Face expression recognition system based on ripplet transform type II and least square SVM

Multimedia Tools and Applications - This paper discusses the development of an efficient and automated system for the recognition of facial expressions, which is essentially an application...     

Fast periocular authentication in handheld devices with reduced phase intensive local pattern

To ensure highest security in handheld devices, biometric authentication has emerged as a reliable methodology. Deployment of mobile biometric authentication struggles due to computational complexity. For a fast response from a mobile biometric authentication method, it is desired that the feature extraction and matching should take least time. In this article, the periocular region captured through frontal camera of a mobile device is considered under investigation for its suitability to produce a reduced feature that takes least time for feature extraction and matching. A recently developed feature Phase Intensive Local Pattern (PILP) is subjected to reduction giving birth to a feature termed as Reduced PILP (R-PILP), which yields a matching time speed-up of 1.56 times while the vector is 20% reduced without much loss in authentication accuracy. The same is supported by experiment on four publicly available databases. The performance is also compared with one global feature: Phase Intensive Global Pattern, and three local features: Scale Invariant Feature Transform, Speeded-up Robust Features, and PILP. The amount of reduction can be varied with the requirement of the system. The amount of reduction and the performance of the system bears a trade-off. Proposed R-PILP attempts to make periocular suitable for mobile devices.     

A new healthcare diagnosis system using an IoT-based fuzzy classifier with FPGA

The Journal of Supercomputing - In this study, an Internet of Things (IoT)-based analysis system is proposed, which can be used to design a consumer electronic device. This IoT-based analysis...     

Illumination and scale invariant relevant visual features with hypergraph-based learning for multi-shot person re-identification

Person re-identification which aims at matching people across disjoint cameras has received increasing attention due to the widespread use of video surveillance applications. Existing methods concentrate either on robust feature extraction or view-invariant feature transformation. However, the extracted features suffer from various limitations such as color inconsistency and scale variations. Besides, during matching, a probe is compared against each gallery instance which represents only the pairwise relationship and ignores the high order relationship among them. To address these issues, we propose a multi-shot person re-identification framework that first performs a preprocessing task on images to address illumination variations for maintaining the color consistency. Subsequently, we formulate an approach to handle scale variations in the pedestrian appearances for keeping them with relatively a fixed scale ratio. Overlapped visual patches representing appearance cues are then extracted from the processed images. A structured multi-class feature selection approach is employed to select a set of relevant patches that simultaneously discriminates all distinct persons. These selected patches use a hypergraph to represent the visual association among a probe and gallery images. Finally, for matching, we formulate a hypergraph-based learning scheme, which considers both the pairwise and high-order association among the probe and gallery images. The hypergraph structure is then optimized to yield an improved similarity score for a probe against each gallery instance. The effectiveness of our proposed framework is validated on three public datasets and comparison with state-of-the-art methods shows the superior performance of our framework.

     

Effect of particle size distribution on rheology of high concentration limestone–water slurry for economic pipeline transportation

The preparation of high concentration slurry requires careful selection of particle size distribution to achieve the required rheological properties for economic and efficient pipeline transportation. In the present study, the maximum static settled concentration (C_W-max) tests pertaining to limiting achievable concentration and rheological measurements were performed for limestone samples in the slurry concentration range of 60–78.5% by mass. The limestone samples with four distinct particle size ranges; i.e., <38, 38–90, 90–210, 210–300?µm were used to prepare five representative experimental samples by blending the fines (<38?µm) with other three coarse size ranges in definite mass proportions. The rheological behavior of the five limestone slurry samples with mono-modal, bimodal and multimodal packing characteristics indicated non-Newtonian flow behavior and fitted well to Bingham Plastic model in the concentration range of 60–78.5% by mass. The slurry samples with bimodal and multimodal packing characteristics indicated substantial reduction in slurry viscosity, yield stress and improved solids loading as compared to mono-modal one. The higher C_W-max values obtained for the specific blended (fines with coarse) limestone slurry samples were attributed to the packing effect and was correlated to the ratio of surface to surface separation for the coarse particles (β) to the average fine particle size (d_50-f) to achieve higher solids concentration. The reduction in slurry viscosity observed for the specific limestone samples was further substantiated by correlating the distribution modulus (ψ) derived from Farris theory. It may be inferred that these theoretical treatments correlating the experimental data can provide highly reliable guidance to the preparation of high concentration limestone slurry for economic pipeline transportation.     

Single image super resolution for texture images through neighbor embedding

This article proposes an improved learning based super resolution scheme using manifold learning for texture images. Pseudo Zernike moment (PZM) has been employed to extract features from the texture images. In order to efficiently retrieve similar patches from the training patches, feature similarity index matrix (FSIM) has been used. Subsequently, for reconstruction of the high resolution (HR) patch, a collaborative optimal weight is generated from the least square (LS) and non-negative matrix factorization (NMF) methods. The proposed method is tested on some color texture, gray texture, and some standard images. Results of the proposed method on texture images advocate its superior performance over established state-of-the-art methods.

     

Autonomic cloud resource provisioning and scheduling using meta-heuristic algorithm

We investigate that resource provisioning and scheduling is a prominent problem due to heterogeneity as well as dispersion of cloud resources. Cloud service providers are building more and more datacenters due to demand of high computational power which is a serious threat to environment in terms of energy requirement. To overcome these issues, we need an efficient meta-heuristic technique that allocates applications among the virtual machines fairly and optimizes the quality of services (QoS) parameters to meet the end user objectives. Binary particle swarm optimization (BPSO) is used to solve real-world discrete optimization problems but simple BPSO does not provide optimal solution due to improper behavior of transfer function. To overcome this problem, we have modified transfer function of binary PSO that provides exploration and exploitation capability in better way and optimize various QoS parameters such as makespan time, energy consumption, and execution cost. The computational results demonstrate that modified transfer function-based BPSO algorithm is more efficient and outperform in comparison with other baseline algorithm over various synthetic datasets.

     

Deceptive review detection using labeled and unlabeled data

Availability of millions of products and services on e-commerce sites makes it difficult to search the best suitable product according to the requirements because of existence of many alternatives. To get rid of this the most popular and useful approach is to follow reviews of others in opinionated social medias, who have already tried them. Almost all e-commerce sites provide facility to the users for giving views and experience of the product and services they experienced. The customers reviews are increasingly used by individuals, manufacturers and retailers for purchase and business decisions. As there is no scrutiny over the reviews received, anybody can write anything unanimously which conclusively leads to review spam. Moreover, driven by the desire of profit and/or publicity, spammers produce synthesized reviews to promote some products/brand and demote competitors products/brand. Deceptive review spam has seen a considerable growth overtime. In this work, we have applied supervised as well as unsupervised techniques to identify review spam. Most effective feature sets have been assembled for model building. Sentiment analysis has also been incorporated in the detection process. In order to get best performance some well-known classifiers were applied on labeled dataset. Further, for the unlabeled data, clustering is used after desired attributes were computed for spam detection. Additionally, there is a high chance that spam reviewers may also be held responsible for content pollution in multimedia social networks, because nowadays many users are giving the reviews using their social network logins. Finally, the work can be extended to find suspicious accounts responsible for posting fake multimedia contents into respective social networks.     

Measurement of probe displacement to the thermal resolution limit in photonic force microscopy using a miniature quadrant photodetector

A photonic force microscope comprises of an optically trapped micro-probe and a position detection system to track the motion of the probe. Signal collection for motion detection is often carried out using the backscattered light off the probe-however, this mode has problems of low S/N due to the small backscattering cross sections of the micro-probes typically used. The position sensors often used in these cases are quadrant photodetectors. To ensure maximum sensitivity of such detectors, it would help if the detector size matched with the detection beam radius after the condenser lens (which for backscattered detection would be the trapping objective itself). To suit this condition, we have used a miniature displacement sensor whose dimensions makes it ideal to work with 1:1 images of micrometer-sized trapped probes in the backscattering detection mode. The detector is based on the quadrant photo-integrated chip in the optical pick-up head of a compact disc player. Using this detector, we measured absolute displacements of an optically trapped 1.1 μm probe with a resolution of ～10 nm for a bandwidth of 10 Hz at 95% significance without any sample or laser stabilization. We characterized our optical trap for different sized probes by measuring the power spectrum for each probe to 1% accuracy, and found that for 1.1 μm diameter probes, the noise in our position measurement matched the thermal resolution limit for averaging times up to 10 ms. We also achieved a linear response range of around 385 nm with cross talk between axes ?4% for 1.1 μm diameter probes. The detector has extremely high bandwidth (few MHz) and low optical power threshold-other factors that can lead to its widespread use in photonic force microscopy.