正面人脸图像中眼睛的定位算法

简单有效的投影函数广泛用于眼睛定位中,然而成像产生的反光亮点降低了该方法的准确率,为此提出一种正面人脸图像中眼睛定位的算法.考虑到反光亮点对成像的影响,首先定义了3种眼睛成像模型,分析了每个模型的投影曲线特点;然后针对每个模型给出了相应的定位方法;最后综合上述3种模型提出了统一的定位算法.利用标准模型的投影曲线可近似为对称抛物线型的特点来降低亮点对眼睛定位精度的影响.实验结果表明,利用文中算法能够准确地定位眼睛的位置,准确率大于93%.     

人脸正面图像中眼睛的精确定位

提出了人脸图像中眼睛定位方法,首先对灰度图像进行灰度形态学操作,所得谷图中像素数总体上随灰度级的增加而显著减少;用位于高端的灰度值作为阈值,逐次对谷图进行二值化,从所得二值图像中获得眼睛的候选位置;用PCA方法进行眼睛位置的最佳选择．该方法解决了二值化阈值步长的自适应选取问题．用AR,Yale和自建人脸数据库进行实验,得到眼睛定位成功率大于93％,平均定位偏差小于眼球半径的1／3．     

正面人脸图像方位的自动校正

人脸识别中图像方位的变化对识别结果有很大的影响。本文在研究眼睛的精确定位的基础上,提出一种图像方位的自动校正的方法,采用本方法对40多幅人脸图像进行了实验,结果验证了方法的有效性。     

一种改进的对称变换应用于人脸图像眼睛定位

在研究人类眼部梯度信息的基础上,提出了一种改进的对称变换算法,并通过小波变换进行目标区域选择,去除了人脸图像大面积的灰度均匀区域,减少了计算量。试验结果表明该方法不仅提高了眼睛定位的准确性,而且缩短了定位时间。     

一种改进的人脸图像眼睛快速定位方法

基于广义对称变换的原理，对传统的眼睛检测算法进行了改进。首先，利用面部眉眼区域的灰度呈纵向明显变化的特点，应用积分图法快速地初步定位眉眼区域，在这些候选的区域内用对称变换法检测候选的眼睛点，最后根据先验知识确定眼睛对。实验结果表明本方法简便易行，速度快，定位准确度较高。     

基于地形学标定的人脸眼睛定位

根据眼睛区域灰度固有特征,从灰度图像地形学标定的角度出发,找到眼睛点的候选集合.然后对平滑后的图像进行数学形态学的膨胀操作,以精简候选集合.最后用投影法对精简后的眼睛对进行验证,从而达到眼睛对精确定位的目的.通过不同质量的人脸数据库进行实验,该方法具有很好的光照鲁棒性,取得了较为满意的定位效果.     

人脸检测中的眼睛定位算法研究

眼睛是一个在人脸检测中极为重要的人脸特征,因此一种快速可靠的精确定位眼睛的算法对许多实际的应用是十分重要的。本文分析了几种常用的眼睛定位算法,并提出了一种基于肤色信息、人脸面部几何特征和人眼灰度信息的算法。算法采用由粗到细的检测策略,先对AnilK.Jain的Cb、Cr椭圆聚类方法进行了改进,用改进的算法进行肤色提取,经过肤色区域的分析,对人脸区域进行预检测,然后结合人眼几何特征进行初步定位,再利用人眼的灰度信息进行精确定位。该算法定位效率高,并对背景、尺寸等细节具有很好的适应性,在人脸实时检测系统中具有很好的应用价值。     

人脸检测中眼睛精确定位的研究

人脸检测在许多应用中都是重要的一个处理阶段,例如人脸识别、电视会议、人机界面等。眼睛是一个在人脸检测中极为重要的人脸特征,因此一种快速可靠的精确定位眼睛的算法对许多实际的应用是十分重要的。该文提出了一种新颖的精确定位眼睛的方法,该方法由两部分组成:第一部分,通过人脸区域分割、五官定位、人脸确认三个步骤在复杂背景中进行人脸区域的检测;第二部分,在检测到人脸区域和眼睛大致位置的基础上,用一种新的椭圆检测算法精确定位眼睛虹膜的位置。实验证明该文所提出的算法是快速可靠的。     

梯度估计在眼睛特征提取中的应用

眼睛特征的提取在表情分析、人脸合成和人脸识别等应用中起着非常重要的作用.目前广泛使用的眼睛特征提取算法是基于模板匹配的算法和基于特征点的算法.然而基于模板匹配的算法通常需要进行多个参数的选择,而且匹配过程非常耗时;特征点的算法则存在准确度较低的问题.针对以上问题提出了一种基于梯度信息提取虹膜中心、虹膜外轮廓和眼睛角点等眼睛特征的算法,算法用改进霍夫变换提取虹膜中心和外轮廓.在获得虹膜中心点后,参照中心位置选择眼睛角点候选区域并对其梯度进行分析.依据角点附近梯度分布与其它区域梯度分布不同的性质定位角点.仿真结果表明提出的算法能有效的提取眼睛特征,在测试数据上得到了比其它算法更高的准确率和更快的速度.     

基于综合特征的虹膜定位的算法

提出了利用虹膜边缘图像的几何特征以及相交弦的性质进行虹膜定位的快速算法.首先利用虹膜图像的整体灰度分布信息,利用边缘检测算子提取虹膜的内边缘图像,然后利用圆的相交弦性质以及投票策略提取虹膜中心,最后利用Hough变换的半径直方图投票提取虹膜内外径.实验结果证明其算法提高了虹膜定位的速度并具有较强的抗干扰能力.     

眨眼检测与眼睛跟踪

提出一种在视频人脸图像序列中,进行眼睛检测,跟踪和睁、闭状态判别的方法。通过眨眼检测,对眼睛进行定位;使用针对性强的内眼角提取算子,确定内眼角精确位置;利用内眼角特征来动态跟踪眼睛;睁眼模板在线生成和更新,通过当前眼睛区域和睁眼模板的相关分析来判别睁、闭眼状态。实验结果表明,算法在50场/s的处理速度下,内眼角点定位准确率达到98%以上,眨眼检测正确率为97.5%。     

用于驾驶疲劳检测的人眼定位及状态判别算法

基于驾驶疲劳研究中的视频图像,提出一种新的适用于驾驶疲劳检测的人眼定位及状态判别算法。定位算法利用Gabor小波变换对图像进行预处理,提取基于图像的地形图特征,并利用支持向量机的方法对候选特征区域进行验证,然后在定位区域对人眼状态进行判断。经实验验证,该方法可以定位睁眼及闭眼图像,并可以定性和定量判断人眼状态。采用该方法获得的人眼状态信息将为后续的驾驶疲劳分析提供重要的数据。     

Study on directional eye movements in non‐frontal face images for eye‐controlled interaction

Directional eye movements based eye‐controlled interaction focuses on interpreting the horizontal, vertical, and diagonal eye movements or their combinations as inputs to design user interfaces for people who suffer with severe mobility disabilities. In this paper, we take into consideration the inherent eye jitter and evaluate the accuracy of dynamic tracking of horizontal, vertical, diagonal, and rectangular eye movements prior to using them. We observe that the rectangular eye gesture composed of short horizontal and vertical eye movements has the best tracking accuracy in the presence of jitter. Finally, we present methods for identifying horizontal and vertical eye movements based on the trajectory of eye pupil centers from non‐frontal face images. We find that the methods are robust and effective within ±20°deflective azimuths of non‐frontal faces. This effectiveness is demonstrated by using the rectangular eye gesture as an interface to perform a painting task.     

Discrete area filters in accurate detection of faces and facial features

This paper introduces a new method for detection of faces and facial features. Proposed algorithm denies the thesis that bottom-up solutions can't work at reasonable speed. It introduces fast detection – about 9 frames per second for a 384 × 256 image – while preserving accurate details of the detection. Main experiments focus on the detection of the eye centers — crucial in many computer vision systems such as face recognition, eye movement detection or iris recognition, however algorithm is tuned to detect 15 fiducial face points. Models were trained on nearly frontal faces. Bottom-up approach allows to detect objects under partial occlusion — particularly two out of four face parts (left eye, right eye, nose, mouth) must be localized. Precision of the trained model is verified on the Feret dataset. Robustness of the face detection is evaluated on the BioID, LFPW, Feret, GT, Valid and Helen databases in comparison to the state of the art detectors.     

一种基于肤色模型和眼睛定位的人脸检测方法

提出了一种基于肤色模型和眼睛定位的人脸检测方法。首先利用肤色模型和掩膜进行粗定位,确定人脸可能区域,然后通过基于方向模板的眼睛定位进行人脸存在的确认和精确定位。实验证明了该方法对于复杂背景下人脸检测的有效性。     

基于光学图像的舰船航迹检测

舰船航迹检测在海上交通管制和军事领域中有着广泛的应用前景．现有的航迹检测方法大都基于SAR图像，信息不直观而且算法复杂．提出了一种新颖的基于光学图像的航迹检测方法．首先利用纹理分析方法判断待检测区域中是否含有航迹，然后在通过判断的区域中使用条纹增强算法，凸现航迹特征，最后使用Radon变换检测航迹，并对结果进行了优化．与现有的检测方法相比，该方法针对性强，复杂度低．使用该方法对实际航拍图片进行了检测实验，取得了很好的效果．     

Projection functions for eye detection

In this paper, the generalized projection function (GPF) is defined. Both the integral projection function (IPF) and the variance projection function (VPF) can be viewed as special cases of GPF. Another special case of GPF, i.e. the hybrid projection function (HPF), is developed through experimentally determining the optimal parameters of GPF. Experiments on three face databases show that IPF, VPF, and HPF are all effective in eye detection. Nevertheless, HPF is better than VPF, while VPF is better than IPF. Moreover, IPF is found to be more effective on occidental than on oriental faces, and VPF is more effective on oriental than on occidental faces. Analysis of the detections shows that this effect may be owed to the shadow of the noses and eyeholes of different races of people.     

Frame detection using gradients fuzzy logic and morphological processing for distant color eye images in an intelligent iris recognition system

The capture of an eye image with the occlusion of spectacles in a non-cooperative environment compromises the accuracy in identifying a person in an iris recognition system. This is due to the obstruction of the iris by the frame which tends to produce an incorrect estimation of the initial center of the iris and the pupil during the iris segmentation process. In addition, it also causes incorrect localization of the upper eyelid during the process of iris segmentation and sometimes, the edges of the frame are wrongly identified as the edges of the upper eyelid. A frame detection method which involves the combination of two gradients, namely the Sobel operator and high pass filter, followed by fuzzy logic and the dilation operation of morphological processing is proposed to identify the frame on the basis of different frame factors in the capture of a distant eye image. In addition, a different color space is applied and only a single channel is used for the process of frame detection. The proposed frame detection method provides the highest frame detection rate compared to the other methods, with a detection rate of more than 80.0%. For the accuracy of the iris localization, upper eyelid localization and iris recognition system, the proposed method gives more than 96.5% accuracy compared to the other methods. The index of decidability showed that the proposed method gives more than 2.35 index compared to the existing methods.     

一种快速人眼定位与跟踪算法的实现

提出了一种在复杂背景、光照、姿势变化条件下的人脸眼睛定位与跟踪算法。首先采用基于OpenCV的级联式AdaBoost对象检测算法进行人脸检测,并提出了解决平面和深度旋转的方法;然后采用基于瞳孔定位后的分类器法精确定位眼睛;最后采用卡尔曼粒子滤波算法进行人眼跟踪。实验结果表明,该算法在复杂背景下极大地提高了眼睛定位与跟踪的速度和精确度,并对光照、姿势不敏感。     

Hough transform for feature detection in panoramic images

Omni-directional sensors are useful in obtaining a 360° field-of-view. With a radially symmetric mirror and conventional lens system this can be achieved with a single camera. There are several proposed profiles for the mirror, but most violate the single viewpoint (SVP) criteria necessary to allow functional equivalence to the standard perspective projection, posing challenges that have not yet been addressed in the literature. Such a imaging system with a non-SVP optical system do not benefit from the affine quality of straight line features being represented as collinear points in the image plane. To utilize these non-SVP mirrors, a new method to recognize such features is required. This work describes an approach to detecting features in panoramic non-SVP images using a modified Hough transform. A mathematical model for this feature extraction process is given. Experimental results are presented to validate this model and show robust performance in identifying line features with only estimated calibration.