本文選題：服飾檢索 + 網(wǎng)絡(luò)爬蟲(chóng)　； 參考：《吉林大學(xué)》2017年碩士論文

【摘要】：網(wǎng)上購(gòu)物因其便捷性自從出現(xiàn)以來(lái)一直備受青睞,服裝類商品是在線銷售量最高的產(chǎn)品之一,為了促進(jìn)網(wǎng)絡(luò)銷售服裝的發(fā)展,快速準(zhǔn)確的搜索出心儀的衣服,成為了關(guān)鍵。本課題構(gòu)建了一個(gè)基于語(yǔ)義信息的服飾檢索平臺(tái),將低層次的物理特征與高層次語(yǔ)義特征聯(lián)系起來(lái),將語(yǔ)義特征作為檢索依據(jù)完成圖像檢索。本課題的主要研究?jī)?nèi)容包括:1、應(yīng)用網(wǎng)絡(luò)爬蟲(chóng)技術(shù)建立初始數(shù)據(jù)集。使用網(wǎng)絡(luò)爬蟲(chóng)技術(shù)獲取各大服飾網(wǎng)站中不同種類的服飾信息,如縮略圖、價(jià)格、購(gòu)買(mǎi)量等,建立初始訓(xùn)練數(shù)據(jù)集,將圖像檢索與網(wǎng)絡(luò)爬蟲(chóng)相結(jié)合,對(duì)大規(guī)模圖像數(shù)據(jù)進(jìn)行處理,提高了在線搜索性能。2、采用語(yǔ)義分類的方法對(duì)圖像進(jìn)行分類處理。相對(duì)于低級(jí)特征,屬性對(duì)服裝變化有較好的魯棒性,從大量的不同訓(xùn)練數(shù)據(jù)中提取的不同的視覺(jué)屬性,將作為圖像的高級(jí)語(yǔ)義表示,將服飾圖像的物理層面的特征進(jìn)行提取和歸納,與高層次語(yǔ)義標(biāo)簽建立連接,應(yīng)用最小二乘概率分類算法生成語(yǔ)義模型,該模型作為圖像的分類標(biāo)準(zhǔn),為圖像的相似性度量提供支持。3、引入倒排索引技術(shù),將相似圖片定位到候選對(duì)象中,減少相似性度量計(jì)算次數(shù),有利于縮短檢索時(shí)間。基于語(yǔ)義信息的服飾檢索開(kāi)發(fā)的工作流程為,首先應(yīng)用爬蟲(chóng)技術(shù)抓取服飾網(wǎng)站數(shù)據(jù),構(gòu)建初始數(shù)據(jù)集,為檢索提供數(shù)據(jù)支持。對(duì)數(shù)據(jù)集中圖像提取局部描述算子,使用k-means算法對(duì)特征進(jìn)行歸一化處理后得到的特征向量,與顏色特征組合共同代表一幅圖像。采用分類算法將初始數(shù)據(jù)集中的所有圖像特征描述向量分類訓(xùn)練得到語(yǔ)義模型。在檢索階段,采用上傳本地圖片的方法,通常情況下,用戶所捕獲的服裝圖像有著不規(guī)范性,受到幾何變形,閉塞,雜亂的背景和光度變化等因素的影響,將對(duì)服飾檢索造成極大的挑戰(zhàn),因此對(duì)查詢圖像進(jìn)行圖像分割、特征提取等圖像預(yù)處理操作,進(jìn)而提取圖像的目標(biāo)區(qū)域的特征。將提取到的特征經(jīng)由語(yǔ)義分類模型分類,得到屬于該圖像的分類概率向量,該特征向量作為圖像的相似度度量依據(jù),歐式距離越小則圖像越相似。檢索時(shí)引入倒排索引方法,從數(shù)據(jù)集中快速而準(zhǔn)確的獲取相同或相似的服飾信息,如服飾圖片、價(jià)格、來(lái)源、訪問(wèn)量等,提高檢索速度。實(shí)驗(yàn)結(jié)果表明,本系統(tǒng)具有健壯性,并能提供優(yōu)質(zhì)的檢索結(jié)果。
[Abstract]:Online shopping has been favored since its convenience, clothing products is one of the highest online sales products, in order to promote the development of online sales of clothing, fast and accurate search for the desired clothing, has become the key. In this paper, a clothing retrieval platform based on semantic information is constructed, which connects the low-level physical features with the high-level semantic features, and uses semantic features as the basis for image retrieval. The main research contents of this thesis include: 1, using web crawler technology to establish initial data set. Using web crawler technology to obtain different kinds of clothing information, such as thumbnail, price, purchase amount, etc., establish initial training data set, combine image retrieval with web crawler, and process large scale image data. The method of semantic classification is used to classify images. Relative to low-level features, attributes are more robust to clothing changes. Different visual attributes extracted from a large number of different training data will be used as high-level semantic representation of images. The features of the physical level of dress image are extracted and induced, connected with the high-level semantic label, and the semantic model is generated by using the least square probability classification algorithm, which is used as the classification standard of the image. The inverted indexing technique is introduced to locate the similarity images in the candidate objects, which can reduce the number of similarity measurement calculations and shorten the retrieval time. The workflow of costume retrieval development based on semantic information is: firstly, crawler technology is used to capture clothing website data, and the initial data set is constructed to provide data support for retrieval. The local description operator is extracted from the image in the dataset, and the feature vector is obtained by using the k-means algorithm to normalize the feature, and the color feature combination is used to represent a single image. The classification algorithm is used to train all the feature description vectors in the initial data set to obtain the semantic model. In the retrieval stage, the method of uploading local images is used. In general, the clothing images captured by users are irregular, affected by geometric deformation, block, clutter background and photometric changes, etc. It will pose a great challenge to dress retrieval, so image preprocessing operations such as image segmentation and feature extraction are carried out to extract the features of the target region of the image. The extracted features are classified by semantic classification model to obtain the classification probability vector of the image. The feature vector is used as the measure of image similarity. The smaller the Euclidean distance is, the more similar the image is. The inverted index method is introduced to obtain the same or similar dress information quickly and accurately from the data set, such as dress picture, price, source, visit quantity and so on, so as to improve the retrieval speed. The experimental results show that the system is robust and can provide high quality retrieval results.
【學(xué)位授予單位】：吉林大學(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2條

1 韓昌剛;郭玉堂;;基于CCA的圖像語(yǔ)義特征提取的分析與研究[J];計(jì)算機(jī)應(yīng)用研究;2012年05期

2 張秋霞;谷平;;文檔主題提取算法VSM改進(jìn)與應(yīng)用[J];情報(bào)雜志;2007年12期

相關(guān)博士學(xué)位論文 前1條

1 劉偉;圖像檢索中若干問(wèn)題的研究[D];浙江大學(xué);2007年

，

本文編號(hào)：2094384