【摘要】：圖像分類識別主要是從原始圖像里劃分興趣區(qū)域并進行準確分割,并在此基礎(chǔ)上進行分類識別任務(wù)。近年來計算機視覺與模式識別特別是卷積神經(jīng)網(wǎng)絡(luò)的發(fā)展,為分類識別提供了良好的技術(shù)支持。由于圖像分類識別在視頻監(jiān)控、人臉識別、圖像分類檢索等方面有著廣泛的應(yīng)用前景,因此越來越受到計算機視覺領(lǐng)域研究者的廣泛關(guān)注與研究。卷積神經(jīng)網(wǎng)絡(luò)是近年發(fā)展起來,并引起廣泛重視的一種圖像分類方法,傳統(tǒng)識別方法需要訓(xùn)練大量網(wǎng)絡(luò)參數(shù),造成了訓(xùn)練時間的增加和網(wǎng)絡(luò)的過擬合;輸入集需要進行前期預(yù)處理,丟失了圖像的原有特征。與傳統(tǒng)方法不同,卷積神經(jīng)網(wǎng)絡(luò)不需要針對特定的任務(wù)采集圖像的特征,而是模擬人類的視覺系統(tǒng)層次化、抽象的產(chǎn)生分類結(jié)果,卷積神經(jīng)網(wǎng)絡(luò)創(chuàng)新的采用了局部感受野,權(quán)值共享,卷積采樣技術(shù),減少了網(wǎng)絡(luò)的訓(xùn)練參數(shù)數(shù)量,提高了識別速度,使得其在圖像識別領(lǐng)域得到了廣泛應(yīng)用。本文從神經(jīng)網(wǎng)絡(luò)的基本概念和算法入手,深入研究神經(jīng)網(wǎng)絡(luò)理論,進而研究卷積神經(jīng)網(wǎng)絡(luò),通過闡述常見卷積神經(jīng)網(wǎng)絡(luò)的不足,在傳統(tǒng)卷積神經(jīng)網(wǎng)絡(luò)上修改網(wǎng)絡(luò)結(jié)構(gòu),提出了基于動態(tài)隨機卷積神經(jīng)網(wǎng)絡(luò),并基于此理論進一步開展手寫數(shù)字識別方向的研究,最后通過實驗驗證其網(wǎng)絡(luò)模型的有效性和實用性。論文的主要工作如下:(1)整理和總結(jié)了近年來闡述了圖像識別的研究背景和國內(nèi)外研究現(xiàn)狀,特別是卷積神經(jīng)網(wǎng)絡(luò)的國內(nèi)外發(fā)展現(xiàn)狀,介紹了神經(jīng)網(wǎng)絡(luò)和卷積神經(jīng)網(wǎng)絡(luò)的基本概念,詳細闡述了網(wǎng)絡(luò)框架和網(wǎng)絡(luò)參數(shù),包括網(wǎng)絡(luò)的卷積層,池化層和梯度下降訓(xùn)練方法。(2)針對傳統(tǒng)卷積神經(jīng)網(wǎng)絡(luò)對于原始圖像大小的局限性,本文提出了一種動態(tài)隨機卷積神經(jīng)網(wǎng)絡(luò)結(jié)構(gòu)和一種隨機池化方法,避免了原始圖像大小的局限性,更大程度了保留了圖像的紋理特征和局部特征。實驗結(jié)果表明,改進的卷積神經(jīng)網(wǎng)絡(luò)精度優(yōu)于傳統(tǒng)卷積神經(jīng)網(wǎng)絡(luò)。(3)對全文做總結(jié),提出了自身的不足和未來的研究方向。
[Abstract]:Image classification and recognition is mainly based on dividing the region of interest from the original image and accurately segmenting, and on the basis of this, the task of classification and recognition is carried out. In recent years, the development of computer vision and pattern recognition, especially convolution neural network, provides a good technical support for classification recognition. Because image classification and recognition have wide application prospects in video surveillance, face recognition, image classification and retrieval, more and more researchers in the field of computer vision pay more and more attention to it. Convolution neural network is a kind of image classification method which has been developed in recent years and has attracted wide attention. Traditional recognition methods need to train a large number of network parameters, which results in the increase of training time and the over-fitting of network. The input set needs to be preprocessed and the original feature of the image is lost. Different from the traditional methods, the convolution neural network does not need to capture the characteristics of the image for a specific task, but simulates the human visual system hierarchy, abstractly produces the classification result, the convolutional neural network innovatively adopts the local receptive field. Weight sharing and convolution sampling technology reduce the number of network training parameters and improve the recognition speed. It is widely used in the field of image recognition. In this paper, the basic concept and algorithm of neural network are introduced, the theory of neural network is deeply studied, and then the convolutional neural network is studied. By expounding the deficiency of common convolutional neural network, the network structure is modified on the traditional convolution neural network. Based on the dynamic random convolution neural network, the recognition direction of handwritten numerals is further studied based on this theory. Finally, the validity and practicability of the network model are verified by experiments. The main work of this paper is as follows: (1) the research background of image recognition in recent years and the current research situation at home and abroad, especially the development of convolution neural network, are summarized. The basic concepts of neural network and convolutional neural network are introduced. The network framework and network parameters, including the convolution layer of the network, are described in detail. (2) in view of the limitation of traditional convolution neural network to the original image size, a dynamic random convolution neural network structure and a random pool method are proposed. The limitation of the original image size is avoided, and the texture feature and local feature of the image are preserved to a greater extent. The experimental results show that the accuracy of the improved convolution neural network is better than that of the traditional convolution neural network. (3) the paper summarizes the full text and puts forward its own shortcomings and future research directions.
【學(xué)位授予單位】：吉林大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP391.41;TP183

【參考文獻】

相關(guān)期刊論文 前6條

1 俱瑩;;前向神經(jīng)網(wǎng)絡(luò)中的遞推學(xué)習(xí)算法[J];無線電工程;2015年04期

2 常建娥;蔣太立;;層次分析法確定權(quán)重的研究[J];武漢理工大學(xué)學(xué)報(信息與管理工程版);2007年01期

3 邱德紅,陳傳波;融合無監(jiān)督和監(jiān)督學(xué)習(xí)策略生成的多分類決策樹[J];小型微型計算機系統(tǒng);2004年04期

4 宋衛(wèi)榮,翁桂榮;BP網(wǎng)絡(luò)激勵函數(shù)的研究[J];蘇州大學(xué)學(xué)報(自然科學(xué));2002年04期

5 鐘曉;馬少平;張鈸;俞瑞釗;;數(shù)據(jù)挖掘綜述[J];模式識別與人工智能;2001年01期

6 李曉峰,劉光中;人工神經(jīng)網(wǎng)絡(luò)BP算法的改進及其應(yīng)用[J];四川大學(xué)學(xué)報(工程科學(xué)版);2000年02期

，

本文編號：2303366