We propose a novel approach for content based color image classification using Support Vector Machine (SVM). Traditional classification approaches deal poorly on content based image classification tasks being one of the reasons of high dimensionality of the feature space. In this paper, color image classification is done on features extracted from histograms of color components. The benefit of using color image histograms are better efficiency, and insensitivity to small changes in camera view-point i.e. translation and rotation. As a case study for validation purpose, experimental trials were done on a database of about 500 images divided into four different classes has been reported and compared on histogram features for RGB, CMYK, Lab, YUV, YCBCR, HSV, HVC and YIQ color spaces. Results based on the proposed approach are found encouraging in terms of color image classification accuracy.
Recommender systems based on collaborative filtering are highly vulnerable to data poisoning attacks, where a determined attacker injects fake users with false user-item feedback, with an objective to either corrupt the recommender system or promote/demote a target set of items. Recently, differential privacy was explored as a defense technique against data poisoning attacks in the typical machine learning setting. In this paper, we study the effectiveness of differential privacy against such attacks on matrix factorization based collaborative filtering systems. Concretely, we conduct extensive experiments for evaluating robustness to injection of malicious user profiles by simulating common types of shilling attacks on real-world data and comparing the predictions of typical matrix factorization with differentially private matrix factorization.
Mapping Intellectual Property (IP) cores onto a Networkon-Chip (NoC) architecture is an important phase of NoC design and the performance and energy consumption of the chip are the major issues that affect the design. In this paper, we analyze the preexistent mapping algorithms and present a new efficient energy and bandwidth aware topological mapping of IPs onto regular tile-based NoC architecture. The proposed algorithm has been implemented and evaluated for randomly generated benchmarks as well as reallife applications like Video Object Plane Decoder (VOPD) and Telecom. The experimental results have also been compared with existing mapping algorithms for the same set of benchmarks which clearly demonstrate significant reduction in maximum allocated bandwidth and energy for future NoC architectures with large number of IP cores. Further, there is a significant reduction in execution time of the proposed algorithm as compared to the other techniques.
The Walker circulation is linked to extratropical waves that are deflected from the Northern Hemisphere polar regions and travel southeastward over central Asia toward the western Pacific warm pool during northern winter. The wave pattern resembles the east Atlantic-west Russia pattern and influences the El Niño-Southern Oscillation (ENSO) region. A tripole pattern between the West Siberian Plain and the two centers of action of ENSO indicates that the background state of ENSO with respect to global sea level pressure (SLP) has a significant negative correlation to the West Siberian Plain. The correlation with the background state, which is defined by the sum of the two centers of action of ENSO, is higher than each of the pairwise correlations with either of the ENSO centers alone. The centers are defined with a clustering algorithm that detects regions with similar characteristics. The normalized monthly SLP time series for the two centers of ENSO (around Darwin, Australia, and Tahiti) are area averaged, and the sum of both regions is considered as the background state of ENSO. This wave train can be detected throughout the troposphere and the lower stratosphere. Its origins can be traced back to Rossby wave activity triggered by convection over the subtropical North Atlantic that emanates wave activity toward the West Siberian Plain. The same wave train also propagates to the central Pacific Ocean around Tahiti and can be used to predict the background state over the ENSO region. This background state also modifies the subtropical bridge between tropical eastern Pacific and subtropical North Atlantic leading to a circumglobal wave train.
This paper deals with the performance of vertically reinforced 1-3 piezoelectric composite material as a constraining layer for active constrained layer damping of sandwich beams. A finite element model, considering both in-plane and out-of-plane actuation of the constraining layer of the active constrained layer damping, has been developed for analyzing the active damping of sandwich beams integrated with the patches of such active constrained layer damping treatment. The analysis revealed that vertically reinforced 1-3 piezo composites, which are popularly used as sensor materials, can be used as distributed actuators of smart sandwich beam.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.