Browsing by Subject "Computationally efficient"

Now showing 1 - 13 of 13

Open Access
Additive neural network for forest fire detection
(Springer, 2020) Pan, H.; Badawi, D.; Zhang, X.; Çetin, Ahmet Enis
In this paper, we introduce a video-based wildfire detection scheme based on a computationally efficient additive deep neural network, which we call AddNet. This AddNet is based on a multiplication-free vector operator, which performs only addition and sign manipulation operations. In this regard, we construct a dot product-like operation from the mf-operator and use it to define dense and convolutional feed-forwarding passes in AddNet. We train AddNet on images taken from forestry surveillance cameras. Our experiments show that AddNet can achieve a time-saving by 12.4% when compared to an equivalent regular convolutional neural network (CNN). Furthermore, the smoke recognition performance of AddNet is as good as regular CNNs and substantially better than binary-weight neural networks.
Open Access
Blind phase noise estimation in OFDM systems by sequential Monte Carlo method
(Springer, 2006) Panayırcı, Erdal; Çırpan, H. A.; Moeneclaey, M.; Noels, N.
In this paper, based on a sequential Monte Carlo method, a computationally efficient algorithm is presented for estimating the residual phase noise, blindly, generated at the output the phase tracking loop employed in OFDM systems. The basic idea is to treat the transmitted symbols as "missing data" and draw samples sequentially of them based on the observed signal samples up to time t. This way, the Bayesian estimates of the phase noise is obtained through these samples, sequentially drawn, together with their importance weights. The proposed receiver structure is seen to be ideally suited for high-speed parallel implementation using VLSI technology.
Open Access
Camera tamper detection using wavelet analysis for video surveillance
(IEEE, 2007-09) Aksay, A.; Temizel, A.; Çetin, A. Enis
It is generally accepted that video surveillance system operators lose their concentration after a short period of time and may miss important events taking place. In addition, many surveillance systems are frequently left unattended. Because of these reasons, automated analysis of the live video feed and automatic detection of suspicious activity have recently gained importance. To prevent capture of their images, criminals resort to several techniques such as deliberately obscuring the camera view, covering the lens with a foreign object, spraying or defocusing the camera lens. In this paper, we propose some computationally efficient wavelet domain methods for rapid camera tamper detection and identify some real-life problems and propose solutions to these. © 2007 IEEE.
Open Access
Co-difference based object tracking algorithm for infrared videos
(IEEE, 2016) Demir, H. Seçkin; Çetin, A. Enis
This paper presents a novel infrared (IR) object tracking algorithm based on the co-difference matrix. Extraction of co-difference features is similar to the well known covariance method except that the vector product operator is redefined in a multiplication-free manner. The new operator yields a computationally efficient implementation for real time object tracking applications. Experiments on an extensive set of IR image sequences indicate that the new method performs better than covariance tracking and other tracking algorithms without requiring any multiplication operations.
Open Access
A computational homogenization framework for soft elastohydrodynamic lubrication
(Springer, 2012) Budt, M.; Temizer, İlker; Wriggers, P.
The interaction between microscopically rough surfaces and hydrodynamic thin film lubrication is investigated under the assumption of finite deformations. Within a coupled micro-macro analysis setting, the influence of roughness onto the macroscopic scale is determined using FE 2-type homogenization techniques to reduce the overall computational cost. Exact to within a separation of scales assumption, a computationally efficient two-phase micromechanical test is proposed to identify the macroscopic interface fluid flux from a lubrication analysis performed on the deformed configuration of a representative surface element. Parameter studies show a strong influence of both roughness and surface deformation on the macroscopic response for isotropic and anisotropic surfacial microstructures.
Open Access
Covariance matrix-based fire and flame detection method in video
(Springer, 2011-09-17) Habiboğlu, Y. H.; Günay, O.; Çetin, A. Enis
This paper proposes a video-based fire detection system which uses color, spatial and temporal information. The system divides the video into spatio-temporal blocks and uses covariance-based features extracted from these blocks to detect fire. Feature vectors take advantage of both the spatial and the temporal characteristics of flame-colored regions. The extracted features are trained and tested using a support vector machine (SVM) classifier. The system does not use a background subtraction method to segment moving regions and can be used, to some extent, with non-stationary cameras. The computationally efficient method can process 320×240 video frames at around 20 frames per second in an ordinary PC with a dual core 2.2 GHz processor. In addition, it is shown to outperform a previous method in terms of detection performance.
Open Access
Design and analysis of a modular learning based cross-coupled control algorithm for multi-axis precision positioning systems
(Institute of Control, Robotics and Systems, 2016) Ulu, N. G.; Ulu E.; Cakmakci, M.
Increasing demand for micro/nano-technology related equipment resulted in growing interest for precision positioning systems. In this paper a modular controller combining cross-coupled control and iterative learning control approaches to improve contour and tracking accuracy at the same time is presented. Instead of using the standard error estimation technique, a computationally efficient and modular contour error estimation technique is used. The new controller is more suitable for tracking arbitrary nonlinear contours and easier to implement to multi-axis systems. Stability and convergence analysis for the proposed controller is presented with the necessary conditions. Effectiveness of the control design is verified with simulations and experiments on a two-axis positioning system. The resulting positioning system achieves nanometer level contouring and tracking performance. © 2016, Institute of Control, Robotics and Systems and The Korean Institute of Electrical Engineers and Springer-Verlag Berlin Heidelberg.
Open Access
An empirical eigenvalue-threshold test for sparsity level estimation from compressed measurements
(IEEE, 2014) Lavrenko, A.; Römer, F.; Del Galdo, G.; Thoma, R.; Arıkan, Orhan
Compressed sensing allows for a significant reduction of the number of measurements when the signal of interest is of a sparse nature. Most computationally efficient algorithms for signal recovery rely on some knowledge of the sparsity level, i.e., the number of non-zero elements. However, the sparsity level is often not known a priori and can even vary with time. In this contribution we show that it is possible to estimate the sparsity level directly in the compressed domain, provided that multiple independent observations are available. In fact, one can use classical model order selection algorithms for this purpose. Nevertheless, due to the influence of the measurement process they may not perform satisfactorily in the compressed sensing setup. To overcome this drawback, we propose an approach which exploits the empirical distributions of the noise eigenvalues. We demonstrate its superior performance compared to state-of-the-art model order estimation algorithms numerically.
Open Access
Learning based cross-coupled control for multi-axis high precision positioning systems
(ASME, 2012-10) Geçer-Ulu, Nurcan; Ulu, Erva; Çakmakçı, Melih
In this paper, a controller featuring cross-coupled control and iterative learning control schemes is designed and implemented on a modular two-axis positioning system in order to improve both contour and tracking accuracy. Instead of using the standard contour estimation technique proposed with the variable gain cross-coupled control, a computationally efficient contour estimation technique is incorporated with the presented control design. Moreover, implemented contour estimation technique makes the presented control scheme more suitable for arbitrary nonlinear contours. Effectiveness of the control design is verified with simulations and experiments on a two-axis positioning system. Also, simulations demonstrating the performance of the control method on a three-axis positioning system are provided. The resulting controller is shown to achieve nanometer level contouring and tracking performance. Simulation results also show its applicability to three-axis nano-positioning systems. Copyright © 2012 by ASME.
Open Access
Moving region detection in compressed video
(Springer, 2004) Töreyin, B. U.; Çetin, A. Enis; Aksay, A.; Akhan, M. B.
In this paper, an algorithm for moving region detection in compressed video is developed. It is assumed that the video can be compressed either using the Discrete Cosine Transform (DOT) or the Wavelet Transform (WT). The method estimates the WT of the background scene from the WTs of the past image frames of the video. The WT of the current image is compared with the WT of the background and the moving objects are determined from the difference. The algorithm does not perform inverse WT to obtain the actual pixels of the current image nor the estimated background. In the case of DOT compressed video, the DC values of 8 by 8 image blocks of Y, U and V channels are used for estimating the background scene. This leads to a computationally efficient method and a system compared to the existing motion detection methods. © Springer-Verlag 2004.
Open Access
A multiplication-free framework for signal processing and applications in biomedical image analysis
(IEEE, 2013) Suhre, A.; Keskin F.; Ersahin, T.; Cetin-Atalay, R.; Ansari, R.; Cetin, A.E.
A new framework for signal processing is introduced based on a novel vector product definition that permits a multiplier-free implementation. First a new product of two real numbers is defined as the sum of their absolute values, with the sign determined by product of the hard-limited numbers. This new product of real numbers is used to define a similar product of vectors in RN. The new vector product of two identical vectors reduces to a scaled version of the l1 norm of the vector. The main advantage of this framework is that it yields multiplication-free computationally efficient algorithms for performing some important tasks in signal processing. An application to the problem of cancer cell line image classification is presented that uses the notion of a co-difference matrix that is analogous to a covariance matrix except that the vector products are based on our new proposed framework. Results show the effectiveness of this approach when the proposed co-difference matrix is compared with a covariance matrix. © 2013 IEEE.
Open Access
Online text classification for real life tweet analysis
(IEEE, 2016) Yar, Ersin; Delibalta, İ.; Baruh, L.; Kozat, Süleyman Serdar
In this paper, we study multi-class classification of tweets, where we introduce highly efficient dimensionality reduction techniques suitable for online processing of high dimensional feature vectors generated from freely-worded text. As for the real life case study, we work on tweets in the Turkish language, however, our methods are generic and can be used for other languages as clearly explained in the paper. Since we work on a real life application and the tweets are freely worded, we introduce text correction, normalization and root finding algorithms. Although text processing and classification are highly important due to many applications such as emotion recognition, advertisement selection, etc., online classification and regression algorithms over text are limited due to need for high dimensional vectors to represent natural text inputs. We overcome such limitations by showing that randomized projections and piecewise linear models can be efficiently leveraged to significantly reduce the computational cost for feature vector extraction from the tweets. Hence, we can perform multi-class tweet classification and regression in real time. We demonstrate our results over tweets collected from a real life case study where the tweets are freely-worded, e.g., with emoticons, shortened words, special characters, etc., and are unstructured. We implement several well-known machine learning algorithms as well as novel regression methods and demonstrate that we can significantly reduce the computational complexity with insignificant change in the classification and regression performance.
Open Access
Rayleigh-bloch waves in CMUT arrays
(Institute of Electrical and Electronics Engineers Inc., 2014) Atalar, Abdullah; Köymen, Hayrettin; Oğuz, H. K.
Using the small-signal electrical equivalent circuit of a capacitive micromachined ultrasonic transducer (CMUT) cell, along with the self and mutual radiation impedances of such cells, we present a computationally efficient method to predict the frequency response of a large CMUT element or array. The simulations show spurious resonances, which may degrade the performance of the array. We show that these unwanted resonances are due to dispersive Rayleigh-Bloch waves excited on the CMUT surface-liquid interface. We derive the dispersion relation of these waves for the purpose of predicting the resonance frequencies. The waves form standing waves at frequencies where the reflections from the edges of the element or the array result in a Fabry-Pérot resonator. High-order resonances are eliminated by a small loss in the individual cells, but low-order resonances remain even in the presence of significant loss. These resonances are reduced to tolerable levels when CMUT cells are built from larger and thicker lates at the expense of reduced bandwidth. © 2014 IEEE.