Amirkabir University of Technology

About: Amirkabir University of Technology is a education organization based out in Tehran, Iran. It is known for research contribution in the topics: Nonlinear system & Finite element method. The organization has 15254 authors who have published 31165 publications receiving 487551 citations. The organization is also known as: Tehran Polytechnic & Tehran Polytechnic University.

Adversarially Learned One-Class Classifier for Novelty Detection

Abstract: Novelty detection is the process of identifying the observation(s) that differ in some respect from the training observations (the target class). In reality, the novelty class is often absent during training, poorly sampled or not well defined. Therefore, one-class classifiers can efficiently model such problems. However, due to the unavailability of data from the novelty class, training an end-to-end deep network is a cumbersome task. In this paper, inspired by the success of generative adversarial networks for training deep models in unsupervised and semi-supervised settings, we propose an end-to-end architecture for one-class classification. Our architecture is composed of two deep networks, each of which trained by competing with each other while collaborating to understand the underlying concept in the target class, and then classify the testing samples. One network works as the novelty detector, while the other supports it by enhancing the inlier samples and distorting the outliers. The intuition is that the separability of the enhanced inliers and distorted outliers is much better than deciding on the original samples. The proposed framework applies to different related applications of anomaly and outlier detection in images and videos. The results on MNIST and Caltech-256 image datasets, along with the challenging UCSD Ped2 dataset for video anomaly detection illustrate that our proposed method learns the target class effectively and is superior to the baseline and state-of-the-art methods.

A New Multilevel Converter Topology With Reduced Number of Power Electronic Components

Abstract: In this paper, a new topology for cascaded multilevel converter based on submultilevel converter units and full-bridge converters is proposed. The proposed topology significantly reduces the number of dc voltage sources, switches, IGBTs, and power diodes as the number of output voltage levels increases. Also, an algorithm to determine dc voltage sources magnitudes is proposed. To synthesize maximum levels at the output voltage, the proposed topology is optimized for various objectives, such as the minimization of the number of switches, gate driver circuits and capacitors, and blocking voltage on switches. The analytical analyses of the power losses of the proposed converter are also presented. The operation and performance of the proposed multilevel converter have been evaluated with the experimental results of a single-phase 125-level prototype converter.

Modification of polysiloxane polymers for biomedical applications: a review

Abstract: This paper reviews methods of modifying polydimethylsiloxane (PDMS) polymers to improve their properties for biomedical applications. The modification methods are discussed under three different categories: bulk, surface and other modification techniques. Surface modification techniques include physical and chemical techniques to modify polymer surfaces. Bulk modification techniques include blending, copolymerization, interpenetrating polymer networks (IPNs) and functionalization. The third category includes less common modification techniques. © 2001 Society of Chemical Industry

EMG feature evaluation for movement control of upper extremity prostheses

Abstract: A variety of EMG features have been evaluated for control of myoelectric upper extremity prostheses. Movement class discrimination, robustness, and computational complexity of these features have been investigated for different time window sizes and noise levels. The measurements include novel application of the Davies-Bouldin index, a measure of cluster separability, and the K-nearest neighbor nonparametric classifier. The features evaluated are the integral of average value, the variance, the number of zero crossings, the Willison amplitude, the v-order and log detectors, and autoregressive model parameters. A new feature, the EMG Histogram, is introduced and shown to be the most effective of the group. The experiments were done on the data acquired from the residual biceps and triceps muscle of an above-elbow amputee.