本文選題：圖像標(biāo)注　切入點：多標(biāo)記學(xué)習(xí)　出處：《山東師范大學(xué)》2016年博士論文

【摘要】：計算機技術(shù)和移動拍照技術(shù)快速發(fā)展,網(wǎng)絡(luò)空間中的圖像信息爆炸式增長。為滿足人們對圖像的檢索,研究人員提出了大量的圖像檢索算法。圖像檢索方法可以分為三類,分別是基于文本的圖像檢索、基于內(nèi)容的圖像檢索和基于語義的圖像檢索。其中,基于語義的圖像檢索系統(tǒng)中的核心技術(shù)是圖像的語義標(biāo)注。本文的重點研究了圖像標(biāo)注的技術(shù)問題。研究人員已經(jīng)提出了大量的圖像標(biāo)注算法,但語義鴻溝問題、維數(shù)災(zāi)難問題、數(shù)據(jù)不平衡問題等重要的問題仍然沒有從根本上得到解決。針對上述問題,本文基于多標(biāo)記學(xué)習(xí)框架,改進了四種經(jīng)典的機器學(xué)習(xí)方法用于圖像標(biāo)注,取得了很好的實驗效果:1.基于懶惰學(xué)習(xí)的多標(biāo)記圖像標(biāo)注算法ML-KNN在計算貝葉斯最大化后驗概率時,只使用了主樣例與近鄰樣例在數(shù)量上的相關(guān)性,沒有考慮主樣例與近鄰樣例在距離上的相關(guān)性。本文把上述兩種相關(guān)性同時考慮,提出了一種改進的基于懶惰學(xué)習(xí)的多標(biāo)記圖像標(biāo)注算法ML-WKNN。在Image和Yeast兩個經(jīng)典多標(biāo)記數(shù)據(jù)集上的實驗結(jié)果表明,ML-WKNN算法比其它四個經(jīng)典的多標(biāo)記算法的總體標(biāo)注效果更好。2.在基于樸素貝葉斯理論的多標(biāo)記樸素貝葉斯算法MLNB中,使用主成分分析方法預(yù)處理樣本的屬性特征。處理之后的樣例屬性之間是不相關(guān)的,但是仍然不能滿足樸素貝葉斯算法需要屬性特征相互獨立的要求。本文中我們使用獨立成分分析方法來預(yù)處理樣例的屬性特征,處理之后的樣例屬性特征之間是相互獨立的,符合樸素貝葉斯算法對于樣例屬性特征的要求。在Image和Yeast兩個經(jīng)典多標(biāo)記數(shù)據(jù)集上的實驗結(jié)果表明,IMLNB算法的在多個評價指標(biāo)上的綜合標(biāo)注效果比其它四個經(jīng)典多標(biāo)記算法更好。3.基于改進構(gòu)建類屬屬性的思想,本文提出了一種改進的多標(biāo)記圖像標(biāo)注算法LTFML。LTFML只使用每個類標(biāo)記的正樣例為每個類標(biāo)記構(gòu)建類屬屬性,并使用一種新的評價函數(shù)對不同類屬屬性聚類簇的進行加權(quán)。在Image和Yeast兩個經(jīng)典多標(biāo)記數(shù)據(jù)集上的實驗結(jié)果表明,LTFML算法的標(biāo)注效果在五個評價指標(biāo)上整體最優(yōu)。4.針對多標(biāo)記圖像標(biāo)注中常見的數(shù)據(jù)不平衡問題,本文對Bagging算法進行改進,提出多標(biāo)記圖像標(biāo)注集成學(xué)習(xí)方法BM3。該算法使用Bagging方法對每個類標(biāo)記的正負樣例分別抽取相等數(shù)量的樣例,然后組成規(guī)模相對較小且正負樣例完全平衡的訓(xùn)練子集。對基分類器的預(yù)測結(jié)果集成時,本文使用了一種新的融合策略—最小最大模塊化方法。在Image和Yeast兩個經(jīng)典多標(biāo)記數(shù)據(jù)集上的實驗結(jié)果表明,3BM算法整體標(biāo)注結(jié)果比BR等經(jīng)典的多標(biāo)記算法的結(jié)果更好。
[Abstract]:With the rapid development of computer technology and mobile photography technology and the explosive growth of image information in cyberspace, a large number of image retrieval algorithms have been proposed by researchers to meet the needs of image retrieval. Image retrieval methods can be divided into three categories. They are text-based image retrieval, content-based image retrieval and semantic-based image retrieval. Semantic tagging is the key technology in semantic-based image retrieval system. This paper focuses on the technical problems of image tagging. Researchers have proposed a large number of image tagging algorithms, but the semantic gap problem. Some important problems, such as dimensionality disaster and data imbalance, have not been solved fundamentally. In view of the above problems, this paper improves four classical machine learning methods for image tagging based on multi-label learning framework. Good experimental results are obtained: 1. ML-KNN, a lazy learning based multi-label image tagging algorithm, only uses the quantitative correlation between the main sample and the nearest neighbor sample when calculating Bayesian maximization posteriori probability. The distance correlation between the main sample and the nearest neighbor sample is not considered. In this paper, the above two correlations are considered at the same time. An improved multi-label image tagging algorithm ML-WKNN based on lazy learning is proposed. The experimental results on two classical multi-label datasets Image and Yeast show that the ML-WKNN algorithm is more effective than the other four classical multi-label algorithms. Better .2. in MLNB, a multi-label naive Bayesian algorithm based on naive Bayesian theory, The principal component analysis (PCA) method is used to preprocess the attribute characteristics of the sample. In this paper, we use independent component analysis (ICA) method to preprocess the attribute features of the sample, which is independent of each other. The experimental results on two classical multi-label datasets of Image and Yeast show that the algorithm has more comprehensive labeling effect on multiple evaluation indexes than the other four classical ones. The tagging algorithm is better. 3. Based on the idea of improving the construction of generic attributes, In this paper, an improved multi-label image tagging algorithm, LTFML.LTFML, is proposed to construct class attributes for each class tag using only positive samples of each class tag. Using a new evaluation function, the clustering of different generic attributes is weighted. The experimental results on two classical multi-label data sets, Image and Yeast, show that the tagging effect of the algorithm is the best in five evaluation indexes. Aiming at the problem of data imbalance in multi-label image annotation, In this paper, the Bagging algorithm is improved, and an integrated learning method BM3 is proposed. The algorithm uses the Bagging method to extract an equal number of positive and negative samples from each class label. Then the training subset with relatively small scale and complete balance of positive and negative samples is formed. When the prediction results of the base classifier are integrated, In this paper, a new fusion strategy, minimum maximum modularization method, is used. The experimental results on two classical multi-label datasets, Image and Yeast, show that the global labeling result of the algorithm is better than that of classical multi-label algorithms such as Br.
【學(xué)位授予單位】：山東師范大學(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】

相關(guān)期刊論文 前10條

1 孟勇,洪丹輝,毛丹;測度熵在圖像紋理分析中的應(yīng)用[J];計算機應(yīng)用與軟件;2000年08期

2 吳濤;秦昆;;圖像紋理特征數(shù)據(jù)挖掘的理論與方法探討[J];計算機時代;2006年08期

3 方玲玲;王相海;;圖像挖掘研究[J];計算機科學(xué);2009年08期

4 高振宇;楊曉梅;龔劍明;金海;;圖像復(fù)雜度描述方法研究[J];中國圖象圖形學(xué)報;2010年01期

5 劉勇,施萬昌,徐玉蘭;圖像差異的分析與識別[J];復(fù)旦學(xué)報(自然科學(xué)版);2000年05期

6 羅l，

本文編號：1690271