 |
 |
 |
 |
 |
 |
||
 |
 |
 |
 |
 |
| 4 | 1/1 | 返回列表 |
| 查看: 1047 | 回復(fù): 3 | |||
likaiaiswt新蟲 (小有名氣)
|
[求助]
人臉識(shí)別翻譯！
|
|
PCA: Principal Components Analysis (PCA) PCA, commonly referred to as the use of eigenfaces, is the technique pioneered by Kirby and Sirivich in 1988. With PCA, the probe and gallery images must be the same size and must first be normalized to line up the eyes and mouth of the subjects within the images. The PCA approach is then used to reduce the dimension of the data by means of data compression basics2 and reveals the most effective low dimensional structure of facial patterns. This reduction in dimensions removes information that is not useful4 and precisely decomposes the face structure in toorthogonal (uncorrelated) components known as eigenfaces. Each face image may be represented as a weighted sum (feature vector) of the eigenfaces, which are stored in a 1D array. A probe image is compared against a gallery image by measuring the distance between their respective feature vectors. The PCA approach typically requires the full frontal face to be presented each time; otherwise the image results in poor performance.4 The primary advantage of this technique is that it can reduce the data needed to identify the individual to 1/1000th of the data presented. 5This Document Last Updated: 7 August 2006 Page 2 of 10 Face Recognition Figure 1: Standard Eigenfaces: Feature vectors are derived using eigenfaces.6 LDA: Linear Discriminant Analysis LDA is a statistical approach for classifying samples of unknown classes based on training samples with known classes.4 (Figure 2)This technique aims to maximize between-class (i.e., across users) variance and minimize within-class (i.e., within user)variance. In Figure 2 where each block represents a class, there are large variances between classes, but little variance within classes. When dealing with high dimensional face data, this technique faces the small sample size problem that arises where there are a small number of available training samples compared to the dimensionality of the sample space.7 Figure 2: Example of Six Classes Using LDA 8This Document Last Updated: 7 August 2006 Page 3 of 10Face Recognition EBGM: Elastic Bunch Graph Matching EBGM relies on the concept that real face images have many nonlinearcharacteristics that are not addressed by the linear analysismethods discussed earlier, such as variations in illumination(outdoor lighting vs. indoor fluorescents), pose (standing straightvs. leaning over) and expression (smile vs. frown). A Gaborwavelet transform creates a dynamic link architecture thatprojects the face onto an elastic grid.4 The Gabor jet is a node onthe elastic grid, notated by circles on the image below, whichdescribes the image behavior around a given pixel. It is the resultof a convolution of the image with a Gabor filter, which is used todetect shapes and to extract features using image processing. [Aconvolution expresses the amount of overlap from functions,blending the functions together.] Recognition is based on thesimilarity of the Gabor filter response at each Gabor node.4 Thisbiologically-based method using Gabor filters is a processexecuted in the visual cortex of higher mammals. The difficultywith this method is the requirement of accurate landmarklocalization, which can sometimes be achieved by combining PCAand LDA methods.4 Figure 4: Elastic Bunch Map Graphing.9 |
回復(fù)此樓» 猜你喜歡
南京大學(xué)化學(xué)調(diào)劑
已經(jīng)有11人回復(fù)
-
材料專碩調(diào)劑
已經(jīng)有5人回復(fù)
-
一志愿華東師范大學(xué)有機(jī)化學(xué)專業(yè)，初試351分，復(fù)試被刷求調(diào)劑!
已經(jīng)有5人回復(fù)
-
283求調(diào)劑（080500）
已經(jīng)有7人回復(fù)
-
311求調(diào)劑
已經(jīng)有6人回復(fù)
-
275求調(diào)劑
已經(jīng)有20人回復(fù)
-
各位老師好，我的一志愿為北京科技大學(xué)085601材料專碩
已經(jīng)有9人回復(fù)
-
294分080500材料科學(xué)與工程求調(diào)劑
已經(jīng)有8人回復(fù)
-
一志愿北京工業(yè)大學(xué)，324分求調(diào)劑
已經(jīng)有5人回復(fù)
-
一志愿武漢理工，總分321，英一數(shù)二，求老師收留。
已經(jīng)有11人回復(fù)
sklyer_123
銀蟲 (小有名氣)
- 翻譯EPI: 2
- 應(yīng)助: 4 (幼兒園)
- 金幣: 277.4
- 紅花: 1
- 帖子: 241
- 在線: 12.5小時(shí)
- 蟲號(hào): 2198228
- 注冊(cè): 2012-12-20
- 性別: MM
- 專業(yè): 會(huì)計(jì)與審計(jì)
【答案】應(yīng)助回帖
愛與雨下: 請(qǐng)解釋一下 2012-12-26 21:46:02
|
主成分分析：主成分分析（常設(shè)仲裁法院） 主成分分析，通常稱為使用的特征，是首創(chuàng)的技術(shù)和sirivich柯比在1988。主成分分析，探頭和畫廊的圖片必須是相同的大小，必須首先正常化線的眼睛和嘴的科目內(nèi)的圖像。主成分分析法是用來減少維度的數(shù)據(jù)通過數(shù)據(jù)壓縮basics2揭示最有效的低維結(jié)構(gòu)的面部形態(tài)。這減少了尺寸，消除信息不useful4和精確分解結(jié)構(gòu)在toorthogonal（無關(guān)）部件被稱為特征。每個(gè)人臉圖像可以表示為一個(gè)加權(quán)（特征向量）的特征，這是存儲(chǔ)在一個(gè)一維數(shù)組。探針圖像比對(duì)一個(gè)畫廊圖像之間的距離測(cè)量它們各自的特征向量。主成分分析法通常需要全臉是每次；否則圖像導(dǎo)致性能較差。4主要利用這一技術(shù)，它可以減少所需的數(shù)據(jù)確定個(gè)人1 /第一千的數(shù)據(jù)。 這文件最后更新：7威嚴(yán)的2006頁2 10 圖1：標(biāo)準(zhǔn)特征臉的人臉識(shí)別：特征向量推導(dǎo)使用6的特征。 分析：線性判別分析 是一種統(tǒng)計(jì)方法進(jìn)行分類未知樣品類別根據(jù)訓(xùn)練樣本與已知類別。4（圖2）這種技術(shù)的目的是最大限度地類（即，在用戶）方差最小化類內(nèi)（即，在用戶）方差。在圖2中，每一塊代表一個(gè)階級(jí)，有很大的差異，階級(jí)之間，但很少差額內(nèi)部類。在處理高維數(shù)據(jù)，這種技術(shù)所面臨的小樣本問題出現(xiàn)在有少量訓(xùn)練樣本的比較維度的樣本空間。7 圖2：例如六類激光8文件最后更新：7威嚴(yán)的2006頁3 10face識(shí)別 彈性圖像匹配算法：彈性束圖匹配 彈性圖像匹配算法依賴的概念，真正的人臉圖像有許多nonlinearcharacteristics，沒有處理的線性分析方法討論，如光照的變化（戶外照明和室內(nèi)熒光燈），姿勢(shì)（站straightvs。俯身）和表達(dá)（笑比皺眉）。一個(gè)gaborwavelet變換創(chuàng)建動(dòng)態(tài)鏈接結(jié)構(gòu)thatprojects面臨到一個(gè)網(wǎng)格。4該射流是一個(gè)節(jié)點(diǎn)在彈性網(wǎng)格，標(biāo)記圓圈下面的圖像，whichdescribes圖像的行為在一個(gè)給定的像素。這后者是卷積的圖像提供了一個(gè)濾波器，這是用來檢測(cè)的形狀和特征提取的圖像處理。[ aconvolution表示數(shù)量的重疊的功能，融合功能。]識(shí)別是基于thesimilarity的伽柏濾波器響應(yīng)每個(gè)伽柏節(jié)點(diǎn)。4 thisbiologically-based方法使用濾波器是一個(gè)processexecuted在視覺皮層較高哺乳動(dòng)物。該difficultywith是這種方法的要求，準(zhǔn)確landmarklocalization，有時(shí)可通過結(jié)合4 pcaand測(cè)速方法。 圖4：彈性束圖9圖表。  |
2樓2012-12-25 16:44:44
likaiaiswt
新蟲 (小有名氣)
- 應(yīng)助: 0 (幼兒園)
- 金幣: 495
- 帖子: 271
- 在線: 137.5小時(shí)
- 蟲號(hào): 2104068
- 注冊(cè): 2012-11-03
- 性別: GG
- 專業(yè): 歷史社會(huì)學(xué)
3樓2012-12-25 22:03:26
fjtony163
版主 (文壇精英)
米米
- 翻譯EPI: 386
- 應(yīng)助: 361 (碩士)
- 貴賓: 6.876
- 金幣: 70610.6
- 散金: 36050
- 紅花: 222
- 沙發(fā): 616
- 帖子: 21912
- 在線: 3793.7小時(shí)
- 蟲號(hào): 1397876
- 注冊(cè): 2011-09-11
- 專業(yè): 遺傳學(xué)研究新技術(shù)與方法
- 管轄: 論文翻譯
4樓2012-12-26 19:01:36
相關(guān)版塊跳轉(zhuǎn)
我要訂閱樓主 likaiaiswt 的主題更新
| 4 | 1/1 | 返回列表 |
| 最具人氣熱帖推薦 [查看全部] | 作者 | 回/看 | 最后發(fā)表 | |
|---|---|---|---|---|
|
[考研] 311求調(diào)劑 +6 | 冬十三 2026-03-24 | 6/300 |
|
|---|---|---|---|---|
|
|
[考研] 291求調(diào)劑 +7 | Y-cap 2026-03-29 | 8/400 |
|
|
[考研] 330分求調(diào)劑 +5 | qzenlc 2026-03-29 | 5/250 |
|
|
|
[碩博家園] 求調(diào)劑 有機(jī)化學(xué)考研356分 +3 | Nadiums 2026-03-25 | 4/200 |
|
|
|
[考研] 一志愿華理，數(shù)一英一285求A區(qū)調(diào)劑 +8 | AZMK 2026-03-25 | 12/600 |
|
|
|
[考研] 317求調(diào)劑 +6 | 十閑wx 2026-03-24 | 6/300 |
|
|
|
[考研] 286求調(diào)劑 +4 | 丟掉懶惰 2026-03-27 | 7/350 |
|
|
|
[考研] 265求調(diào)劑11408 +3 | 劉小鹿lu 2026-03-27 | 3/150 |
|
|
|
[考研] 一志愿211院校 344分 東北農(nóng)業(yè)大學(xué)生物學(xué)學(xué)碩，求調(diào)劑 +5 | 丶風(fēng)雪夜歸人丶 2026-03-26 | 8/400 |
|
|
|
[考研] 考研調(diào)劑 +4 | Sanmu-124 2026-03-26 | 4/200 |
|
|
|
[考研] 308求調(diào)劑 +7 | 墨墨漠 2026-03-25 | 7/350 |
|
|
|
[考研] 08開頭275求調(diào)劑 +4 | 拉誰不重要 2026-03-26 | 4/200 |
|
|
|
[論文投稿] Journal of Mechanical Science and Technology +3 | Russ_ss 2026-03-25 | 5/250 |
|
|
|
[考研] 284求調(diào)劑 +11 | junqihahaha 2026-03-26 | 12/600 |
|
|
|
[考研] 325求調(diào)劑 +3 | Aoyijiang 2026-03-23 | 3/150 |
|
|
|
[考研] 303求調(diào)劑 +6 | 藍(lán)山月 2026-03-25 | 6/300 |
|
|
|
[考研] 332求調(diào)劑 +6 | 032500 2026-03-25 | 6/300 |
|
|
|
[考研] 求b區(qū)院校調(diào)劑 +4 | 周56 2026-03-24 | 5/250 |
|
|
|
[考研] 各位老師您好：本人初試372分 +5 | jj涌77 2026-03-25 | 6/300 |
|
|
|
[考研] 調(diào)劑 +4 | 13853210211 2026-03-24 | 4/200 |
|