Weighted Score Level Fusion Research Articles

Presentation attack detection based on score level fusion and challenge-response technique

Biometrics is the state of the art in dealing with identity identification and verification based on the physical and behavioral characteristics and widely used in the fields of Fintech, such as mobile payment and online banking due to its security and convenience. However, there are various attacks against the biometrics system. The presentation attack is one of the most common attacks that an imposter presents fake biometrics to the sensor trying to fool the system. This paper proposes a multimodal presentation attack detection (PAD) method against photo-attack and video-attack in face recognition system by using score level fusion and challenge-response scenario. The proposed challenge-response scenario is that requesting the user to speak out the randomly prompted words. Then, the recognized speech text and the user’s mouth motion are detected simultaneously to verify if the user is liveness. Two weighted score level fusion rules, namely weighted sum and weight product, are used to combine the speech and mouth motion traits as a matching score. The final score is fed into supervised machine learning algorithms and trained for classifying spoofing. The experiments are conducted in the self-built database. Experimental results show that the proposed method can achieve the best half total error rate at 3.64% and can effectively improve facial recognition system security.

Human identification after plastic surgery using region based score level fusion of local facial features

Plastic surgery alters original facial features of an individual thereby making Face Recognition after plastic surgery difficult. Cosmetic procedures introduce geometrical deviations which are difficult to analyze by state of the art facial identification procedures. Here a region based score level fusion approach for local facial features is proposed to equalize former and latter surgery images. Steps involved in the recognition process are; firstly identifying the ROIs (areas/regions of interest/concern) of before and after surgery images. ROIs are eyes, nose and mouth regions; feature extraction from identified regions via Speeded Up Robust Features and K Nearest Neighbour techniques; region wise and full face geometrical distance calculation between matched feature vectors of pre and post surgery image samples by a distance metric (sum of squared differences); final recognition rate calculation by weighted score level fusion. The projected procedure gives recognition of 89.7% for local surgical treatment and 87% for global surgery.

An improved SRC method based on virtual samples for face recognition

The sparse representation classifier (SRC) performs classification by evaluating which class leads to the minimum representation error. However, in real world, the number of available training samples is limited due to noise interference, training samples cannot accurately represent the test sample linearly. Therefore, in this paper, we first produce virtual samples by exploiting original training samples at the aim of increasing the number of training samples. Then, we take the intra-class difference as data representation of partial noise, and utilize the intra-class differences and training samples simultaneously to represent the test sample in a linear way according to the theory of SRC algorithm. Using weighted score level fusion, the respective representation scores of the virtual samples and the original training samples are fused together to obtain the final classification results. The experimental results on multiple face databases show that our proposed method has a very satisfactory classification performance.

Joint features classifier with genetic set for undersampled face recognition

Face recognition with limited training samples is a very difficult task. Especially in face recognition featuring one training image per individual, it even seems to be impossible to enable a superb accuracy. In this paper, we present a novel joint features classification approach with an external generic set for face recognition. The presented scheme leverages two representations based on Gabor feature and local Gabor binary patterns (LGBP) feature. Firstly, Gabor feature-based representation with an external generic set and LGBP feature-based representation with an external generic set are obtained independently. Then a weighted score level fusion scheme is adopted to automatically combine Gabor feature and LGBP feature, and to output the final decision. Three metrics, i.e., recognition rate, stability and execution time, are investigated in our evaluation of the performance of the presented method. The comprehensive experimental results on three large face databases (i.e., AR, FERET and WLF) demonstrated that the presented approach can always achieve very satisfactory accuracy and stability and that it is computationally tractable.

Video-based emotion recognition in the wild using deep transfer learning and score fusion

Multimodal recognition of affective states is a difficult problem, unless the recording conditions are carefully controlled. For recognition “in the wild”, large variances in face pose and illumination, cluttered backgrounds, occlusions, audio and video noise, as well as issues with subtle cues of expression are some of the issues to target. In this paper, we describe a multimodal approach for video-based emotion recognition in the wild. We propose using summarizing functionals of complementary visual descriptors for video modeling. These features include deep convolutional neural network (CNN) based features obtained via transfer learning, for which we illustrate the importance of flexible registration and fine-tuning. Our approach combines audio and visual features with least squares regression based classifiers and weighted score level fusion. We report state-of-the-art results on the EmotiW Challenge for “in the wild” facial expression recognition. Our approach scales to other problems, and ranked top in the ChaLearn-LAP First Impressions Challenge 2016 from video clips collected in the wild.

Kernel discriminant analysis of Block-wise Gaussian Derivative Phase Pattern Histogram for palmprint recognition

This paper presents an efficient palmprint recognition technique for palmprints collected with visible as well as multispectral imaging system. ROI extraction is a challenging task for palmprint captured in unconstrained environment. ROI extracted by gaps between fingers and width of palm makes system rotation and translation invariant. Approximation ROI obtained by First-level decomposition of ROI using Haar wavelet reduces computational overhead as well as noise. Phase quantization of AROI by Gaussian derivative filter gives Gaussian derivative phase pattern image and its block-wise histograms are concatenated to form a single vector referred as BGDPPH descriptor. Dimension reduction is performed by increasing discrimination between genuine and impostor scores using chi-RBF kernel discriminant analysis (KDA). Weighted score level fusion of spectral palmprints on Fisher criterion improves recognition rate. Robustness of the proposed BGDPPH descriptor against blur and noise is evaluated on four gray-scale and two multispectral palmprint databases collected through touch-based and touch-less acquisition devices.

Multimodal biometrics: Weighted score level fusion based on non-ideal iris and face images

The iris and face are among the most promising biometric traits that can accurately identify a person because their unique textures can be swiftly extracted during the recognition process. However, unimodal biometrics have limited usage since no single biometric is sufficiently robust and accurate in real-world applications. Iris and face biometric authentication often deals with non-ideal scenarios such as off-angles, reflections, expression changes, variations in posing, or blurred images. These limitations imposed by unimodal biometrics can be overcome by incorporating multimodal biometrics. Therefore, this paper presents a method that combines face and iris biometric traits with the weighted score level fusion technique to flexibly fuse the matching scores from these two modalities based on their weight availability. The dataset use for the experiment is self established dataset named Universiti Teknologi Malaysia Iris and Face Multimodal Datasets (UTMIFM), UBIRIS version 2.0 (UBIRIS v.2) and ORL face databases. The proposed framework achieve high accuracy, and had a high decidability index which significantly separate the distance between intra and inter distance.