本文選題：深度學習　切入點：自編碼機　出處：《武漢大學》2017年碩士論文

【摘要】：近年來,隨著成像技術的發(fā)展,圖像、視頻等數(shù)據(jù)越來越豐富,如何從海量數(shù)據(jù)中提取有效信息也成為了一個難題。深度學習已被證明是有效的解決途徑。深度神經(jīng)網(wǎng)絡是一種多層次特征學習模型,能夠自動從原始視覺數(shù)據(jù)中提取出較為抽象的特征,用于后續(xù)的圖像分類、目標檢測等工作。然而,深度神經(jīng)網(wǎng)絡的訓練往往需要大量的標記樣本,并且模型參數(shù)數(shù)量龐大,容易過擬合。為了解決這些問題,本文基于前人的模型,提出了改進的深度神經(jīng)網(wǎng)絡模型以及防止過擬合的方法,以提高深度神經(jīng)網(wǎng)絡在圖像識別中的性能。本文的創(chuàng)新點如下:(1)提出一種新型的非監(jiān)督神經(jīng)網(wǎng)絡模型——深度卷積降噪自編碼機,從未標記的圖像樣本中學習有效的特征表達,從而改善了當前主流深度網(wǎng)絡訓練需要大量標記樣本的問題。(2)提出了一種新的正則化方法——結構化去相關約束,能夠有效地規(guī)范化神經(jīng)網(wǎng)絡模型,防止模型陷入過擬合,同時使得模型學習結構化和不冗余的特征表達,極大地提升了網(wǎng)絡模型的特征學習能力和圖像分類能力。
[Abstract]:In recent years, with the development of imaging technology, image, video and other data are more and more abundant. How to extract effective information from massive data has also become a difficult problem. Deep learning has been proved to be an effective solution. Depth neural network is a multi-level feature learning model. Abstract features can be automatically extracted from the original visual data for subsequent image classification and target detection. However, the training of depth neural networks often requires a large number of labeled samples and a large number of model parameters. In order to solve these problems, an improved depth neural network model and a method to prevent over-fitting are proposed based on the previous models. In order to improve the performance of depth neural network in image recognition, the innovation of this paper is as follows: 1) A new unsupervised neural network model-depth convolution noise reduction self-coding machine is proposed, which can learn effective feature expression in unmarked image samples. Therefore, the problem of large number of tagged samples is improved in the training of current mainstream deep network. A new regularization method, structured decorrelation constraint, is proposed, which can effectively standardize the neural network model and prevent the model from falling into overfitting. At the same time, it makes the model learning structure and non-redundant feature representation, greatly improve the network model feature learning ability and image classification ability.
【學位授予單位】：武漢大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：TP391.41;TP18

【相似文獻】

相關期刊論文 前2條

1 胡宏宇;王殿海;李志慧;曲昭偉;魏巍;;基于視頻的車輛特征表達與分類算法[J];交通與計算機;2008年06期

2 ;[J];;年期

相關博士學位論文 前1條

1 施建宇;蛋白質亞細胞定位特征表達與分類算法研究[D];西北工業(yè)大學;2006年

相關碩士學位論文 前1條

1 熊維;面向視覺特征表達的深度學習算法研究[D];武漢大學;2017年

，

本文編號：1694376