Face and eyes localization algorithm in thermal images for temperature measurement of the inner canthus of the eyes

In this paper, a novel algorithm for the detection and localization of the face and eyes in thermal images is presented, particularly the temperature measurement of the human body by measuring the eye corner (inner canthus) temperature. The algorithm uses a combination of the template-matching, knowledge-based and morphological methods, particularly the modified Randomized Hough Transform (RHT) in the localization process, also growing segmentation to increase accuracy of the localization algorithm. In many solutions, the localization of the face and/or eyes is made by manual selection of the regions of the face and eyes and then the average temperature in the region is measured. The paper also discusses experimental studies and the results, which allowed the evaluation of the effectiveness of the developed algorithm. The standardization of measurement, necessary for proper temperature measurement with the use of infrared thermal imaging, are also presented.     

A novel algorithm for automatic localization of human eyes

Based on geometrical facial features and image segmentation, we present a novel algorithm for automatic localization of human eyes in grayscale or color still images with complex background. Firstly, a determination criterion of eye location is established by the prior knowledge of geometrical facial features. Secondly, a range of threshold values that would separate eye blocks from others in a segmented face image (i.e., a binary image) are estimated. Thirdly, with the progressive increase of the threshold by an appropriate step in that range, once two eye blocks appear from the segmented image, they will be detected by the determination criterion of eye location. Finally, the 2D correlation coefficient is used as a symmetry similarity measure to check the factuality of the two detected eyes. To avoid the background interference, skin color segmentation can be applied in order to enhance the accuracy of eye detection. The experimental results demonstrate the high efficiency of the algorithm and correct     

基于人脸姿态估计的虚拟眼镜试戴技术

为解决用户线上眼镜的最优选购,提出了一种基于人脸姿态估计的虚拟眼镜试戴技术。首先采用肤色模型与形状模型结合的算法对场景中的人脸区域进行检测,然后根据人眼在人脸中的几何位置关系实现人眼的精确定位;进一步利用人眼对称性先验知识来估计脸在三维空间中的姿态信息,即人脸与正面的偏移角度;最后,依据人眼位置和人脸姿态将眼镜图像融合到眼睛区域,即完成眼镜的虚拟试戴。该方法为3D环境下客户与商品之间虚拟视觉化的实现提供了一种可靠的技术支撑和应用思路。     

Adaptive optics fundus camera using a liquid crystal phase modulator

We have developed an adaptive optics (AO) fundus camera to obtain high resolution retinal images of eyes. We use a liquid crystal phase modulator to compensate the aberrations of the eye for better resolution and better contrast in the images. The liquid crystal phase modulator has a wider dynamic range to compensate aberrations than most mechanical deformable mirrors and its linear phase generation makes it easy to follow eye movements. The wavefront aberration was measured in real time with a sampling rate of 10 Hz and the closed loop system was operated at around 2 Hz. We developed software tools to align consecutively obtained images. From our experiments with three eyes, the aberrations of normal eyes were reduced to less than 0.1 μm (RMS) in less than three seconds by the liquid crystal phase modulator. We confirmed that this method was adequate for measuring eyes with large aberrations including keratoconic eyes. Finally, using the liquid crystal phase modulator, high resolution images of retinas could be obtained.     

基于图像处理的人眼注视方向检测研究

提出了一种基于图像处理技术以实现人眼注视方向的实时检测方法。研究了人眼瞳孔、眼眶的位置、形状及其与视线方向的关系。以人眼注视不同方向时,眼睛瞳孔位置和眼眶形状产生的差异为基础,通过对所拍摄的眼部图片进行实时处理与分析,实现了对人眼注视方向的实时检测。实验结果表明了该方法的可行性及有效性。     

Contrast Adaptation Effects under Interocular Suppression for Normal and Strabismic Observers

Our vision system receives two different images from our two eyes. Since these two images usually have high correlation they are fused into a single stable image. In binocular rivalry suppression a stimulus presented in one eye is suppressed and a different stimulus in the other eye is perceived for a certain time interval. For strabismic observers the suppression is continuous between different retinal images caused by misalignment of eye positions. Previous research investigated the effects of this binocular rivalry suppression on a number of visual processes. The present study looked at the effects of contrast adaptation for the adaptation stimulus presented to eyes under interocular suppression. We compared normal and strabismic observers and found that in both types, regardless of the existence of suppression, there was the effect of contrast adaptation. These findings provide evidence that the suppression occurs at a site in the visual system after the locus of contrast adaptation. The contrast adaptation strength differed in the two types of observers.     

利用几何特性及神经网络进行人脸探测技术的研究

在人脸识别过程中 ,首先也是最重要的一个环节是人脸探测 ,因为一旦从图像中定位并提取到了人脸 ,那么下一步的人脸识别工作就变得非常容易。眼睛是人脸图像中最容易探测的部位 ,而且通过探测双眼来发现人脸最符合人的视觉习惯。提出了一种基于几何特征分析和人工神经网络的由粗到细的两级人脸探测方法。在第一级中 ,眼睛和脸是通过测量眼睛的尺寸和眼睛与脸的位置关系探测到的 ,第一级的输出是一个尺寸归一化的人脸 ,但偶尔也伴随着一个或多个因对复杂背景中与眼睛类似的物体的误判而得到的非人脸图像 ;第二级神经网络正是用来过滤掉第一级中被误判的人脸。实验表明 ,这种由粗到细的两级人脸探测系统具有很高的稳定性和探测正确率     

一种通用的红外人脸器官定位算法

红外人脸热像图是红外成像领域的研究热点之一,而对作为其基础性研究的通用红外人脸定位算法却研究的较少。利用由人脸生理结构造成的温度特异性和由温度变化形成的图像边缘,给出了一种通用的人脸器官定位算法。首先利用温度特异性定位鼻孔等温度特异性强的区域,然后以此为参考点,结合面部器官分布特点,对融合提取的图像边缘信息进行眼睛等其他器官的定位。实验结果表明,该方法对各器官的平均定位精度均达到了90%以上,具有一定的实用价值。另外,该方法不仅可以对常规的眼、鼻、口等器官进行定位,而且还可以对眉骨、瞳孔等位置进行定位。     

Reconstruction of eye movements during blinks

In eye movement research in reading, the amount of data plays a crucial role for the validation of results. A methodological problem for the analysis of the eye movement in reading are blinks, when readers close their eyes. Blinking rate increases with increasing reading time, resulting in high data losses, especially for older adults or reading impaired subjects. We present a method, based on the symbolic sequence dynamics of the eye movements, that reconstructs the horizontal position of the eyes while the reader blinks. The method makes use of an observed fact that the movements of the eyes before closing or after opening contain information about the eyes movements during blinks. Test results indicate that our reconstruction method is superior to methods that use simpler interpolation approaches. In addition, analyses of the reconstructed data show no significant deviation from the usual behavior observed in readers.     

Design of a computer game using an eye-tracking device for eye's activity rehabilitation

An eye mouse interface that can be used to operate a computer using the movement of the eyes is described. We developed this eye-tracking system for eye motion disability rehabilitation. When the user watches the screen of a computer, a charge-coupled device will catch images of the user's eye and transmit it to the computer. A program, based on a new cross-line tracking and stabilizing algorithm, will locate the center point of the pupil in the images. The calibration factors and energy factors are designed for coordinate mapping and blink functions. After the system transfers the coordinates of pupil center in the images to the display coordinate, it will determine the point at which the user gazed on the display, then transfer that location to the game subroutine program. We used this eye-tracking system as a joystick to play a game with an application program in a multimedia environment. The experimental results verify the feasibility and validity of this eye-game system and the rehabilitation effects for the user's visual movement.     

Sucrose cryo-protection facilitates imaging of whole eye sections by MALDI mass spectrometry

Sucrose is used as a cryo-preservation agent on large mammalian eyes post formalin fixation and is shown to reduce freezing artefacts allowing the collection of 12-μm thick sections from these large aqueous samples. The suitability of this technique for use in MALDI imaging experiments is demonstrated by the acquisition of the first images of lipid distributions within whole sagittal porcine eye sections.     

Eye location under different eye poses,scales,and illuminations

Robust non-intrusive eye location plays an important role in vision-based man-mechine interaction. A modified Hausdorff distance based measure to localize the eyes is proposed, which could tolerate various changes in eye pose, shape, and scale. To eliminate the effects of the illumination variations, an 8- neighbour-based transformation of the gray images is proposed. The transformed image is less sensitive to illumination changes while preserves the appearance information of eyes. All the localized candidates of eyes are identified by back-propagation neural networks. Experiments demonstrate that the robust method for eye location is able to localize eyes with different eye sizes, shapes, and poses under different illuminations.     

Measurement of Wavefront Aberration of Human Eye Using Talbot Image of Two-Dimensional Grating

The measurement of human ocular aberration is now frequently performed because of the increase in refractive surgery on the human cornea. The Hartmann-Shack (H-S) wavefront sensor is considered to be the most useful wavefront sensor, and a calculation method for wavefront aberration has been established. New methods of measuring wavefront aberrations of human eyes, using the Talbot image of a two-dimensional grating as a wavefront sensor and local shift of the Talbot image to calculate tilt of the wavefront are shown. The shift of the Talbot image was determined by comparing the phases of fundamental spatial frequency between the grating and the local patch of the Talbot image by Fourier transformation. The actual experiment was performed using a modified commercially available wavefront analyzer. Using these methods, Talbot images were obtained from model eyes and a human eye, and wavefront shapes were successfully reconstructed. Wavefront aberrations can be measured even when the obtained image is degraded by defocusing or scattering.     

基于人眼视觉系统的假彩色融合图像质量的评价方法

随着图像融合技术的发展,各种融合算法层出不穷,而很多情况下最终的融合图像是由人眼观察的,因此基于人眼视觉系统的图像融合质量评价显得尤为重要.为了能够模拟人眼对于融合图像的感知,得到融合后图像质量的客观评价,本文提出了一种基于色差理论的假彩色融合图像质量的评价方法.首先将源图像和融合图像转化到CIE L*a*b*均匀色空间,在频域对图像进行对比度敏感函数滤波,通过计算滤波后融合图像的色差判断图像的细节信息,在一定程度上色差越大信息越丰富;通过计算融合图像与源图像的色差判断融合图像与源图像的相关性,相关性越高,融合算法越好.通过融合图像的色差大小以及与源图像的相关性两个参量,得出融合算法的优劣.实验表明,与其他评价方法相比,本文提出的评价方法与人眼观察的结果较为一致.     

Face recognition based on PCA image reconstruction and LDA

Face recognition has become a research hotspot in the field of pattern recognition and artificial intelligence. Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) are two traditional methods in pattern recognition. In this paper, we propose a novel method based on PCA image reconstruction and LDA for face recognition. First, the inner-classes covariance matrix for feature extraction is used as generating matrix and then eigenvectors from each person is obtained, then we obtain the reconstructed images. Moreover, the residual images are computed by subtracting reconstructed images from original face images. Furthermore, the residual images are applied by LDA to obtain the coefficient matrices. Finally, the features are utilized to train and test SVMs for face recognition. The simulation experiments illustrate the effectivity of this method on the ORL face database.     

红外图像的疲劳状态识别方法

目的：为了解决光照变化对疲劳检测系统造成的准确性不高的问题,提出了一种近红外环境下判断人眼状态的方法,即针对红外光补图像的人眼状态判断。方法：首先,利用Adaboost算法对人眼区域进行定位,在网格法标记人眼瞳孔部分的基础上,进行Retinex红外图像增强。接着,对二值化与边缘检测后的红外图像分别进行网格法闭合度计算,得到人眼闭合度。最后,根据闭合度计算结果设定双阈值并结合PERCLOS来判断眼睛状态。结果：在DM642硬件平台上进行疲劳检测试验,实验结果表明,人眼状态识别率达到了90%以上,且平均每秒能处理21帧图片。结论：证明了该方法不仅能有效解决光照变化带来的问题,而且满足疲劳状态检测系统的快速性、可靠性和有效性等要求。     

高准确度光学系统中材料非均匀性对成像质量的影响

利用Zernike多项式对用Zygo干涉仪测得的离散材料折射率数据进行了拟合,再使用光线光学的方法评价了系统的成像质量. 由于材料折射率分布的无规则性,在对包含非均匀介质的实际光学系统的模拟仿真和优化时,需要考虑选取材料不同部位加工成的透镜会对系统成像质量有不同的影响,而且加工好的透镜在装配过程中,绕着光轴旋转不同的角度同样会影响成像质量. 通过计算机模拟的方法预先选取材料的最佳部位以及找到最好的装配位置,从而提高了光学系统的性能.     

Closed-loop adaptive optics in the human eye

We have developed a prototype apparatus for real-time closed-loop measurement and correction of aberrations in the human eye. The apparatus uses infrared light to measure the wave-front aberration at 25 Hz with a Hartmann-Shack sensor. Defocus is removed by a motorized optometer, and higher-order aberrations are corrected by a membrane deformable mirror. The device was first tested with an artificial eye. Correction of static aberrations takes approximately five iterations, making the system capable of following aberration changes at 5 Hz. This capability allows one to track most of the aberration dynamics in the eye. Results in living eyes showed effective closed-loop correction of aberrations, with a residual uncorrected wave front of 0.1microm for a 4.3-mm pupil diameter. Retinal images of a point source in different subjects with and without adaptive correction of aberrations were estimated in real time. The results demonstrate real-time closed-loop correction of aberration in the living eye. An application of this device is as electro-optic "spectacles" to improve vision.     

一种以超分辨率理论为基础的磁共振眼球成像方法

磁共振成像（MRI）是一种无电离辐射的非介入性的眼内肿瘤检测方法,但分辨率和运动伪影是成像过程中不易克服的困难．以往的扫描方法或是不可避免的引入运动伪影,或是需要受试者做精确的配合,增加了成像的难度,给受试者带来不舒适的体验．本文提出了一种以超分辨率理论为基础的新的磁共振眼球成像方法,使用一种特制的眼球线圈,对眼部区域扫描一系列动态的图像,使得不同方向上的采集分辨率互补．最后经过预处理、配准、超分辨率重建等操作,得到高质量的磁共振眼球图像．实验结果表明,这种方法可以在不需要受试者做额外配合工作的情况下,得到更加清晰的磁共振眼球图像．