增强边缘梯度二值卷积神经网络的人脸姿态识别

为了提高人脸姿态识别的识别精度,设计了一种增强边缘梯度二值卷积神经网络用于识别.首先,提出ROILBC(Region of Interest Local Binary Convolution)在人脸姿态图像上提取二值特征并归类,根据二值特征图谱和原像的对比情况选择人脸姿态图像ROI(Region of Interest)以供后续网络学习.其次,提出DR-MGPC(Dimensionality Reduced Modified Gradient Pattern Convolution)提取图像边缘梯度二值特征,在此基础上,提出Enhanced DR-LDPC(Enhanced Dimensionality Reduced Local Directional Pattern Convolution)提取图像增强边缘梯度方向特征.网络采用直方图相似度、卡方检验、常态分布比对的巴氏距离法作为测量依据来进行识别;实验在FERET和CAS-PEAL-R1数据集上进行,相比其他人脸姿态识别方法,提出的二值模式卷积神经网络在识别精度和计算效率上更优异.

The Choice of Effective Vocabulary Learning Strategies

This paper aims to make a comparison between good and poor language learners in the use of vocabulary learning strategies. It will introduce some helpful vocabulary learning strategies to help those frustrated Chinese college non-English major learners.

What Your Poo Says About Your Income?

A laboratory in the University of Queensland is storing some unusual specimens:the human waste of more than one-fifth of the Australian population.Thesamples,collectedfrom wastewater treatment plants across the country,frozen and mailed to researchers,have been described as a treasure trove of insight into the dietary and drug habits of different communities.

多囊卵巢综合征的病机初探

随着现代医学对多囊卵巢综合征(PCOS)70多年的深入研究,祖国医学对PCOS的认识有了很大进展。本文试从PCOS患者肥胖型和非肥胖型的不同激素水平方面,用中医特有的辨证思想阐述PCOS的中医病机。