结合多尺度HOG特征和语义属性的合成素描人脸识别

合成素描的人脸识别问题属于异质人脸识别研究领域,在刑侦领域具有重要的实际应用.由于合成素描与人脸照片属于不同模态,对不同模态人脸进行鲁棒的表征是识别的关键.针对合成素描人脸在某些区域缺乏纹理细节,单纯依赖局部细节特征识别率较低的问题,文中提出一种融合多尺度HOG特征并加以语义属性约束的合成素描人脸识别的算法.首先提取出合成素描人脸的全局HOG特征以及五官等关键部位的局部HOG特征来表征人脸的整体结构特征和细节特征,之后将得到的整体结构特征和各个部位的细节特征进行分数层融合,最后用语义属性特征对匹配结果进行重排序.在PRIP-VSGC和UoM-SGFS数据集上进行验证,文中算法rank10的识别率分别达到88.6%和96.7%,与现有算法相比有明显的提高.     

信息熵加权的HOG特征提取算法研究

针对人脸图像中不同部位所含的信息熵不同,对识别的影响程度不同等因素,提出了一种信息熵加权的HOG特征提取方法。该算法将待识别的人脸图像进行分块,对分块后的图像进行HOG特征提取,计算每块图像所含的信息熵作为权重系数加到各个分块中形成新的HOG特征,通过PCA算法对特征进行降维,得到信息熵加权的HOG特征。通过在ORL和YALE实验结果表明,该算法相较于其他传统识别方法具有更高的识别精度和准确度,并且对于人脸在光照、姿态表情等干扰因素下均具有良好的有效性和鲁棒性。     

面向人脸识别的WPD-HOG金字塔特征提取方法

人脸识别技术可应用于各监控和安保领域,它涉及特征提取、识别模型等关键技术。其中特征提取方法直接影响识别效果,目前所用的特征提取方法存在特征表达不全面、计算复杂度高等问题。据此,提出一种基于WPD-HOG金字塔的人脸特征提取方法,该方法结合小波包分解（Wavelet Packet Decomposition,WPD）、图像金字塔以及方向梯度直方图（Histograms of Oriented Gradients,HOG）对人脸图像特征进行有效表征,最终将WPD-HOG金字塔特征通过SVM分类器进行分类。通过在ORL人脸库上进行实验,与四种对比方法HOG、HOG金字塔、FWPD-HOG以及FWPD-HOG金字塔进行比较,实验结果表明,WPD-HOG金字塔特征提取方法的识别率要高于对比方法,且在噪声方面具有较好的鲁棒性。     

改进的HOG和Gabor,LBP性能比较

为了实现复杂环境下的人脸特征有效表达,提出一种改进的梯度方向直方图(HOG)人脸识别方法.首先以人脸图像网格作为采样窗口并在其上提取HOG特征;然后将所有网格HOG特征向量进行组合,实现整个人脸特征表达;最后采用最近邻分类器进行识别.另外,比较了该方法与Gabor小波和局部二值模式(LBP)2种著名的人脸局部特征表示方法的优劣.实验结果表明,在调优的HOG参数下,在具有光照和时间环境等复杂变化的FERET人脸库中,较少维数的HOG特征比LBP特征有更好的表现,而且HOG特征提取时间和特征向量维数比Gabor小波方法更具有优势.     

一种自适应加权HOG特征的人脸识别算法

为了提高人脸识别在复杂条件下的识别率,提出一种基于自适应加权梯度方向直方图特征（AW-HOG）的人脸识别方法。该方法首先将人脸图像分成均匀子块,并利用HOG描述算子提取分块人脸特征,根据各分块对识别的贡献率自适应地计算各分块的权重,然后融合权重系数以及各分块的HOG特征,形成AW-HOG特征并采用主成分分析（PCA）算法进行降维,最后利用支持向量机（SVM）进行分类识别。在Yale B 以及AR标准人脸库上的实验结果表明,提出的人脸识别方法在识别率上优于传统算法且对光照具有较强的鲁棒性。     

基于Haar特性的改进HOG的人脸特征提取算法

现有的大多数特征提取算法在提取人脸特征时,容易受到光照等外界因素的影响,从而导致后期人脸识别率下降。而方向梯度直方图(Histogram of Oriented Gradient,HOG)具有较强的光照鲁棒性,能够很好地减少由光照带来的干扰,但传统HOG在计算梯度幅值和方向时只计算水平和垂直方向上4个像素点对中间像素的影响,当外界环境变化时不能保证稳定性,因此提出一种基于Haar特性的改进HOG的人脸特征提取算法。该算法在计算梯度幅值和方向时考虑水平、垂直以及对角线上8个像素点对中间像素的影响,由于增加计算量导致特征提取时间也随之增加,因此引入Haar,借助Haar型特征运算简单、快捷的特点设计4组Haar型特征编码模式,按照改进的HOG特征计算方式提取人脸特征。在有光照等外界因素影响的FERET人脸数据库和Yale B扩展的人脸测试库中进行实验,实验结果表明,与GFC,LBP和其他文献中的HOG算法相比,该算法对光照具有更好的鲁棒性,能够在光照变化的环境下提高人脸识别率。该算法在FERET探测集fb,fc,dup1和dup2上的识别率分别为95.1%,80.9%,70.1%和63.2%,在Yale B中的识别率为89.1%。     

基于多级纹理特征的深度信念网络人脸识别算法

为解决人脸特征提取过程中局部特征缺失的问题,借助局部二值模式(LBP)与方向梯度直方图(HOG)提出一种基于多级纹理特征融合的深度信念网络人脸识别算法。以提取局部纹理特征以及边缘纹理特征为出发点,对人脸图像进行三级纹理特征提取。使用MB-LBP提取初级纹理特征;在此基础上进行改进的CS-LBP图像特征提取作为二级纹理特征;使用HOG算子在二级纹理特征上完成三级纹理特征提取。将二级和三级纹理特征直方图顺序串联融合后输入到深度信念网络(DBN)逐层贪婪训练,优化网络参数,并用优化的网络在ORL、YELA人脸标准库中进行测试,识别率均在92%以上。该算法与传统算法(SVM、PCA)相比较拥有更好的人脸识别效果,同时也表明了局部纹理特征的改善为识别过程的特征提取提供强有力的保障,为人脸识别的进一步研究开拓新思路。     

Recognizing expressions from face and body gesture by temporal normalized motion and appearance features

Recently, recognizing affects from both face and body gestures attracts more attentions. However, it still lacks of efficient and effective features to describe the dynamics of face and gestures for real-time automatic affect recognition. In this paper, we combine both local motion and appearance feature in a novel framework to model the temporal dynamics of face and body gesture. The proposed framework employs MHI-HOG and Image-HOG features through temporal normalization or bag of words to capture motion and appearance information. The MHI-HOG stands for Histogram of Oriented Gradients (HOG) on the Motion History Image (MHI). It captures motion direction and speed of a region of interest as an expression evolves over the time. The Image-HOG captures the appearance information of the corresponding region of interest. The temporal normalization method explicitly solves the time resolution issue in the video-based affect recognition. To implicitly model local temporal dynamics of an expression, we further propose a bag of words (BOW) based representation for both MHI-HOG and Image-HOG features. Experimental results demonstrate promising performance as compared with the state-of-the-art. Significant improvement of recognition accuracy is achieved as compared with the frame-based approach that does not consider the underlying temporal dynamics.     

Pattern recognition using feature feedback: Application to face recognition

In this paper, we propose a new pattern recognition method using feature feedback and present its application to face recognition. Conventional pattern recognition methods extract the features employed for classification using PCA, LDA and so on. On the other hand, in the proposed method, the extracted features are analyzed in the original space using feature feedback. Using reverse mapping from the extracted features to the original space, we can identify the important part of the original data that affects the classification. In this way, we can modify the data to obtain a higher classification rate, make it more compact or abbreviate the required sensors. To verify the applicability of the proposed method, we apply it to face recognition using the Yale Face Database. Each face image is divided into two parts, the important part and unimportant part, using feature feedback, and the classification performed using the feature mask obtained from feature feedback. Also, we combine face recognition with image compression. The experimental results show that the proposed method works well.     

Efficient railway tracks detection and turnouts recognition method using HOG features

Railway tracks detection and turnouts recognition are the basic tasks in driver assistance systems, which can determine the interesting regions for detecting obstacles and signals. In this paper, a novel railway tracks detection and turnouts recognition method using HOG (Histogram of Oriented Gradients) features was presented. At first, the approach computes HOG features and establishes integral images, and then extracts railway tracks by region-growing algorithm. Then based on recognizing the open direction of the turnout, we find the path where the train will travel through. Experiments demonstrated that our method was able to correctly extract tracks and recognize turnouts even in very bad illumination conditions and run fast enough for practical use. In addition, our approach only needs a computer and a cheap camera installed in the railroad vehicle, not specialized hardwares and equipment.     

HOG在人脸识别中的性能研究

介绍梯度方向直方图(HOG)人脸识别算法,设计基于脸部识别技术人脸库的HOG人脸识别实验,以测试不同HOG参数对人脸识别的影响,从而进行最优参数设置。实验结果表明,HOG特征在行人检测和人脸识别上对梯度方向空间和区间的选择是一致的,不同分块模式对人脸识别的影响与行人检测不同,HOG特征描述子用较少的特征维数就能有效地表达人脸,采用块内标准化方式后,识别性能有大幅度提升。     

基于脊谷特征提取的三维人脸识别

针对三维人脸数据庞大及识别效率低的问题,提出采用提取脊点及谷点表征人脸。脊点和谷点作为曲面局部区域内主曲率沿主方向变化的极值点,能够很好地表征三维人脸特征。对三维人脸提取脊点模型和谷点模型,通过对它们栅格化后生成对应的空间分布密度直方图实现人脸粗匹配,采用计算LTS-Hausdorff距离实现人脸的精确匹配。在GavabDB三维人脸库的实验结果表明,该方法具有较高的识别率。