Privacy preserving Back-propagation neural network learning over arbitrarily partitioned data

Neural networks have been an active research area for decades. However, privacy bothers many when the training dataset for the neural networks is distributed between two parties, which is quite common nowadays. Existing cryptographic approaches such as secure scalar product protocol provide a secure way for neural network learning when the training dataset is vertically partitioned. In this paper, we present a privacy preserving algorithm for the neural network learning when the dataset is arbitrarily partitioned between the two parties. We show that our algorithm is very secure and leaks no knowledge (except the final weights learned by both parties) about other party’s data. We demonstrate the efficiency of our algorithm by experiments on real world data.     

Mobility-Aware RSVP: A framework for improving the performance of multimedia services over wireless IP-based mobile networks

This paper proposes the Mobility-Aware Resource Reservation Protocol (MARSVP) in which mobility and QoS signaling are performed as a single functional block. The key concept of MARSVP is to convey mobility-specific information (binding updates and their associated acknowledgments) by using newly defined RSVP objects embedded in existing RSVP messages. An appealing feature of MARSVP is that it adheres to the current RSVP standard (RFC 2205) and thus requires minimal changes to end nodes without affecting any of the conventional RSVP routers in between. The proposed mechanism is evaluated using a simulation model for application-level performance and an analytical model for network-level signaling cost. Simulation results indicate a 27.9% improvement in QoS interruption when using Mobile IPv6 (MIPv6), 12.5% when using Hierarchical Mobile IPv6 (HMIPv6), and no improvement when using Fast Handovers for MIPv6 (FMIPv6). On the network-level, signaling cost savings of 9.4% and 11.9% are achieved for MIPv6 and HMIPv6, respectively, while FMIPv6 achieves savings of 17.9% when using Voice-over-IP traffic and 26.7% for Video-over-IP traffic. The results of the conducted studies indicate MARSVP’s superiority to conventional RSVP when deployed over wireless networks.     

SVM based robust watermarking for enhanced medical image security

Medical images are more typical than any other ordinary images, since it stores patient’s information for diagnosis purpose. Such images need more security and confidentiality as total diagnosis depends on it. In telemedicine applications, transmission of medical image via open channel, demands strong security and copyright protection. In our proposed robust watermarking model, a double layer security is introduced to ensure the robustness of embedded data. The embedded data is scrambled using a unique key and then a transform domain based hybrid watermarking technique is used to embed the scrambled data into the transform coefficients of the host image. The data embedding in medical images involves more attention, so that the diagnosis part must not be affected by any modification. Therefore, Support Vector Machine (SVM) is used as a classifier, which classify a medical image into two regions i.e. Non Region of Interest (NROI) and Region of Interest (ROI) to embed watermark data into the NROI part of the medical image, using the proposed embedding algorithm. The objective of the proposed model is to avoid any quality degradation to the medical image. The simulation is performed to measure the Peak Signal to Noise Ratio (PSNR) for imperceptibility and Structural Similarity Index (SSIM) to test the robustness. The experimented result shows, robustness and imperceptibility with SSIM of more than 0.50 and PSNR of more than 35 dB for proposed watermarking model.

     

Linearized motion estimation for articulated planes

In this paper, we describe the explicit application of articulation constraints for estimating the motion of a system of articulated planes. We relate articulations to the relative homography between planes and show that these articulations translate into linearized equality constraints on a linear least-squares system, which can be solved efficiently using a Karush-Kuhn-Tucker system. The articulation constraints can be applied for both gradient-based and feature-based motion estimation algorithms and to illustrate this, we describe a gradient-based motion estimation algorithm for an affine camera and a feature-based motion estimation algorithm for a projective camera that explicitly enforces articulation constraints. We show that explicit application of articulation constraints leads to numerically stable estimates of motion. The simultaneous computation of motion estimates for all of the articulated planes in a scene allows us to handle scene areas where there is limited texture information and areas that leave the field of view. Our results demonstrate the wide applicability of the algorithm in a variety of challenging real-world cases such as human body tracking, motion estimation of rigid, piecewise planar scenes, and motion estimation of triangulated meshes.     

Automated Cellular Modeling and Prediction on a Large Scale

We describe CHAMP (CHurn Analysis, Modeling, andPrediction), an automated system for modeling cellularsubscriber churn that is predicting which customerswill discontinue cellular phone service. We describevarious issues related to developing and deployingthis system including automating data access from aremote data warehouse, preprocessing, featureselection, model validation, and optimization toreflect business tradeoffs. Using data from GTE'sdata warehouse for cellular phone customers, CHAMP iscapable of developing churn models customized byregion for over one hundred GTE cellular phone marketstotaling over 5 million customers. Every month churnfactors are identified for each geographic region andmodels are updated to generate churn scores predictingwho is likely to churn in the short term. Learningmethods such as decision trees and genetic algorithmsare used for feature selection and a cascade neuralnetwork is used for predicting churn scores. Inaddition to producing churn scores, CHAMP alsoproduces qualitative results in the form of rules andcomparison of market trends that are disseminatedthrough a web based interface.     

A novel microfluidic switch for pH control using Chitosan based hydrogels

We report a novel pH-sensitive hydrogel based micro-valve for metered flow that has applications in a laboratory made “Intelligent valving system”. The hydrogel solution was prepared through Chitosan and poly vinyl alcohol in acetic acid and crystallized using gluteraldehyde as the crosslinking agent in the form of thin wafers and it was found to be very sensitive to pH changes. The pore structure of hydrogel was investigated through Field Emission Scanning Electron Microscopy and thin wafers of the gel were physically placed inside PDMS microchannels. Flow metering in these channels was observed by controlled expansion of the hydrogel plug till complete valving was realized. This valving device was further precisely characterized with micro Particle Image Velocimetry using a solution containing fluorescent polymeric micro beads. The principle advantage of this hydrogel device is the smaller range of pH (varying between pH 3 and 7) over which the valving response is observed.     

Non‐dominated sorting particle swarm optimization (NSPSO) and network security policy enforcement for Policy Space Analysis

Network operators depend on security services with the aim of safeguarding their IT infrastructure. Various types of network security policies are employed on a global scale and are disseminated among several security middleboxes implemented in networks. But, owing to the complications in security policies, it is not quite efficient to directly use the path‐wise enforcement schemes that are prevalent. The major motivation of this work is to improve security levels and solve the policy enforcement problem. For the first time, this work reports the issue of policy enforcement on middleboxes. The major contribution of this work is to design security policy enforcement as a Weighted K Set Covering Problem, and we designed a Policy Space Analysis (PSA) tool intended for a group of operations in the security policy. This PSA tool was developed based on range‐signified hyper‐rectangles, which are indexed by the Hilbert R‐tree. Leveraging the PSA, we first investigated the topological features of various kinds of policies. Balancing the PSA tool in a non‐dominated sorting particle swarm optimization technique exposes the intrinsic difficulties of this security strategy and provides guidance for designing the enforcement approach. In addition, in this research, a new fuzzy rule‐based classification system is introduced for packet classification. A scope‐wise policy enforcement algorithm was proposed, which chooses a moderate number of enforcement network nodes for deploying multiple policy subsets in a greedy manner. This scheme is much quicker compared with the first one and therefore has found its application in real‐time deployments.     

Vision-based action recognition of earthmoving equipment using spatio-temporal features and support vector machine classifiers

Video recordings of earthmoving construction operations provide understandable data that can be used for benchmarking and analyzing their performance. These recordings further support project managers to take corrective actions on performance deviations and in turn improve operational efficiency. Despite these benefits, manual stopwatch studies of previously recorded videos can be labor-intensive, may suffer from biases of the observers, and are impractical after substantial period of observations. This paper presents a new computer vision based algorithm for recognizing single actions of earthmoving construction equipment. This is particularly a challenging task as equipment can be partially occluded in site video streams and usually come in wide variety of sizes and appearances. The scale and pose of the equipment actions can also significantly vary based on the camera configurations. In the proposed method, a video is initially represented as a collection of spatio-temporal visual features by extracting space–time interest points and describing each feature with a Histogram of Oriented Gradients (HOG). The algorithm automatically learns the distributions of the spatio-temporal features and action categories using a multi-class Support Vector Machine (SVM) classifier. This strategy handles noisy feature points arisen from typical dynamic backgrounds. Given a video sequence captured from a fixed camera, the multi-class SVM classifier recognizes and localizes equipment actions. For the purpose of evaluation, a new video dataset is introduced which contains 859 sequences from excavator and truck actions. This dataset contains large variations of equipment pose and scale, and has varied backgrounds and levels of occlusion. The experimental results with average accuracies of 86.33% and 98.33% show that our supervised method outperforms previous algorithms for excavator and truck action recognition. The results hold the promise for applicability of the proposed method for construction activity analysis.