Sharifah Mumtazah Syed Ahmad

Integration of multiple soft biometrics for human identification

A new local feature descriptor is proposed for facial shape representation.The performance on still and surveillance face datasets is comparable to the state of the arts.We present experimental findings on integration of face and body based soft biometrics.Five score fusion techniques are examined to determine the most reliable method.Fuzzy logic is discovered as the most effective score fusion. We propose a computational approach to human identification based on the integration of face and body related soft biometric traits. In previous studies on soft biometrics, several methods for human identification using semantic descriptions have been introduced. Though the results attained exhibit the effectiveness of such techniques in image retrieval and short term tracking of subjects, semantics literally limits the ability of a biometric system to provide conclusive identification. This paper presents a new framework for biometric identification based solely on multiple measured soft biometric traits. The paper describes techniques for extracting/estimating face and body based soft biometric traits from frame set. Furthermore, we utilized a sequential attribute combination method to perform attribute selection prior to integration at match score level. Finally, an evaluation of five score fusion techniques is performed. The results show that the proposed framework can be utilized to model an adequate soft biometric system with rank-1 identification rate of 88%. Display Omitted

Online Handwritten Signature Verification Using Neural Network Classifier Based on Principal Component Analysis

One of the main difficulties in designing online signature verification (OSV) system is to find the most distinctive features with high discriminating capabilities for the verification, particularly, with regard to the high variability which is inherent in genuine handwritten signatures, coupled with the possibility of skilled forgeries having close resemblance to the original counterparts. In this paper, we proposed a systematic approach to online signature verification through the use of multilayer perceptron (MLP) on a subset of principal component analysis (PCA) features. The proposed approach illustrates a feature selection technique on the usually discarded information from PCA computation, which can be significant in attaining reduced error rates. The experiment is performed using 4000 signature samples from SIGMA database, which yielded a false acceptance rate (FAR) of 7.4% and a false rejection rate (FRR) of 6.4%.

Assessment of predictive models for chlorophyll-a concentration of a tropical lake

BackgroundThis study assesses four predictive ecological models; Fuzzy Logic (FL), Recurrent Artificial Neural Network (RANN), Hybrid Evolutionary Algorithm (HEA) and multiple linear regressions (MLR) to forecast chlorophyll- a concentration using limnological data from 2001 through 2004 of unstratified shallow, oligotrophic to mesotrophic tropical Putrajaya Lake (Malaysia). Performances of the models are assessed using Root Mean Square Error (RMSE), correlation coefficient (r), and Area under the Receiving Operating Characteristic (ROC) curve (AUC). Chlorophyll-a have been used to estimate algal biomass in aquatic ecosystem as it is common in most algae. Algal biomass indicates of the trophic status of a water body. Chlorophyll- a therefore, is an effective indicator for monitoring eutrophication which is a common problem of lakes and reservoirs all over the world. Assessments of these predictive models are necessary towards developing a reliable algorithm to estimate chlorophyll- a concentration for eutrophication management of tropical lakes.ResultsSame data set was used for models development and the data was divided into two sets; training and testing to avoid biasness in results. FL and RANN models were developed using parameters selected through sensitivity analysis. The selected variables were water temperature, pH, dissolved oxygen, ammonia nitrogen, nitrate nitrogen and Secchi depth. Dissolved oxygen, selected through stepwise procedure, was used to develop the MLR model. HEA model used parameters selected using genetic algorithm (GA). The selected parameters were pH, Secchi depth, dissolved oxygen and nitrate nitrogen. RMSE, r, and AUC values for MLR model were (4.60, 0.5, and 0.76), FL model were (4.49, 0.6, and 0.84), RANN model were (4.28, 0.7, and 0.79) and HEA model were (4.27, 0.7, and 0.82) respectively. Performance inconsistencies between four models in terms of performance criteria in this study resulted from the methodology used in measuring the performance. RMSE is based on the level of error of prediction whereas AUC is based on binary classification task.ConclusionsOverall, HEA produced the best performance in terms of RMSE, r, and AUC values. This was followed by FL, RANN, and MLR.

Recent advances in facial soft biometrics

Face as a biometric attribute has been extensively studied over the past few decades. Even though, satisfactory results are already achieved in controlled environments, the practicality of face recognition in realistic scenarios is still limited by several challenges, such as, expression, pose, occlusion, etc. Recently, the research direction is concentrating on the prospects of complementing face recognition systems with facial soft biometric traits. The ease of extracting facial soft biometrics under several varying conditions has mainly resulted in the ability of using the traits to, either improve the performance of traditional face recognition systems, or performing recognition solely based on many facial soft biometrics. This paper presents state-of-the-art techniques in facial soft biometrics research by describing the type of traits, feature extraction methods, and the application domains. It indicates the most recent and valuable results attained, while also highlighting some possible future scientific research directions to be investigated.

Gender recognition on real world faces based on shape representation and neural network

Gender as a soft biometric attribute has been extensively investigated in the domain of computer vision because of its numerous potential application areas. However, studies have shown that gender recognition performance can be hindered by improper alignment of facial images. As a result, previous experiments have adopted face alignment as an important stage in the recognition process, before performing feature extraction. In this paper, the problem of recognizing human gender from unaligned real world faces using single image per individual is investigated. The use of feature descriptor to form shape representation of face images with any arbitrary orientation from the cropped version of Labeled Faces in the Wild (LFW) dataset is proposed. By combining the feature extraction technique with artificial neural network for classification, a recognition rate of 89.3% is attained.

Estimating body related soft biometric traits in video frames

Soft biometrics can be used as a prescreening filter, either by using single trait or by combining several traits to aid the performance of recognition systems in an unobtrusive way. In many practical visual surveillance scenarios, facial information becomes difficult to be effectively constructed due to several varying challenges. However, from distance the visual appearance of an object can be efficiently inferred, thereby providing the possibility of estimating body related information. This paper presents an approach for estimating body related soft biometrics; specifically we propose a new approach based on body measurement and artificial neural network for predicting body weight of subjects and incorporate the existing technique on single view metrology for height estimation in videos with low frame rate. Our evaluation on 1120 frame sets of 80 subjects from a newly compiled dataset shows that the mentioned soft biometric information of human subjects can be adequately predicted from set of frames.

Analysis of the Effects and Relationship of Perceived Handwritten Signature's Size, Graphical Complexity, and Legibility with Dynamic Parameters for Forged and Genuine Samples

This article presents an analysis of handwritten signature dynamics belonging to two authentication groups, namely genuine and forged signature samples. Genuine signatures are initially classified based on their relative size, graphical complexity, and legibility as perceived by human examiners. A pool of dynamic features is then extracted for each signature sample in the two groups. A two‐way analysis of variance (ANOVA) is carried out to investigate the effects and the relationship between the perceived classifications and the authentication groups. Homogeneity of variance was ensured through Bartletts test prior to ANOVA testing. The results demonstrated that among all the investigated dynamic features, pen pressure is the most distinctive which is significantly different for the two authentication groups as well as for the different perceived classifications. In addition, all the relationships investigated, namely authenticity group versus size, graphical complexity, and legibility, were found to be positive for pen pressure.

Online signature verification using neural network and pearson correlation features

In this paper, we proposed a method for feature extraction in online signature verification. We first used signature coordinate points and pen pressure of all signatures, which are available in the SIGMA database. Then, Pearson correlation coefficients were selected for feature extraction. The obtained features were used in back-propagation neural network for verification. The results indicate an accuracy of 82.42%.

Face Liveness Detection Using Dynamic Local Ternary Pattern (DLTP)

Face spoofing is considered to be one of the prominent threats to face recognition systems. However, in order to improve the security measures of such biometric systems against deliberate spoof attacks, liveness detection has received significant recent attention from researchers. For this purpose, analysis of facial skin texture properties becomes more popular because of its limited resource requirement and lower processing cost. The traditional method of skin analysis for liveness detection was to use Local Binary Pattern (LBP) and its variants. LBP descriptors are effective, but they may exhibit certain limitations in near uniform patterns. Thus, in this paper, we demonstrate the effectiveness of Local Ternary Pattern (LTP) as an alternative to LBP. In addition, we adopted Dynamic Local Ternary Pattern (DLTP), which eliminates the manual threshold setting in LTP by using Weber’s law. The proposed method was tested rigorously on four facial spoof databases: three are public domain databases and the other is the Universiti Putra Malaysia (UPM) face spoof database, which was compiled through this study. The results obtained from the proposed DLTP texture descriptor attained optimum accuracy and clearly outperformed the reported LBP and LTP texture descriptors.

The Design and Compilation of a Facial Spoof Database on Various Textures

Face biometrics plays an important role in various authentication applications. However, one of the main design challenges for an accurate face biometrics is detecting spoof artificial artifacts. The development of robust anti-spoofing algorithm requires for rigorous system training and testing on facial database that includes various possible spoof specimen which reflects different variations of spoofing attacks. Currently, the databases which are used to test face anti-spoofing algorithms are limited in terms of texture variations. Therefore, in this paper, our focus is to introduce a face spoofing database which is compiled from various types of spoofing textures. Our database was collected from 30 different subjects and fake faces were recaptured on various forms which include four different types of paper-based textures and three different digital display devices. All images were collected using a high precision camera device. This database would provide a more realistic and challenging platform for facial anti-spoofing research.