本文關(guān)鍵詞：基于P-RBF神經(jīng)網(wǎng)絡(luò)的人臉識(shí)別算法研究　出處：《南昌航空大學(xué)》2017年碩士論文　論文類(lèi)型：學(xué)位論文

  更多相關(guān)文章： PCA主成分分析 LDA線(xiàn)性判別分析 P-RBF神經(jīng)網(wǎng)絡(luò) FCM模糊C均值 DE差分進(jìn)化

【摘要】：數(shù)據(jù)預(yù)處理時(shí),主成分分析算法(PCA算法)能夠降低特征空間的維數(shù)。但是,由于涉及整體面部圖像,使得在改變視點(diǎn)的情況下不能保證具有相同的識(shí)別率,故為了彌補(bǔ)局限性,在PCA算法的基礎(chǔ)上,提出了線(xiàn)性判別分析算法(LDA算法)來(lái)提高處理不同類(lèi)別圖像時(shí)的識(shí)別率。本文首先闡述了由PCA與LDA相結(jié)合的新型算法,接著詳細(xì)介紹了P-RBF神經(jīng)網(wǎng)絡(luò)的設(shè)計(jì)方法和具體實(shí)現(xiàn)過(guò)程,最后,在A(yíng)TT數(shù)據(jù)庫(kù)和耶魯數(shù)據(jù)庫(kù)中進(jìn)行人臉識(shí)別試驗(yàn),從而為P-RBF神經(jīng)網(wǎng)絡(luò)的人臉識(shí)別系統(tǒng)設(shè)計(jì)了一個(gè)最優(yōu)的人臉識(shí)別方案。在本文中,提出了基于多項(xiàng)式的徑向基函數(shù)神經(jīng)網(wǎng)絡(luò)(P-RBF NNS)作為主要識(shí)別部分的人臉識(shí)別系統(tǒng)。該系統(tǒng)有助于解決高維圖像識(shí)別問(wèn)題,其主要由圖像數(shù)據(jù)預(yù)處理部分和圖像數(shù)據(jù)識(shí)別部分構(gòu)成。提出的P-RBF神經(jīng)網(wǎng)絡(luò)體系結(jié)構(gòu)分為三個(gè)功能模塊:條件部分、結(jié)論部分、和聚集部分。在P-RBF神經(jīng)網(wǎng)絡(luò)的條件部分,輸入空間通過(guò)使用模糊C均值(FCM)算法來(lái)實(shí)現(xiàn)模糊聚類(lèi)的分配。在P-RBF神經(jīng)網(wǎng)絡(luò)的結(jié)論部分中,使用如下三種多項(xiàng)式,如常數(shù)型、線(xiàn)性型和二次多項(xiàng)式型來(lái)作為連接函數(shù)。在P-RBF神經(jīng)網(wǎng)絡(luò)的聚集部分,通過(guò)采用模糊推理法獲得P-RBF神經(jīng)網(wǎng)絡(luò)模型的系數(shù)。同時(shí),將“如果-那么”規(guī)則作為該神經(jīng)網(wǎng)絡(luò)聚集部分的模糊規(guī)則集合。該神經(jīng)網(wǎng)絡(luò)的基本設(shè)計(jì)參數(shù)(包括學(xué)習(xí)速率,動(dòng)量系數(shù),模糊化系數(shù)和特征選擇項(xiàng))由差分進(jìn)化(DE)算法進(jìn)行優(yōu)化。最后,在A(yíng)TT數(shù)據(jù)庫(kù)和耶魯數(shù)據(jù)庫(kù)進(jìn)行人臉識(shí)別試驗(yàn),實(shí)驗(yàn)結(jié)果表明,PCA-LDA結(jié)合算法具有更好的可行性和有效性,能有效實(shí)時(shí)的給出測(cè)試者的人臉識(shí)別結(jié)果。
[Abstract]:Data preprocessing, principal component analysis algorithm (PCA algorithm) can reduce the dimension of feature space. However, due to the whole face image, the change of viewpoint in case of guarantee has the same recognition rate, so in order to make up for the limitations, based on the PCA algorithm, presents a linear discriminant analysis algorithm (LDA to improve the recognition algorithm) when the rate of different categories of images. This paper describes the new algorithm by the combination of PCA and LDA, then introduces the design method of P-RBF neural network and realization process, finally, face recognition test in the ATT database and Yale database, so as to design a scheme for face recognition the optimal P-RBF neural network for face recognition system. In this paper, radial basis function neural network is proposed based on polynomial (P-RBF NNS) as the main part of the recognition of face recognition system System. The system is helpful to solve the high-dimensional image recognition problem, which is mainly composed of image data preprocessing and image data to identify parts. P-RBF neural network architecture proposed is divided into three functional modules: part, conclusion, and aggregation. In P-RBF neural network conditions, the input space by the use of fuzzy C means (FCM) algorithm to realize the distribution of fuzzy clustering. The P-RBF neural network in the conclusion part, use the following three kinds of polynomials, as usual number type, linear type and two polynomial as the connection function in the aggregation part of P-RBF neural network by using fuzzy inference method, coefficient of P-RBF neural network model. At the same time, the "If then" rules as the neural network fuzzy aggregation rules part of the collection. The basic design parameters of the neural network (including learning rate, momentum coefficient, fuzzy Feature selection coefficient) by differential evolution (DE) algorithm was optimized. Finally, face recognition test in the ATT database and Yale database, the experimental results show that the PCA-LDA algorithm has better feasibility and effectiveness of the face recognition results can effectively give the real-time testing.

【學(xué)位授予單位】：南昌航空大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類(lèi)號(hào)】：TP391.41;TP183

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 周慧;張尤賽;龔淼;;基于RBF神經(jīng)網(wǎng)絡(luò)的醫(yī)學(xué)圖像分類(lèi)算法研究[J];電子設(shè)計(jì)工程;2017年03期

2 張?jiān)苿?王志春;;基于RBF神經(jīng)網(wǎng)絡(luò)的磨礦粒度軟測(cè)量模型及實(shí)現(xiàn)[J];計(jì)算機(jī)測(cè)量與控制;2017年01期

3 宋永強(qiáng);劉議聰;朱泓光;;基于RBF神經(jīng)網(wǎng)絡(luò)和單神經(jīng)元PID的PMSM速度控制器[J];電子世界;2017年02期

4 邵婷婷;亓宏濤;肖劍;秦俊;王可可;;基于神經(jīng)網(wǎng)絡(luò)的井眼姿態(tài)測(cè)量誤差補(bǔ)償研究[J];微型電腦應(yīng)用;2017年01期

5 嚴(yán)s，

本文編號(hào)：1391063