基于圖像處理的自動試衣軟件設(shè)計與實現(xiàn)
[Abstract]:With the development of science and technology, especially the rapid rise of Internet technology in recent years, there have been a number of virtual fitting software, one of which is based on three-dimensional scene, good effect, but high cost and limited choice of clothing; The other kind of software is based on mobile phone, in the fitting process only changes the head, does not achieve the effect. This paper designs a fitting software combined with the related technology of digital image processing. The basic idea of the fitting software designed in this paper is to segment the model clothing region image and match it to the clothing area of the clothing person image. Therefore, the emphasis of this paper is to introduce the application and implementation of image segmentation and image matching algorithms in garment fitting system. Image segmentation mainly includes human body region segmentation and clothing region segmentation. This paper mainly integrates edge detection segmentation, threshold segmentation, region growth segmentation, gray projection segmentation, aiming at the characteristics of clothing region. In this paper, an improved algorithm for region growing segmentation is proposed to make clothing region segmentation more accurate. Image matching mainly includes size matching, position matching and skin color matching. The emphasis of the size matching of clothing region is to change the proportion in the geometric transformation and adjust the size of the model area according to the size of the clothing area of the fitter; the position matching of the clothing area mainly aligns the clothing center axis with the face center axis. The face recognition method is used to determine the center axis of the face and the color matching is used to unify the skin color between the skin color and the color of the fitting person in the model clothing area by the transformation of the color model. The fitting system has the characteristics of high automation, simple interaction, wide range of application, and also has certain market value.
【學(xué)位授予單位】:東華理工大學(xué)
【學(xué)位級別】:碩士
【學(xué)位授予年份】:2017
【分類號】:TP391.41
【參考文獻】
相關(guān)期刊論文 前10條
1 亢潔;李靜;;基于積分投影和模板匹配的人眼定位算法研究[J];陜西科技大學(xué)學(xué)報(自然科學(xué)版);2017年01期
2 劉杰;程云章;譚啟璐;韓承航;;基于區(qū)域生長法的冠狀動脈三維分割[J];生物醫(yī)學(xué)工程研究;2016年04期
3 周曉寧;郝建敏;陳悅;;一種基于灰度直方圖的閾值分割算法的研究[J];數(shù)字技術(shù)與應(yīng)用;2016年10期
4 孔祥陽;徐保根;;均值濾波及其改進算法去噪效果研究[J];商丘師范學(xué)院學(xué)報;2016年06期
5 權(quán)亞楠;卜麗靜;武文波;;改進的閾值加權(quán)平均HSV與小波變換圖像融合[J];遼寧工程技術(shù)大學(xué)學(xué)報(自然科學(xué)版);2016年01期
6 韓冬越;覃曉;劉步實;;基于分水嶺區(qū)域合并的彩色服裝圖像分割[J];廣西師范學(xué)院學(xué)報(自然科學(xué)版);2015年04期
7 蘭瑜潔;鐘舜聰;;基于區(qū)域生長法的自適應(yīng)圖像分割的網(wǎng)眼織物瑕疵檢測[J];機電工程;2015年11期
8 莫林利;趙秀紹;鄭偉;王旭;;二維表查找和雙線性插值算法的設(shè)計與應(yīng)用[J];華東交通大學(xué)學(xué)報;2015年05期
9 任曉娜;;基于Otsu算法的圖像閥值自動選取算法研究與應(yīng)用[J];電子設(shè)計工程;2015年11期
10 藺宏偉;;幾何迭代法及其應(yīng)用綜述[J];計算機輔助設(shè)計與圖形學(xué)學(xué)報;2015年04期
相關(guān)碩士學(xué)位論文 前5條
1 吳傳文;人臉識別技術(shù)在招生考試系統(tǒng)中的研究與應(yīng)用[D];北京交通大學(xué);2015年
2 劉珊珊;基于JPEG2000的感興趣區(qū)域編碼的研究與實現(xiàn)[D];電子科技大學(xué);2014年
3 李帥;圖像檢索中分割與感興趣區(qū)域提取技術(shù)的研究[D];北京郵電大學(xué);2014年
4 楊阿麗;肺部CT圖像感興趣區(qū)域提取方法研究[D];北京理工大學(xué);2011年
5 段珊珊;基于PC攝像頭的考試監(jiān)控系統(tǒng)研究與設(shè)計[D];北京交通大學(xué);2007年
,本文編號:2131831
本文鏈接:http://sikaile.net/kejilunwen/ruanjiangongchenglunwen/2131831.html