Muhammad Atif Tahir

Multiscale Local Phase Quantization for Robust Component-Based Face Recognition Using Kernel Fusion of Multiple Descriptors

Face recognition subject to uncontrolled illumination and blur is challenging. Interestingly, image degradation caused by blurring, often present in real-world imagery, has mostly been overlooked by the face recognition community. Such degradation corrupts face information and affects image alignment, which together negatively impact recognition accuracy. We propose a number of countermeasures designed to achieve system robustness to blurring. First, we propose a novel blur-robust face image descriptor based on Local Phase Quantization (LPQ) and extend it to a multiscale framework (MLPQ) to increase its effectiveness. To maximize the insensitivity to misalignment, the MLPQ descriptor is computed regionally by adopting a component-based framework. Second, the regional features are combined using kernel fusion. Third, the proposed MLPQ representation is combined with the Multiscale Local Binary Pattern (MLBP) descriptor using kernel fusion to increase insensitivity to illumination. Kernel Discriminant Analysis (KDA) of the combined features extracts discriminative information for face recognition. Last, two geometric normalizations are used to generate and combine multiple scores from different face image scales to further enhance the accuracy. The proposed approach has been comprehensively evaluated using the combined Yale and Extended Yale database B (degraded by artificially induced linear motion blur) as well as the FERET, FRGC 2.0, and LFW databases. The combined system is comparable to state-of-the-art approaches using similar system configurations. The reported work provides a new insight into the merits of various face representation and fusion methods, as well as their role in dealing with variable lighting and blur degradation.

Inverse random under sampling for class imbalance problem and its application to multi-label classification

In this paper, a novel inverse random under sampling (IRUS) method is proposed for the class imbalance problem. The main idea is to severely under sample the majority class thus creating a large number of distinct training sets. For each training set we then find a decision boundary which separates the minority class from the majority class. By combining the multiple designs through fusion, we construct a composite boundary between the majority class and the minority class. The proposed methodology is applied on 22 UCI data sets and experimental results indicate a significant increase in performance when compared with many existing class-imbalance learning methods. We also present promising results for multi-label classification, a challenging research problem in many modern applications such as music, text and image categorization.

Towards cloud based big data analytics for smart future cities

A large amount of land-use, environment, socio-economic, energy and transport data is generated in cities. An integrated perspective of managing and analysing such big data can answer a number of science, policy, planning, governance and business questions and support decision making in enabling a smarter environment. This paper presents a theoretical and experimental perspective on the smart cities focused big data management and analysis by proposing a cloud-based analytics service. A prototype has been designed and developed to demonstrate the effectiveness of the analytics service for big data analysis. The prototype has been implemented using Hadoop and Spark and the results are compared. The service analyses the Bristol Open data by identifying correlations between selected urban environment indicators. Experiments are performed using Hadoop and Spark and results are presented in this paper. The data pertaining to quality of life mainly crime and safety & economy and employment was analysed from the data catalogue to measure the indicators spread over years to assess positive and negative trends.

A Multiple Expert Approach to the Class Imbalance Problem Using Inverse Random under Sampling

In this paper, a novel inverse random under sampling (IRUS) method is proposed for class imbalance problem. The main idea is to severely under sample the negative class (majority class), thus creating a large number of distinct negative training sets. For each training set we then find a linear discriminant which separates the positive class from the negative class. By combining the multiple designs through voting, we construct a composite between the positive class and the negative class. The proposed methodology is applied on 11 UCI data sets and experimental results indicate a significant increase in Area Under Curve (AUC) when compared with many existing class-imbalance learning methods.

Off-line writer identification using an ensemble of grapheme codebook features

A novel approach is proposed for off-line writer identification.The proposed approach utilizes an ensemble of codebook grapheme features.Kernel discriminant analysis is employed for dimensionality reduction.Experiments are conducted using publicly available writer identification data sets.The proposed technique provides a very accurate and efficient solution. Off-line writer identification is the process of matching a handwritten sample with its author. Manual identification is very time-consuming because it requires a meticulous comparison of character shape details. Consequently the automation of writer identification has become an important area of research interest. The codebook (or bag of features) approach is a state-of-the-art computerized technique for writer identification. One way to achieve a high identification rate is to expose the personalized set of character shapes, or allographs, that a writer has adopted over the years. The main problem associated with this approach is the extremely large of number of points of interest that are generated. In this paper we extend the basic model to include an ensemble of codebooks. Additionally, Kernel discriminant analysis using spectral regression (SR-KDA) is used as a dimensionality reduction technique in order to avoid over-fitting. Fusion of multiple codebooks is shown to increase the identification rate by 11% compared with a single codebook approach.

Visual category recognition using Spectral Regression and Kernel Discriminant Analysis

Visual category recognition (VCR) is one of the most important tasks in image and video indexing. Spectral methods have recently emerged as a powerful tool for dimensionality reduction and manifold learning. Recently, Spectral Regression combined with Kernel Discriminant Analysis (SR-KDA) has been successful in many classification problems. In this paper, we adopt this solution to VCR and demonstrate its advantages over existing methods both in terms of speed and accuracy. The distinctiveness of this method is assessed experimentally using an image and a video benchmark: the PASCAL VOC Challenge 08 and the Mediamill Challenge. From the experimental results, it can be derived that SR-KDA consistently yields significant performance gains when compared with the state-of-the art methods. The other strong point of using SR-KDA is that the time complexity scales linearly with respect to the number of concepts and the main computational complexity is independent of the number of categories.

Improving multilabel classification performance by using ensemble of multi-label classifiers

Multilabel classification is a challenging research problem in which each instance is assigned to a subset of labels. Recently, a considerable amount of research has been concerned with the development of “good” multi-label learning methods. Despite the extensive research effort, many scientific challenges posed by e.g. highly imbalanced training sets and correlation among labels remain to be addressed. The aim of this paper is use heterogeneous ensemble of multi-label learners to simultaneously tackle both imbalance and correlation problems. This is different from the existing work in the sense that the later mainly focuses on ensemble techniques within a multi-label learner while we are proposing in this paper to combine these state-of-the-art multi-label methods by ensemble techniques. The proposed ensemble approach (EML) is applied to three publicly available multi-label data sets using several evaluation criteria. We validate the advocated approach experimentally and demonstrate that it yields significant performance gains when compared with state-of-the art multi-label methods.

Combining multiple kernels by augmenting the kernel matrix

In this paper we present a novel approach to combining multiple kernels where the kernels are computed from different information channels. In contrast to traditional methods that learn a linear combination of n kernels of size m ×m, resulting in m coefficients in the trained classifier, we propose a method that can learn n ×m coefficients. This allows to assign different importance to the information channel per example rather than per kernel. We analyse the proposed kernel combination in empirical feature space and provide its geometrical interpretation. We validate the approach on both UCI datasets and an object recognition dataset, and demonstrate that it leads to classification improvements.

A Comparison of L_1 Norm and L_2 Norm Multiple Kernel SVMs in Image and Video Classification

SVM is one of the state-of-the-art techniques for image and video classification. When multiple kernels are available, the recently introduced multiple kernel SVM (MK-SVM) learns an optimal linear combination of the kernels, providing a new method for information fusion. In this paper we study how the behaviour of MK-SVM is affected by the norm used to regularise the kernel weights to be learnt. Through experiments on three image/video classification datasets as well as on synthesised data, new insights are gained as to how the choice of regularisation norm should be made, especially when MK-SVM is applied to image/video classification problems.

Face recognition using multi-scale local phase quantisation and Linear Regression Classifier

Linear Regression Classifier (LRC) is state-of-the-art face recognition method that represent a probe image as a linear combination of class specific models. However, this method views the image as a point in a feature space, and thus LRC cannot accommodate severe luminance alterations. Histogram-based features, such as Multiscale Local Phase Quantisation histogram (MLPQH) have gained reputation as powerful and attractive texture descriptors showing excellent results in terms of accuracy and computational complexity in face recognition. In this paper, MLPQH features are integrated with “face” features to confront the illumination problem in LRC. The main novelty is the fusion of histogram and face features using z-score normalisation and LRC classifier. The proposed system is evaluated on two benchmarks: ORL and Extended Yale B. The results indicate a significant increase in the performance when compared with state-of-the-art face recognition methods.