A Robust Method for Near Infrared Face Recognition Based on Extended Local Binary Pattern

Serving as an effective texture description operator, local binary pattern (LBP) has been applied in the visible face recognition successfully. In order to enhance the recognition performance, the technology of face recognition depending on the near infrared (NIR) image has attracted extensive attention in the recent years. Although the characteristics of LBP have been researched in optical image thoroughly, the development of these do not catch enough attention in the NIR image at all. Therefore, in our paper, we study the characteristics of LBP from the following two aspects in NIR image. On one hand, we come to the probability distribution of various patterns of LBP in the NIR facial image. On the other hand, we discuss the influence to LBP caused by the illumination change in facial image.

A robust face recognition method that utilises a set of combined features is proposed to effectively conduct face recognition with images taken under diverse illumination variations. This method extracts discriminant features from different methods, both of which have different characteristics. To exploit the respective advantage of each method, the respective discriminability of the features extracted by each method is measured based on the discriminant distance criterion of each method. The experimental results show that the proposed features result in improved recognition performance under illumination variation.

Face recognition has been widely studied due to its importance in different applications; however, most of the proposed methods fail when face images are occluded or captured under illumination and pose variations. Recently several low-rank dictionary learning methods have been proposed and achieved promising results for noisy observations. While these methods are mostly developed for single-modality scenarios, recent studies demonstrated the advantages of feature fusion from multiple inputs. We propose a multi-modal structured low-rank dictionary learning method for robust face recognition, using raw pixels of face images and their illumination invariant representation. The proposed method learns robust and discriminative representations from contaminated face images, even if there are few training samples with large intra-class variations. Extensive experiments on different datasets validate the superior performance and robustness of our method to severe illumination variations and occlusion.

Accurate localization of representative points of a face is crucial to many face analysis and synthesis problems. Active shape model (ASM) is a powerful statistical tool for face alignment. However, it suffers from variations of pose, illumination and expressions. In this paper, we analyze the mechanism of active shape model and realize that the ability of normal profiles to describe the local appearance pattern is very limited. For efficient appearance pattern representation, the local binary pattern is used and extended to describe the local patterns of facial key points. For the purpose of retaining the spatial information, sub-images of key points are divided into several regions, which are combined to define the extended local binary pattern (ELBP) histogram. Then we propose an improved ASM method framework, ELBP-ASM, in which local appearance patterns of key points are modelled using extended local binary pattern. Experimental results demonstrate that ELBP-ASM achieves more accurate results compared with original method used in ASM.

Face recognition technology has evolved as a popular identification technique to perform verification of human identity. By using the feature extraction methods and dimensionality reduction techniques in the pattern recognition applications, a number of facial recognition systems has been produced with distinct measure of success. Various face recognition algorithms and their extensions, have been proposed in the past three decades. However, face recognition faces challenging problems in real life applications because of the variation in the illumination of the face images. In the recent years, the research is focused towards Illumination-invariant face recognition system and many approaches have been proposed. But, there are several issues in face recognition across illumination variation which still remains unsolved. This paper provides a research on an efficient illumination-invariant face recognition system using Discrete Cosine Transform (DCT) and Principal Component Analysis (PCA). For processing the illumination invariant image, low frequency components of DCT are used to normalize the illuminated image, odd and even components of DCT is used for compensation in illumination variation and PCA is used for recognition of face images. The existing approaches in illumination Invariant face recognition are comprehensively reviewed and discussed. The proposed approach is validated with Yale Face Database B. Experimental results demonstrate the effectiveness of this approach in the performance of face recognition.

Tumors contain both microscopic and macroscopic areas of necrotic cells. These have abnormal cell membrane permeability, which allows passage of large macromolecules such as antibodies into the cell. 131I-chTNT-1/B antibody (Cotara) is a novel, single administration agent designed to target tumor necrosis that has completed a Phase II clinical trial for recurrent glioblastoma multiforme (GBM), showing 9.3 month median overall survival for patients. Since 131I-chTNT-1/B antibody is a high-molecular-mass protein, it cannot easily pass from the systemic circulation through the blood-brain barrier (BBB) and thus can only be delivered locally by an intratumoral catheter. We have developed a human ME-180 cervical carcinoma xenograft tumor model in nude mice to characterize the targeting of chTNT-1/B to necrotic areas using near infrared (NIR) optical imaging. The use of NIR optical imaging is a powerful translational tool to monitor in vivo antibody targeting. chTNT-1/B was labeled with a NIR fluorescent dye and kinetics of antibody retention in tumors was measured by NIR optical imaging following intratumoral injection of the NIR-chTNT-1/B. NIR-chTNT-1/B was retained in ME180 tumors for a longer duration compared to NIR dye alone. Labeling of chTNT-1/B with a low dye/protein ratio was necessary to demonstrate maximum specific targeting to tumor necrosis. Evaluation of in vivo antibody targeting to tumors using NIR imaging may facilitate improvements for treatment of patients with malignant glioma with Cotara. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 2454. doi:1538-7445.AM2012-2454

There usually exist diverse variations in face images taken under uncontrolled conditions. Most previous work on face recognition focuses on particular variations and usually assume the absence of others. Such work is called controlled face recognition. Instead of the ‘divide and conquer’ strategy adopted by controlled face recognition, this paper presents one of the first attempts directly aiming at uncontrolled face recognition. The solution is based on Individual Stable Neural Network (ISNN) proposed in this paper. ISNN can map a face image into the so-called Individual Stable Space (ISS), the feature space that only expresses personal characteristics, which is the only useful information for recognition. There are no restrictions for the face images fed into ISNN. Moreover, unlike many other robust face recognition methods, ISNN does not require any extra information (such as view angle) other than the personal identities during training. These advantages of ISNN make it a very practical approach for uncontrolled face recognition. In the experiments, ISNN is tested on two large face databases with vast variations and achieves the best performance compared with several popular face recognition techniques.

Face recognition (FR) is a practical application of pattern recognition (PR) and remains a compelling topic in the study of computer vision. However, in real-world FR systems, interferences in images, including illumination condition, occlusion, facial expression and pose variation, make the recognition task challenging. This study explored the impact of those interferences on FR performance and attempted to alleviate it by taking face symmetry into account. A novel and robust FR method was proposed by combining multi-mirror symmetry with local binary pattern (LBP), namely multi-mirror local binary pattern (MMLBP). To enhance FR performance with various interferences, the MMLBP can 1) adaptively compensate lighting under heterogeneous lighting conditions, and 2) generate extracted image features that are much closer to those under well-controlled conditions (i.e., frontal facial images without expression). Therefore, in contrast with the later variations of LBP, the symmetrical singular value decomposition representation (SSVDR) algorithm utilizing the facial symmetry and a state-of-art non-LBP method, the MMLBP method is shown to successfully handle various image interferences that are common in FR applications without preprocessing operation and a large number of training images. The proposed method was validated with four public data sets. According to our analysis, the MMLBP method was demonstrated to achieve robust performance regardless of image interferences.

Facial recognition is a critical biometric identification method in modern security systems, yet it faces significant challenges under varying lighting conditions, particularly when dealing with near-infrared (NIR) images, which exhibit reduced illumination compared to visible light (VIS) images. This study aims to evaluate the performance of Convolutional Neural Networks (CNNs) in addressing the Cross-Spectral Cross-Distance (CSCD) challenge, which involves face identification across different spectra (NIR and VIS) and varying distances. Three CNN models—VGG16, ResNet50, and EfficientNetB0—were assessed using a dataset comprising 800 facial images from 100 individuals, captured at four different distances (1m, 60m, 100m, and 150m) and across two wavelengths (NIR and VIS). The Multi-task Cascaded Convolutional Networks (MTCNN) algorithm was employed for face detection, followed by image preprocessing steps including resizing to 224x224 pixels, normalization, and homomorphic filtering. Two distinct data augmentation strategies were applied: one utilizing 10 different augmentation techniques and the other with 4 techniques, trained with a batch size of 32 over 100 epochs. Among the tested models, VGG16 demonstrated superior performance, achieving 100% accuracy in both training and validation phases, with a training loss of 0.55 and a validation loss of 0.612. These findings underscore the robustness of VGG16 in effectively adapting to the CSCD setting and managing variations in both lighting and distance.

Robust discriminative nonnegative dictionary learning for occluded face recognition

Face recognition is one of the biometric identification methods with the highest potential. The existing face recognition algorithms relying on the texture information of face images are affected greatly by the variation of expression, scale and illumination. Whereas the algorithms based on the shape topology weaken the influence of illumination to some extent, but the impact of expression, scale and illumination on face recognition is still unsolved. To this end, we propose a new method for face recognition by integrating texture information with shape information, called biview face recognition algorithm. The texture models are constructed by using subspace learning methods and shape topologies are formed by building graphs for face images. The proposed biview face recognition method is compared with recognition algorithms merely based on texture or shape information. Experimental results of recognizing faces under the variation of illumination, expression and scale demonstrate that the performance of the proposed biview face recognition outperforms texture-based and shape-based algorithms.

An improved collaborative representation based classification with regularized least square (CRC–RLS) method for robust face recognition

Converting near-infrared (NIR) images into color images is a challenging task due to the different characteristics of visible and NIR images. Most methods of generating color images directly from a single NIR image are limited by the scene and object categories. In this paper, we propose a novel approach to recovering object colors from multi-spectral NIR images using gray information. The multi-spectral NIR images are obtained by a 2-CCD NIR/RGB camera with narrow NIR bandpass filters of different wavelengths. The proposed approach is based on multi-spectral NIR images to estimate a conversion matrix for NIR to RGB conversion. In addition to the multi-spectral NIR images, a corresponding gray image is used as a complementary channel to estimate the conversion matrix for NIR to RGB color conversion. The conversion matrix is obtained from the ColorChecker's 24 color blocks using polynomial regression and applied to real-world scene NIR images for color recovery. The proposed approach has been evaluated by a large number of real-world scene images, and the results show that the proposed approach is simple yet effective for recovering color of objects.

In recent years, single-mode face recognition has achieved great progress under the impetus of a variety of applications, and its recognition rate can reach as high as 99%. Heterogeneous face biometrics, namely face mapping between images captured in different spectral bands, has now become a hotspot. Our paper focuses on the problem of face recognition between registered visual (VIS) images and predicted near infrared (NIR) images, and proposes a new method, based on extant VIS and NIR images sets to synthesize corresponding VIS image by analyzing the imaging model of heterogeneous image pairs, thereby performing the eclectic face mapping. Experiments show that the synthesized images greatly reduce the difference between them, and face recognition based on it shows promising results.

This meta-analysis was conducted to evaluate the diagnostic value of near-infrared (NIR) or fluorescent indocyanine green (ICG) guided sentinel lymph node (SLN) mapping in gastric cancer (GC). This meta-analysis was registered at the PROSPERO. Clinical studies were retrieved from the electronic database Pubmed, Embase, Medline, Web of science, and the Cochrane Library. Quality assessment was conducted by an adapted checklist of QUADAS-2. A bivariate mixed-effects model was used to pool the data. Evaluation of articles quality, analysis for publication bias, summary receiver operator characteristic (SROC) curves, and meta-regression were also performed. Subgroup analysis was used to explain the heterogeneities. A total of 13 clinical studies (971 patients) were included. The NIR or fluorescent imaging (FI) involved infrared ray electronic endoscopy (IREE), infrared ray laparoscopic system (IRLS), and FI system. Significant evidence of heterogeneity was found for sensitivity and specificity (I 2 = 91.1% and I 2 = 98.2%), respectively. The pooled SLN sensitivity (Sen), specificity (Spe), positive likelihood ratio (PLR), negative likelihood ratio (NLR), and diagnostic odds ratio (DOR) were 0.94(95%CI: 0.80-0.99), 1.00(95%CI: 0.60-1.00), 34.0(95%CI: 9.25-125.29), 0.06(95%CI:0.02-0.22), and 252.50(95%CI: 94.93-671.61), respectively. Area under curve (AUC) of SROC curve was 1.00 (95%CI: 0.99-1.00), and the summary operating point (cut-off value) was SENS = 0.94(95%CI: 0.80-0.99) and SPEC = 1.00(95%CI: 0.60-1.00). Subgroup analysis showed that NIR imaging, imaging performed 20 minutes after intraoperative injection, preoperative injection (especially for FI imaging), stained with immunohistochemistry (IHC) (+hematoxylin-eosin [HE]), cT1 stage, submucosa injection (especially for cT1), mean number of SLN ≥ 5, study size > 26 were associated with higher SLN sensitivity. In terms of ICG concentration, diluted ICG concentration that 0.5 mg/mL (compared with 5 mg/mL) in NIR imaging and 0.05 mg/mL (compared with 0.5 mg/mL) in FI system showed higher sensitivities. However, the differences in tumor diameter (≤30 mm vs >30 mm), gastrectomy methods (opening vs laparoscopy), lymphadenectomy methods (LBD vs pick-up), and publication year (≥2010 vs <2010) did not achieve statistical significance. ICG combined with NIR or FI guided SLN mapping is technically feasible for GC. Based on the small sample size evidence, the IREE and IRLS devices may have higher sensitivity than FI in current clinical studies; and there may be an excessive ICG concentration used for current SLN mapping in GC. However, well-designed further studies with large sample size are needed to confirm the best procedure and suitable criteria. This meta-analysis was registered at PROSPERO. Clinical studies on this topic were retrieved from the electronic database Pubmed, Embase, Medline, Web of Science, and the Cochrane Library. The NIR or FI involved IREE, IRLS, and FI techniques. A total of 13 clinical studies (971 patients) were included. Based on the small sample size evidence, NIR or FI ICG guided SLN mapping in GC is technically feasible.