Effect of Severe Image Compression on Iris Recognition Performance

We investigate three schemes for severe compression of iris images in order to assess what their impact would be on recognition performance of the algorithms deployed today for identifying people by this biometric feature. Currently, standard iris images are 600 times larger than the IrisCode templates computed from them for database storage and search; but it is administratively desired that iris data should be stored, transmitted, and embedded in media in the form of images rather than as templates computed with proprietary algorithms. To reconcile that goal with its implications for bandwidth and storage, we present schemes that combine region-of-interest isolation with JPEG and JPEG2000 compression at severe levels, and we test them using a publicly available database of iris images. We show that it is possible to compress iris images to as little as 2000 bytes with minimal impact on recognition performance. Only some 2% to 3% of the bits in the IrisCode templates are changed by such severe image compression, and we calculate the entropy per code bit introduced by each compression scheme. Error tradeoff curve metrics document very good recognition performance despite this reduction in data size by a net factor of 150, approaching a convergence of image data size and template size.     

On techniques for angle compensation in nonideal iris recognition.

The popularity of the iris biometric has grown considerably over the past two to three years. Most research has been focused on the development of new iris processing and recognition algorithms for frontal view iris images. However, a few challenging directions in iris research have been identified, including processing of a nonideal iris and iris at a distance. In this paper, we describe two nonideal iris recognition systems and analyze their performance. The word "nonideal" is used in the sense of compensating for off-angle occluded iris images. The system is designed to process nonideal iris images in two steps: 1) compensation for off-angle gaze direction and 2) processing and encoding of the rotated iris image. Two approaches are presented to account for angular variations in the iris images. In the first approach, we use Daugman's integrodifferential operator as an objective function to estimate the gaze direction. After the angle is estimated, the off-angle iris image undergoes geometric transformations involving the estimated angle and is further processed as if it were a frontal view image. The encoding technique developed for a frontal image is based on the application of the global independent component analysis. The second approach uses an angular deformation calibration model. The angular deformations are modeled, and calibration parameters are calculated. The proposed method consists of a closed-form solution, followed by an iterative optimization procedure. The images are projected on the plane closest to the base calibrated plane. Biorthogonal wavelets are used for encoding to perform iris recognition. We use a special dataset of the off-angle iris images to quantify the performance of the designed systems. A series of receiver operating characteristics demonstrate various effects on the performance of the nonideal-iris-based recognition system.     

Fast discrete curvelet transform based anisotropic iris coding and recognition using k-out-of-n: A fused post-classifier

This paper presents rotation invariant technique for iris feature extraction and fused post-classification at the decision level to improve the performance under non-ideal environmental conditions. In this work, directional iris texture features based on two-dimensional (2D) Fast Discrete Curvelet Transform (FDCT) are computed. This approach divides the normalized iris image into six sub-images. The curvelet transform is applied on each sub-image. The feature vector for each sub-image is derived using the directional energies of these curvelet coefficients. These distances are fused at the decision level through novel post-classifier using k-out-of-n: A scheme to reduce the false rejection rate. The feasibility of the proposed algorithm has been tested using UBIRIS, MMU1 and CASIA-Iris V2.0 databases and performance is compared with some of the well-known existing iris recognition algorithms. The experimental results show that the performance is comparable with some of the state-of-the-art iris recognition algorithms.     

A novel biorthogonal wavelet network system for off-angle iris recognition

One important category of non-ideal conditions for iris recognition is off-angle iris images. Practically it is very difficult for images to be captured with no offset. It then becomes necessary to account for off angle information in order to maintain robust performance. A biorthogonal wavelet based iris recognition system, previously designed at our lab, is modified and demonstrated to perform off-angle iris recognition. Biorthogonal wavelet network (BWN) are developed and trained for each class. The non-ideal factors are adjusted by repositioning the BWN. To test, along with the real data, synthetic iris images are generated by using affine and geometric transforms of 0°, 10° and 20^° experimentally collected images. The tests were carried out on the experimentally collected off-angle data and synthetically generated data for angles from 0° to 60^° with a resolution of 5^°. This approach is shown to have less constraints than a transformation based iris recognition approach. Iris images off-angle by up to 42^° for synthetic data and up to 45^° for experimental data are successfully recognized.     

Meta-Analysis of Third-Party Evaluations of Iris Recognition

Iris recognition has long been widely regarded as a highly accurate biometric despite the lack of independent large-scale testing of its performance. Recently, however, three third-party evaluations of iris recognition were performed. This paper compares and contrasts the results of these independent evaluations. We find that despite differences in methods, hardware, and/or software, all three studies report error rates of the same order of magnitude: observed false nonmatch rates from 0.0122 to 0.03847 at a false match rate of 0.001. Furthermore, the differences between the best performers' error rates are an order of magnitude smaller than the observed error rates.      

CGAN样本生成的遥感图像飞机识别

目的 基于深度学习的飞机目标识别方法在遥感图像解译领域取得了很大进步,但其泛化能力依赖于大规模数据集。条件生成对抗网络（conditional generative adversarial network,CGAN）可用于产生逼真的生成样本以扩充真实数据集,但对复杂遥感场景的建模能力有限,生成样本质量低。针对这些问题,提出了一种结合CGAN样本生成的飞机识别框架。方法 改进条件生成对抗网络,利用感知损失提高生成器对遥感图像的建模能力,提出了基于掩膜的结构相似性（structural similarity,SSIM）度量损失函数（masked-SSIM loss）以提高生成样本中飞机区域的图像质量,该损失函数与飞机的掩膜相结合以保证只作用于图像中的飞机区域而不影响背景区域。选取一个基于残差网络的识别模型,与改进后的生成模型结合,构成飞机识别框架,训练过程中利用生成样本代替真实的卫星图像,降低了对实际卫星数据规模的需求。结果 采用生成样本与真实样本训练的识别模型在真实样本上的进行实验,前者的准确率比后者低0.33%;对于生成模型,在加入感知损失后,生成样本的峰值信噪比（peak signal to noise ratio,PSNR）提高了0.79 dB,SSIM提高了0.094;在加入基于掩膜的结构相似性度量损失函数后,生成样本的PSNR提高了0.09 dB,SSIM提高了0.252。结论 本文提出的基于样本生成的飞机识别框架生成了质量更高的样本,这些样本可以替代真实样本对识别模型进行训练,有效地解决了飞机识别任务中的样本不足问题。     

渐进式虹膜图像质量评估模型

虹膜图像质量评估目前尚无统一评估标准,导致虹膜识别拒识率和误识率较高。针对该问题,提出一种虹膜图像质量评估模型。根据虹膜图像中各干扰因素的不同特点,在先验知识基础上采用区域化、加权的方法,渐近式地实施像素级质量评估,依据像素级评估结果实施图像级质量评估。实验结果表明,像素级虹膜图像质量评估中的虹膜图像干扰项识别率和模糊识别率较高,图像级虹膜图像质量评估与人工评估结果相一致。     

A novel cancellable Iris template generation based on salting approach

The iris has been vastly recognized as one of the powerful biometrics in terms of recognition performance, both theoretically and empirically. However, traditional unprotected iris biometric recognition schemes are highly vulnerable to numerous privacy and security attacks. Several methods have been proposed to generate cancellable iris templates that can be used for recognition; however, these templates achieve lower accuracy of recognition in comparison to traditional unprotected iris templates. In this paper, a novel cancellable iris recognition scheme based on the salting approach is introduced. It depends on mixing the original binary iris code with a synthetic pattern using XOR operation. This scheme guarantees a high degree of privacy/security preservation without affecting the performance accuracy compared to the unprotected traditional iris recognition schemes. Comprehensive experiments on various iris image databases demonstrate similar accuracy to those of the original counterparts. Hence, robustness to several major privacy/security attacks is guaranteed.

     

A new multi-unit iris authentication based on quality assessment and score level fusion for mobile phones

Although iris recognition technology has been reported to be more stable and reliable than other biometric systems, performance can be degraded due to many factors such as small eyes, camera defocusing, eyelash occlusions and specular reflections on the surface of glasses. In this paper, we propose a new multi-unit iris authentication method that uses score level fusion based on a support vector machine (SVM) and a quality assessment method for mobile phones. Compared to previous research, this paper presents the following two contributions. First, we reduced the false rejection rate and improved iris recognition accuracy by using iris quality assessment. Second, if even two iris images were determined to be of bad quality, we captured the iris images again without using a recognition process. If only one iris image among the left and right irises was regarded as a good one, it was used for recognition. However, if both the left and right iris images were good, we performed multi-unit iris recognition using score level fusion based on a SVM. Experimental results showed that the accuracy of the proposed method was superior to previous methods that used only one good iris image or those methods that used conventional fusion methods.     

Fusing the information in visible light and near-infrared images for iris recognition

Automated human identification is a significant issue in real and virtual societies. Iris is a suitable choice for meeting this goal. In this paper, we present an iris recognition system that uses images acquired in both near-infrared and visible lights. These two types of images reveal different textural information of the iris tissue. We demonstrated the necessity to process both VL and NIR images to recognize irides. The proposed system exploits two feature extraction algorithms: one is based on 1D log-Gabor wavelet which gives a detailed representation of the iris region and the other is based on 1D Haar wavelet which represents a coarse model of iris. The Haar wavelet algorithm is proposed in this paper. It makes smaller iris templates than the 1D log-Gabor approach and yet achieves an appropriate recognition rate. We performed the fusion at the match score level and examined the performance of the system in both verification and identification modes. UTIRIS database was used to evaluate the method. The results were compared with other approaches and proved to have better recognition accuracy, while no image enhancement technique is utilized prior to the feature extraction stage. Furthermore, we demonstrated that fusion can compensate the lack of input image information, which can be beneficial in reducing the computation complexity and handling non-cooperative iris images.     

Hierarchical fusion of multi-spectral face images for improved recognition performance

This paper presents a two level hierarchical fusion of face images captured under visible and infrared light spectrum to improve the performance of face recognition. At image level fusion, two face images from different spectrums are fused using DWT based fusion algorithm. At feature level fusion, the amplitude and phase features are extracted from the fused image using 2D log polar Gabor wavelet. An adaptive SVM learning algorithm intelligently selects either the amplitude or phase features to generate a fused feature set for improved face recognition. The recognition performance is observed under the worst case scenario of using single training images. Experimental results on Equinox face database show that the combination of visible light and short-wave IR spectrum face images yielded the best recognition performance with an equal error rate of 2.86%. The proposed image-feature fusion algorithm also performed better than existing fusion algorithms.     

High-Performance Iris Recognition for Mobile Platforms

In spite of a fact that many standalone iris recognition solutions are successfully implemented and deployed around the world, development of a reliable iris recognition solution capable to provide high recognition performance (both in biometric quality and speed) on mobile device is still an actual task. Main issues related to iris recognition in the mobile devices consist in uncontrollable capturing conditions and limitations in computation power. The aim of the proposed approach is to eliminate aforementioned issues by providing user with comprehensive feedback and, at the same time, performing the most computationally complex operations only on the images of the best quality. Key features of the proposed approach are multi-stage algorithm structure, novel iris image quality estimation and adaptive iris feature vector quantization algorithms. These features allow to achieve high recognition accuracy and real-time performance which are proved by experimental results.     

IrisCodeNet：虹膜特征编码网络

使用有效的特征提取算法对虹膜纹理进行准确的表达是虹膜识别技术的关键。基于虹膜识别任务的特殊性,提出了用于虹膜特征编码的网络模型IrisCodeNet。该网络架构使用了改进的BasicBlock,并结合了可以扩大决策边界的损失函数AM-Softmax（additive margin softmax）。为了获取最佳的虹膜识别效果,对AM-Softmax的参数设置、虹膜图像预处理输入形式、数据增强方式、网络输入尺寸做了细致的研究。实验结果表明：使用IrisCodeNet训练得到的特征提取器在CASIA-Iris-Thousand、CASIA-Iris-Distance、IITD虹膜数据库上进行测试,所评估的等错误率（equal error rate,EER）和正确接受率（true acceptance rate,TAR）均远远超过了广泛应用的传统算法。特别地,IrisCodeNet无需传统的虹膜归一化或精确的虹膜分割步骤依然取得了极好的识别效果。并且使用Grad-CAM（gradient-weighted class activation mapping）算法进行了可视化分析,结果表明该网络框架有效地关注了虹膜纹理信息,从而证明了IrisCodeNet具有较强的虹膜纹理特征提取能力。     

融合局部与全局特征提取的虹膜识别方法

传统的虹膜识别系统需要将虹膜图像转换至极坐标系统并进行归一化,通过平移特征向量来达到旋转不变性。为了降低传统虹膜识别方法的复杂性,提出了一种融合局部与全局特征提取的虹膜识别方法,无须对预处理后的虹膜图像进行归一化。该方法首先对分割出的虹膜图像直接采用非张量积小波提取全局特征,接着采用SIFT方法提取选定区域的局部特征,最后对虹膜局部及全局特征采用不同的权值,进行相似性距离测试。结果表明该方法在等错误率为0.935%的情况下,正确识别率达到了99.065%。在不对虹膜图像归一化的情况下,可获得很好的识别性能。     

PI-Unet：异质虹膜精确分割神经网络模型的研究

在虹膜识别系统中,异质虹膜图像（可见光和红外图像）的分割是最重要且最有挑战性的一个任务,该任务的难点在于针对异质虹膜图像,要同时兼顾虹膜分割的准确率和快速性。提出了适用于异质虹膜分割的神经网络模型PI-Unet（Precise Iris Unet）以及用于训练该网络模型的数据增强方法和损失函数。对PI-Unet的Encoder和Decoder进行实验探索,得出能同时兼顾准确率和快速性的网络结构,将提出的数据增强方法和损失函数用于该网络进行训练,在CASIA-iris-intervel-v4和UBIRIS.v2虹膜图像数据库上测试该网络的准确率、参数量和计算量。测试结果表明,提出的数据增强方法和损失函数能有效提高异质虹膜分割准确率,PI-Unet与传统虹膜分割算法和其他虹膜分割神经网络相比,对异质虹膜图像的分割准确率更高且参数量和计算量更少,能够适用于低性能的边缘计算设备。     

基于虹膜纹理深度特征和Fisher向量的人种分类

目的 虹膜是位于人眼表面黑色瞳孔和白色巩膜之间的圆环形区域,有着丰富的纹理信息。虹膜纹理具有高度的区分性和稳定性。人种分类是解决虹膜识别在大规模数据库上应用难题的主要方法之一。现有的虹膜图像人种分类方法主要采用手工设计的特征,而且针对亚洲人和非亚洲人的基本人种分类,无法很好地解决亚种族分类问题。为此提出一种基于虹膜纹理深度特征和Fisher向量的人种分类方法。方法 首先用CNN（convolutional neural network）对归一化后的虹膜纹理图像提取深度特征向量,作为底层特征;然后使用高斯混合模型提取Fisher向量作为最终的虹膜特征表达;最后用支持向量机分类得到最终结果。结果 本文方法在亚洲人和非亚洲人的数据集上采用non-person-disjoint的方式取得99.93%的准确率,采用person-disjoint的方式取得91.94%的准确率;在汉族人和藏族人的数据集上采用non-person-disjoint的方式取得99.69%的准确率,采用person-disjoint的方式取得82.25%的准确率。结论 本文通过数据驱动的方式从训练数据中学习到更适合人种分类的特征,可以很好地实现对基本人种以及亚种族人种的分类,提高了人种分类的精度。同时也首次证明了用虹膜图像进行亚种族分类的可行性,对人种分类理论进行了进一步地丰富和完善。     

基于DSP的虹膜识别系统设计

本文设计了一种用于身份鉴别的虹膜识别系统。它以TI的TMS320DM642DSP为中央处理器,配合以A/D转换器和存储器等,完成虹膜图像的采集、处理和存储。主程序运行在DSP中,完成虹膜定位、特征分析和匹配等算法,得出识别结果。实验证明,系统具有较高的识别率,效果良好。     

Iris image reconstruction from binary templates: An efficient probabilistic approach based on genetic algorithms

A binary iriscode is a very compact representation of an iris image. For a long time it was assumed that the iriscode did not contain enough information to allow for the reconstruction of the original iris. The present work proposes a novel probabilistic approach based on genetic algorithms to reconstruct iris images from binary templates and analyzes the similarity between the reconstructed synthetic iris image and the original one. The performance of the reconstruction technique is assessed by empirically estimating the probability of successfully matching the synthesized iris image against its true counterpart using a commercial matcher. The experimental results indicate that the reconstructed images look reasonably realistic. While a human expert may not be easily deceived by them, they can successfully deceive a commercial matcher. Furthermore, since the proposed methodology is able to synthesize multiple iris images from a single iriscode, it has other potential applications including privacy enhancement of iris-based systems.     

Using empirical mode decomposition for iris recognition

Iris recognition is known as an inherently reliable technique for human identification. Empirical Mode Decomposition (EMD), an adaptive multi-resolution decomposition technique, appears to be suitable for non-linear, non-stationary data analysis. Based on EMD, a fully data-driven method without using any pre-determined filter or wavelet function, an iris recognition scheme is presented by modifying EMD as a low-pass filter to analyze the iris images. To evaluate the efficacy of the proposed approach, three different similarity measures are used. Experimental results show that three metrics have all achieved promising and similar performance. Therefore, the proposed method demonstrates to be feasible for iris recognition and EMD is suitable for feature extraction.