K-Means聚類算法的優(yōu)化及在圖片去重中的應(yīng)用
發(fā)布時間：2018-10-23 10:53
【摘要】：隨著云存儲服務(wù)高速發(fā)展及普及,多媒體數(shù)據(jù)如圖片、視頻等越來越成記錄和分享信息的主要方式。與傳統(tǒng)文字記錄相比,圖片等多媒體數(shù)據(jù)存儲所占用存儲空間要大得許多。因此在應(yīng)對圖片等多媒體數(shù)據(jù)時,如何有效的對圖片去重,減少存儲圖片的容量也是一種新的挑戰(zhàn)。研究觀察發(fā)現(xiàn)在主流的社交網(wǎng)站(如facebook、qq、百度云)中,相似性圖片在總圖片數(shù)量集中占據(jù)很大比例。相似性圖片其定義為經(jīng)過圖片的常見變換,例如圖片的連拍、水印、裁剪、縮放等操作得到的一系列圖片。針對這一發(fā)現(xiàn),設(shè)計出一種圖片去重系統(tǒng)。圖片去重系統(tǒng)可以分為兩部分。第一部分,將圖片集進(jìn)行相似性聚類,對圖片集進(jìn)行基于內(nèi)容的圖片檢索。在圖片檢索技術(shù)方面,先將圖片進(jìn)行預(yù)處理,提取圖片局部特征值,對提取的所有特征值執(zhí)行K-Means聚類算法,將最終的聚類中心作為BOW(Bag-of-Words)模型的視覺單詞,用視覺單詞對SIFT特征點(diǎn)集進(jìn)行量化處理,從而達(dá)到一張圖片只需要用一個特征向量表示。最后采用倒排索引方式,將相似性圖片進(jìn)行聚類。第二部分,由于已聚類好的相似性圖片其相似度很高,采用視頻流壓縮算法對圖片進(jìn)行壓縮,極大減少圖片存儲容量。K-Means聚類算法是圖片相似性聚類過程中的關(guān)鍵技術(shù),其執(zhí)行速度與結(jié)果將直接影響相似性圖片壓縮效果。換句話說,K-Means聚類算法會是整個系統(tǒng)的一個性能瓶頸。當(dāng)處理大數(shù)量特征點(diǎn)時,標(biāo)準(zhǔn)K-Means聚類算法中數(shù)據(jù)點(diǎn)n和中心點(diǎn)k值會變得相當(dāng)大,從而使得K-Means聚類算法效率變得低下。設(shè)計實(shí)現(xiàn)一種K-Means聚類算法優(yōu)化方案,使其能夠在面對大數(shù)據(jù)量n和k值情況下,降低算法時間復(fù)雜度,提升K-Means算法執(zhí)行效率,因而應(yīng)用于圖片去重系統(tǒng)中提升系統(tǒng)執(zhí)行速度。最后根據(jù)實(shí)驗(yàn)測試結(jié)果顯示,優(yōu)化的K-Means算法在大數(shù)量級下有較好的性能提升。
[Abstract]:With the rapid development and popularization of cloud storage services, multimedia data such as pictures and videos have become the main way to record and share information. Compared with traditional text records, multimedia data storage such as pictures takes up much more storage space. Therefore, how to reduce the storage capacity of images is a new challenge when dealing with multimedia data such as pictures. The study found that in mainstream social networking sites, such as the facebook,qq, Baidu cloud, similar images accounted for a large proportion of the total number of images. The similarity picture is defined as a series of pictures which are obtained by the common transformation of the picture, such as continuous shooting, watermark, clipping, zooming and so on. In response to this discovery, a system of image removal was designed. The image removal system can be divided into two parts. In the first part, the similarity clustering is carried out, and the content-based image retrieval is carried out. In the aspect of image retrieval technology, the image is preprocessed, the local feature value is extracted, the K-Means clustering algorithm is implemented for all the extracted feature values, and the final clustering center is regarded as the visual word of the BOW (Bag-of-Words) model. The SIFT feature point set is quantized with visual words, so that only one feature vector is used to represent a picture. Finally, the similarity images are clustered by inverted index. In the second part, due to the high similarity of the well clustered images, the video stream compression algorithm is used to compress the images, which greatly reduces the storage capacity of the images. K-Means clustering algorithm is the key technology in the process of image similarity clustering. Its execution speed and result will directly affect the image compression effect of similarity. In other words, K-Means clustering algorithm is a performance bottleneck for the whole system. When dealing with a large number of feature points, the data point n and the center point k in the standard K-Means clustering algorithm become quite large, which makes the K-Means clustering algorithm inefficient. An optimization scheme of K-Means clustering algorithm is designed and implemented, which can reduce the time complexity of the algorithm and improve the execution efficiency of the K-Means algorithm in the case of large amount of data n and k, so it can be applied to the image removal system to improve the execution speed of the system. Finally, the experimental results show that the optimized K-Means algorithm has better performance in large order of magnitude.
【學(xué)位授予單位】：華中科技大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2016
【分類號】：TP391.41;TP311.13
本文編號：2289022
[Abstract]:With the rapid development and popularization of cloud storage services, multimedia data such as pictures and videos have become the main way to record and share information. Compared with traditional text records, multimedia data storage such as pictures takes up much more storage space. Therefore, how to reduce the storage capacity of images is a new challenge when dealing with multimedia data such as pictures. The study found that in mainstream social networking sites, such as the facebook,qq, Baidu cloud, similar images accounted for a large proportion of the total number of images. The similarity picture is defined as a series of pictures which are obtained by the common transformation of the picture, such as continuous shooting, watermark, clipping, zooming and so on. In response to this discovery, a system of image removal was designed. The image removal system can be divided into two parts. In the first part, the similarity clustering is carried out, and the content-based image retrieval is carried out. In the aspect of image retrieval technology, the image is preprocessed, the local feature value is extracted, the K-Means clustering algorithm is implemented for all the extracted feature values, and the final clustering center is regarded as the visual word of the BOW (Bag-of-Words) model. The SIFT feature point set is quantized with visual words, so that only one feature vector is used to represent a picture. Finally, the similarity images are clustered by inverted index. In the second part, due to the high similarity of the well clustered images, the video stream compression algorithm is used to compress the images, which greatly reduces the storage capacity of the images. K-Means clustering algorithm is the key technology in the process of image similarity clustering. Its execution speed and result will directly affect the image compression effect of similarity. In other words, K-Means clustering algorithm is a performance bottleneck for the whole system. When dealing with a large number of feature points, the data point n and the center point k in the standard K-Means clustering algorithm become quite large, which makes the K-Means clustering algorithm inefficient. An optimization scheme of K-Means clustering algorithm is designed and implemented, which can reduce the time complexity of the algorithm and improve the execution efficiency of the K-Means algorithm in the case of large amount of data n and k, so it can be applied to the image removal system to improve the execution speed of the system. Finally, the experimental results show that the optimized K-Means algorithm has better performance in large order of magnitude.
【學(xué)位授予單位】：華中科技大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2016
【分類號】：TP391.41;TP311.13
【參考文獻(xiàn)】
相關(guān)博士學(xué)位論文 前2條
1 李大瑞;基于大規(guī)模視覺模式學(xué)習(xí)的高性能圖像表示[D];中國科學(xué)技術(shù)大學(xué);2014年
2 葉航軍;面向大規(guī)模圖像庫的索引和檢索機(jī)制研究[D];清華大學(xué);2003年
相關(guān)碩士學(xué)位論文 前5條
1 曹六一;海量圖像檢索系統(tǒng)設(shè)計與實(shí)現(xiàn)[D];山東大學(xué);2014年
2 王琰;JPEG2000圖像壓縮算法軟件優(yōu)化設(shè)計[D];西安電子科技大學(xué);2012年
3 李俊偉;H264幀內(nèi)預(yù)測和去塊濾波的研究和實(shí)現(xiàn)[D];電子科技大學(xué);2010年
4 馬斌;基于DSP的JPEG編碼器的實(shí)現(xiàn)與優(yōu)化[D];西安電子科技大學(xué);2009年
5 趙剛;多媒體數(shù)據(jù)庫降維與索引算法研究及系統(tǒng)實(shí)現(xiàn)[D];吉林大學(xué);2007年
，本文編號：2289022
本文鏈接：http://sikaile.net/kejilunwen/ruanjiangongchenglunwen/2289022.html
最近更新
教材專著
