查询结果:   丁欢欢,杨永红.基于加权DWT和DCT的粒子群神经网络人脸识别算法[J].计算机应用与软件,2016,33(1):156 - 158,229.
中文标题
基于加权DWT和DCT的粒子群神经网络人脸识别算法
发表栏目
人工智能与识别
摘要点击数
923
英文标题
FACE RECOGNITION ALGORITHM OF PARTICLE SWARM OPTIMISATION NEURAL NETWORK BASED ON WEIGHTED DWT AND DCT
作 者
丁欢欢 杨永红 Ding Huanhuan Yang Yonghong
作者单位
江苏科技大学电子与信息工程学院 江苏 镇江 212003     
英文单位
School of Electronic and Information Engineering,Jiangsu University of Science and Technology,Zhenjiang 212003,Jiangsu,China     
关键词
人脸识别 小波变换 离散余弦变换 粒子群优化算法 BP神经网络
Keywords
Face recognition Wavelet transform Discrete cosine transform Particle swarm optimisation BP neural network
基金项目
江苏省高校优秀中青年教师和校长境外研修计划
作者资料
丁欢欢,硕士生,主研领域:图像图形及智能控制。杨永红,副教授。 。
文章摘要
针对人脸识别中出现的维数过高和计算复杂而导致的识别率低的问题,提出一种基于加权DWT(Discrete Wavelet Transform)和DCT(Discrete Cosine Transform)的粒子群神经网络人脸识别新算法。该算法首先用小波变换对人脸图像进行分解,去除对角线分量影响,提取加权低频和高频的离散余弦变换系数作为特征向量,最后利用粒子群优化BP神经网络进行分类识别。在ORL人脸库上验证了该算法的有效性和可行性。
Abstract
Concerning the low recognition rate problem caused by too high dimensions and complex computation in face recognition, in this paper we proposed a new face recognition algorithm of particle swarm optimisation neural network, which is based on weighted discrete wavelet transform and discrete cosine transform. The algorithm first utilises wavelet transform to decompose face image, eliminates the effect of diagonal components, and extracts the weighted low-frequency and high-frequency discrete cosine transform coefficients as the feature vectors. Finally it uses particle swarm optimisation BP neural network for classification and recognition. The effectiveness and feasibility of the algorithm have been verified on ORL face database.
下载PDF全文