and the like. Because of the significance of the significance test, it is more and more important and has been extensively studied, and many significant detection models have been proposed successively. In the field of the significance detection of computer vision, there are two branches according to the difference of the task: the general significance detection and the specific significance detection. and according to the characteristics of the result, each branch can be further divided into two types of visual significance detection and saliency object detection. The purpose of the universal significance detection task is to find areas or objects that are noticed by a person in a natural image that has no clear category. and the particular significance detection is generally used to find some type of region or object in the image according to different tasks, such as the human face in the photograph, the automobile in the monitoring, and the tumor in the medical image. In this paper, the theory and method of the significance detection of the image content are comprehensively studied, and the existing significance detection model is analyzed from a plurality of angles and a plurality of aspects, and a new characteristic, a model and an evaluation method are put forward, and a great contribution is made to the field of significance detection. The main innovation points include: the development of two models of the visual significance detection model and the significance object detection model, and the significance characteristics of the two types of models, the evaluation database and the evaluation method are summarized, and the two types of models are found to have many similar points. Further analysis, the two models have three main components: the feature contrast, the significance extraction direction and the lead-binding method, and the close relationship between the two types of models is described again. A general significance object detection model UFO[10] is proposed. In this model, two significant features of focus degree and object degree are proposed, in which the focus degree can be estimated by the scale space analysis, and the object degree is calculated by the modified object detection algorithm. Finally, the unique degree of non-linearity in combination with a wide range of uses results in a UFO model. The model has made a leading result under the unified evaluation system of the world's largest and most difficult-most significant object detection databases, MSRA1000 and BSD300, and the unified evaluation system. An improved method for detecting a significant object based on diffusion is presented. By analyzing the existing diffusion-based saliency object detection model, we have a brand-new interpretation of the working mechanism of this kind of model, and we find that the performance of the diffusion-based saliency object detection model is related to both the diffusion matrix and the seed vector. and the upper performance limit is determined by the diffusion matrix. Therefore, we propose a method to improve the accuracy and efficiency of the model by re-synthesizing the diffusion matrix and constructing the seed vector. Most of the previous diffusion-based significance object detection models focus only on the generation of seed vectors, but we have passed a number of experiments, including visual saliency enhancement, and the limited optimal seed point efficiency (COSE) we propose, It is proved that the diffusion matrix of the re-synthesis has stronger diffusion ability, and the significance information of the seed vector can be more accurately transmitted to the whole significant object. At the same time, the experiment of the visual significance enhancement capability provides a way to transform the visual saliency detection model to detect the significant object. Finally, we combine the re-synthesized diffusion matrix and the constructed seed vector to get the GP model[11]. The experiment of significant object detection on the two largest databases, MRA10K and ECSSD, was the leading level of GP under most of the evaluation methods. A particular significance object detection model is proposed. In particular, the model implements an algorithm to automatically detect the tumor location in the breast ultrasound image and to map out the tumor profile. The model first uses the AdaBoost classifier to find all potential tumor regions, and further filters the real tumor region by using the SVM classifier. and finally, the detected tumor region and the center of the non-tumor region are taken as a front/ background seed point, and a tumor profile is obtained by using the random Walks segmentation algorithm. The experimental results show that the model can accurately position the tumor position and draw the tumor contour accurately, and the algorithm can also deal with the ultrasound image containing multiple tumors.
【學(xué)位授予單位】：山東大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2016
【分類號】：TP391.41
[Abstract]:The human vision system can quickly position the most attractive content in a large, complex dynamic and static scene, which is called the significance detection. The content of the attraction is therefore called significance (content), and there is a significant difference in some of the characteristics of the general significance, such as some dangerous warning signs. The significance detection capability of the vision system can make us more focused on a local part of the visual scene, and turn a blind eye to other background parts in the scene, so that part of the signal can be preferentially processed and the reaction can be made more quickly when we face the external stimulus. With the development of image video acquisition and capture device, the scale of the data is becoming more and more complex, so the early computer vision algorithm can't be qualified for the current task. Therefore, people try to design the algorithm to simulate the significance detection ability of the human vision system, to find out some important content in the image for subsequent analysis and to ignore the redundant information, so as to accelerate the execution of the whole task. The results of the significance detection can be applied to many computer visual tasks, such as object detection and identification[9][52][80][81], image segmentation[84][85][86], image and video compression[82][83], image redirection[53][79][100],[88][89][90], visual tracking[94][95][96][97][98][99], content-based image retrieval[87], image editing[91][92][93], and the like. Because of the significance of the significance test, it is more and more important and has been extensively studied, and many significant detection models have been proposed successively. In the field of the significance detection of computer vision, there are two branches according to the difference of the task: the general significance detection and the specific significance detection. and according to the characteristics of the result, each branch can be further divided into two types of visual significance detection and saliency object detection. The purpose of the universal significance detection task is to find areas or objects that are noticed by a person in a natural image that has no clear category. and the particular significance detection is generally used to find some type of region or object in the image according to different tasks, such as the human face in the photograph, the automobile in the monitoring, and the tumor in the medical image. In this paper, the theory and method of the significance detection of the image content are comprehensively studied, and the existing significance detection model is analyzed from a plurality of angles and a plurality of aspects, and a new characteristic, a model and an evaluation method are put forward, and a great contribution is made to the field of significance detection. The main innovation points include: the development of two models of the visual significance detection model and the significance object detection model, and the significance characteristics of the two types of models, the evaluation database and the evaluation method are summarized, and the two types of models are found to have many similar points. Further analysis, the two models have three main components: the feature contrast, the significance extraction direction and the lead-binding method, and the close relationship between the two types of models is described again. A general significance object detection model UFO[10] is proposed. In this model, two significant features of focus degree and object degree are proposed, in which the focus degree can be estimated by the scale space analysis, and the object degree is calculated by the modified object detection algorithm. Finally, the unique degree of non-linearity in combination with a wide range of uses results in a UFO model. The model has made a leading result under the unified evaluation system of the world's largest and most difficult-most significant object detection databases, MSRA1000 and BSD300, and the unified evaluation system. An improved method for detecting a significant object based on diffusion is presented. By analyzing the existing diffusion-based saliency object detection model, we have a brand-new interpretation of the working mechanism of this kind of model, and we find that the performance of the diffusion-based saliency object detection model is related to both the diffusion matrix and the seed vector. and the upper performance limit is determined by the diffusion matrix. Therefore, we propose a method to improve the accuracy and efficiency of the model by re-synthesizing the diffusion matrix and constructing the seed vector. Most of the previous diffusion-based significance object detection models focus only on the generation of seed vectors, but we have passed a number of experiments, including visual saliency enhancement, and the limited optimal seed point efficiency (COSE) we propose, It is proved that the diffusion matrix of the re-synthesis has stronger diffusion ability, and the significance information of the seed vector can be more accurately transmitted to the whole significant object. At the same time, the experiment of the visual significance enhancement capability provides a way to transform the visual saliency detection model to detect the significant object. Finally, we combine the re-synthesized diffusion matrix and the constructed seed vector to get the GP model[11]. The experiment of significant object detection on the two largest databases, MRA10K and ECSSD, was the leading level of GP under most of the evaluation methods. A particular significance object detection model is proposed. In particular, the model implements an algorithm to automatically detect the tumor location in the breast ultrasound image and to map out the tumor profile. The model first uses the AdaBoost classifier to find all potential tumor regions, and further filters the real tumor region by using the SVM classifier. and finally, the detected tumor region and the center of the non-tumor region are taken as a front/ background seed point, and a tumor profile is obtained by using the random Walks segmentation algorithm. The experimental results show that the model can accurately position the tumor position and draw the tumor contour accurately, and the algorithm can also deal with the ultrasound image containing multiple tumors.
【學(xué)位授予單位】：山東大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2016
【分類號】：TP391.41
【相似文獻(xiàn)】
相關(guān)期刊論文 前10條
1 楊清;;電網(wǎng)異常檢測模型方法設(shè)計[J];電測與儀表;2009年S2期
2 紀(jì)祥敏;陳秋妹;景林;;面向下一代互聯(lián)網(wǎng)的異常檢測模型研究[J];福建電腦;2013年01期
3 崔艷娜;;一種網(wǎng)絡(luò)流量異常檢測模型[J];計算機(jī)與現(xiàn)代化;2013年08期
4 涂旭平;金海;何麗莉;楊志玲;陶智飛;;一種新的網(wǎng)絡(luò)異常流量檢測模型[J];計算機(jī)科學(xué);2005年08期
5 呂洪柱;張建平;鄧文新;;基于數(shù)據(jù)挖掘技術(shù)的異常檢測模型設(shè)計[J];高師理科學(xué)刊;2007年06期
6 馬琳;蘇一丹;莫錦萍;;協(xié)同推薦系統(tǒng)檢測模型的一種優(yōu)化方法[J];微計算機(jī)信息;2010年03期
7 楊清;;基于模糊序列電網(wǎng)異常檢測建模方法與研究[J];山西電子技術(shù);2009年05期
8 李雪琴;;基于模糊C均值的異常流量檢測模型[J];贛南師范學(xué)院學(xué)報;2009年06期
9 唐彰國;李煥洲;鐘明全;張健;;改進(jìn)的進(jìn)程行為檢測模型及實現(xiàn)[J];計算機(jī)應(yīng)用;2010年01期
10 申利民;李峰;孫鵬飛;牛景春;;開放企業(yè)計算環(huán)境下基于信任的行為檢測模型[J];計算機(jī)集成制造系統(tǒng);2013年01期
相關(guān)會議論文 前7條
1 劉俊榮;王文槿;劉寶旭;;一種基于網(wǎng)絡(luò)行為分析的木馬檢測模型[A];第十六屆全國核電子學(xué)與核探測技術(shù)學(xué)術(shù)年會論文集（下冊）[C];2012年
2 馬文忠;郭江艷;陳科成;楊珊;王艷麗;;基于神經(jīng)網(wǎng)絡(luò)的供熱燃燒系統(tǒng)檢測模型的研究[A];2011中國電工技術(shù)學(xué)會學(xué)術(shù)年會論文集[C];2011年
3 張廣軍;賀俊吉;;基于圓結(jié)構(gòu)光的內(nèi)表面三維視覺檢測模型[A];中國儀器儀表學(xué)會學(xué)術(shù)論文集[C];2004年
4 王建平;張自立;魏華;;戰(zhàn)術(shù)空域沖突檢測模型研究[A];Proceedings of 14th Chinese Conference on System Simulation Technology & Application(CCSSTA’2012)[C];2012年
5 武照東;劉英凱;劉春;吳秀峰;;Overlay網(wǎng)絡(luò)的鏈路故障檢測模型[A];2008通信理論與技術(shù)新發(fā)展——第十三屆全國青年通信學(xué)術(shù)會議論文集（下）[C];2008年
6 李京鵬;楊林;劉世棟;;防火墻狀態(tài)檢測模型研究[A];第十八次全國計算機(jī)安全學(xué)術(shù)交流會論文集[C];2003年
7 周雙娥;熊國平;;基于Petri網(wǎng)的故障檢測模型的設(shè)計與分析[A];第六屆中國測試學(xué)術(shù)會議論文集[C];2010年
相關(guān)博士學(xué)位論文 前5條
1 蔣鵬;圖像內(nèi)容顯著性檢測的理論和方法研究[D];山東大學(xué);2016年
2 趙靜;網(wǎng)絡(luò)協(xié)議異常檢測模型的研究與應(yīng)用[D];北京交通大學(xué);2010年
3 趙斌;基于圖模型的微博數(shù)據(jù)分析與管理[D];華東師范大學(xué);2012年
4 牛清寧;基于信息融合的疲勞駕駛檢測方法研究[D];吉林大學(xué);2014年
5 劉鵬飛;鋁合金點焊質(zhì)量的逆過程檢測方法研究[D];天津大學(xué);2008年
相關(guān)碩士學(xué)位論文 前10條
1 朱遠(yuǎn)文;前端啟發(fā)式滲透檢測模型研究[D];天津理工大學(xué);2015年
2 劉嬌;基于高光譜技術(shù)的不同品種豬肉品質(zhì)檢測模型傳遞方法研究[D];華中農(nóng)業(yè)大學(xué);2015年
3 軒照光;ITS系統(tǒng)防碰撞技術(shù)研究[D];電子科技大學(xué);2015年
4 祝e，
本文編號：2326853
本文鏈接：http://sikaile.net/shoufeilunwen/xxkjbs/2326853.html
最近更新
教材專著
