本文選題：機(jī)器學(xué)習(xí)　切入點(diǎn)：隨機(jī)森林　出處：《湘潭大學(xué)》2017年碩士論文　論文類型：學(xué)位論文

【摘要】：在互聯(lián)網(wǎng)蓬勃發(fā)展的年代,用戶可以隨時(shí)隨地上傳任何圖像。而且隨著智能機(jī)的普及和各種社交平臺(tái)的推廣,圖像分享已成為當(dāng)下潮流。由此可見,圖像數(shù)據(jù)將作為整個(gè)互聯(lián)網(wǎng)信息的重要組成部分,有效理解這些圖像內(nèi)容能夠有助于發(fā)現(xiàn)用戶行為模式,挖掘新知識(shí),在下一代互聯(lián)網(wǎng)競爭中取得制高點(diǎn)。圖像分類作為理解圖像內(nèi)容的重要手段之一,在金融、郵政、公共安全、交通等領(lǐng)域都有成功的應(yīng)用,其重要性不言而喻。對于圖像特征提取算子和分類器的研究一直是圖像分類研究中的重點(diǎn)。特征提取方法和分類器相結(jié)合能夠?qū)υ紙D像進(jìn)行有效的降維處理。而且在有大量樣本的前提下,通過訓(xùn)練樣本調(diào)整特征提取算子和分類器模型的參數(shù)能夠使分類結(jié)果接近最優(yōu)解。目前圖像分類研究存在的主要問題在于:1.對于特征提取來說,人為選擇特定的特征提取算子具有一定的不確定性,而且所提取的圖像特征沒有經(jīng)過篩選,在一定程度上來說,所提取的部分圖像特征也可能干擾分類精度;2.對于分類器來說,要獲得較好的分類準(zhǔn)確率通常需要較長的訓(xùn)練時(shí)間,如何平衡訓(xùn)練時(shí)間和識(shí)別準(zhǔn)確率之間的關(guān)系,在使訓(xùn)練時(shí)間盡可能短的情況下提高分類準(zhǔn)確率是廣大研究者正在考慮的問題。極限學(xué)習(xí)機(jī)是一種訓(xùn)練速度快的分類模型,而且泛化能力強(qiáng),是解決分類問題的一個(gè)較好模型。本文主要的創(chuàng)新點(diǎn)在于提出了一種結(jié)合HOG和隨機(jī)森林(Random Forest)的極限學(xué)習(xí)機(jī)分類模型。通過統(tǒng)計(jì)原始圖像的梯度或邊緣方向,對圖像進(jìn)行HOG特征提取,同時(shí)引入隨機(jī)森林方法,對HOG特征的各個(gè)維度進(jìn)行重要性度量,進(jìn)一步剔除重要性低的冗余信息。最終將經(jīng)過篩選的特征作為極限學(xué)習(xí)機(jī)網(wǎng)絡(luò)的輸入,通過極限學(xué)習(xí)機(jī)進(jìn)行圖像分類。我們在MNIST和USPS數(shù)據(jù)集上進(jìn)行了實(shí)驗(yàn),實(shí)驗(yàn)結(jié)果表明本文方法在訓(xùn)練時(shí)間上優(yōu)于多層極限學(xué)習(xí)機(jī)模型,在準(zhǔn)確率上優(yōu)于傳統(tǒng)的ELM、HOG-ELM和多層ELM。
[Abstract]:In the era of Internet boom, users can upload any image anytime and anywhere. And with the popularity of smartphones and various social platforms, image sharing has become the current trend. Image data will be an important part of the whole Internet information. Understanding these images can help to discover user behavior patterns and to mine new knowledge. Image classification, as one of the important means to understand image content, has been successfully applied in the fields of finance, postal service, public safety, transportation and so on. The importance is self-evident. The research of image feature extraction operator and classifier has always been the focus of image classification. The combination of feature extraction and classifier can effectively reduce the dimension of the original image. And on the premise of having a large number of samples, Adjusting the parameters of feature extraction operator and classifier model by training samples can make the classification result close to the optimal solution. At present, the main problem in image classification research is: 1. For feature extraction, The artificial selection of a specific feature extraction operator is uncertain, and the extracted image features are not filtered. To a certain extent, some of the extracted image features may interfere with the classification accuracy. In order to obtain better classification accuracy, it usually takes longer training time, how to balance the relationship between training time and recognition accuracy, It is a problem that most researchers are considering to improve the classification accuracy under the condition of making the training time as short as possible. The extreme learning machine is a kind of classification model with fast training speed and strong generalization ability. The main innovation of this paper is to put forward a kind of extreme learning machine classification model combining HOG and Random Forest Random Forest. by analyzing the gradient or edge direction of the original image, HOG feature extraction is carried out on the image, and random forest method is introduced to measure the importance of each dimension of the HOG feature. The redundant information of low importance is further eliminated. Finally, the filtered feature is used as the input of the LLM network. The experiments on MNIST and USPS data sets show that the proposed method is superior to the multilayer extreme learning machine model in training time, and the accuracy is better than that of traditional Elmer HOG-ELM and multilayer ELM.
【學(xué)位授予單位】：湘潭大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP391.41

【參考文獻(xiàn)】

相關(guān)期刊論文 前8條

1 楊凱歌;馮學(xué)智;肖鵬峰;朱榴駿;;優(yōu)化子空間SVM集成的高光譜圖像分類[J];遙感學(xué)報(bào);2016年03期

2 DENG ChenWei;HUANG GuangBin;XU Jia;TANG JieXiong;;Extreme learning machines: new trends and applications[J];Science China(Information Sciences);2015年02期

3 張嘉桐;李雪妍;郭樹旭;康建玲;;基于橢圓傅里葉描述子的形狀表示的研究[J];計(jì)算機(jī)工程與應(yīng)用;2014年02期

4 劉麗;匡綱要;;圖像紋理特征提取方法綜述[J];中國圖象圖形學(xué)報(bào);2009年04期

5 馮建輝;楊玉靜;;基于灰度共生矩陣提取紋理特征圖像的研究[J];北京測繪;2007年03期

6 董立巖;苑森淼;劉光遠(yuǎn);賈書洪;;基于貝葉斯分類器的圖像分類[J];吉林大學(xué)學(xué)報(bào)(理學(xué)版);2007年02期

7 邢偉利;圖像檢索中顏色的特征提取及匹配算法[J];微機(jī)發(fā)展;2004年02期

8 閻平凡;對多層前向神經(jīng)網(wǎng)絡(luò)研究的幾點(diǎn)看法[J];自動(dòng)化學(xué)報(bào);1997年01期

相關(guān)博士學(xué)位論文 前1條

1 尚麗;稀疏編碼算法及其應(yīng)用研究[D];中國科學(xué)技術(shù)大學(xué);2006年

相關(guān)碩士學(xué)位論文 前3條

1 張鎧琪;跨媒體搜索關(guān)鍵技術(shù)[D];北京郵電大學(xué);2015年

2 甘露;極限學(xué)習(xí)機(jī)的研究與應(yīng)用[D];西安電子科技大學(xué);2014年

3 楊全銀;基于Hough變換的圖像形狀特征檢測[D];山東大學(xué);2009年

，

本文編號(hào)：1561969