Property:Abstract
From ISLAB/CAISR
Jump to navigationJump to searchThis is a property of type Text.
M
<p>Complex moments of the local power spectrum (CMP) are investigated in a multiscale context. The multiscale CMPs are shown to approximate well the 1D angular Fourier transform of the band in question. This observation is used to derive further properties of the power spectrum in terms of texture orientations or n-folded symmetry patterns. A method is presented to approximate the power spectrum using only separable filtering in the spatial domain. Interesting implications to the Gabor decomposition are shown. The number of orientations in the filter bank is related to the order of n-folded symmetry detectable. Furthermore, the multiscale CMPs can be estimated incrementally in the spatial domain, which is both fast and reliable. Experiments on power spectrum estimation, orientation estimation, and texture segmentation are presented.</p> +
N
Near-infrared and visible-light periocular recognition with Gabor features using frequency-adaptive automatic eye detection +
<p>Periocular recognition has gained attention recently due to demands of increased robustness of face or iris in less controlled scenarios. We present a new system for eye detection based on complex symmetry filters, which has the advantage of not needing training. Also, separability of the filters allows faster detection via one-dimensional convolutions. This system is used as input to a periocular algorithm based on retinotopic sampling grids and Gabor spectrum decomposition. The evaluation framework is composed of six databases acquired both with near-infrared and visible sensors. The experimental setup is complemented with four iris matchers, used for fusion experiments. The eye detection system presented shows very high accuracy with near-infrared data, and a reasonable good accuracy with one visible database. Regarding the periocular system, it exhibits great robustness to small errors in locating the eye centre, as well as to scale changes of the input image. The density of the sampling grid can also be reduced without sacrificing accuracy. Lastly, despite the poorer performance of the iris matchers with visible data, fusion with the periocular system can provide an improvement of more than 20%. The six databases used have been manually annotated, with the annotation made publicly available. © The Institution of Engineering and Technology 2015.</p> +
<p>Creating fault detection software for complex mechatronic systems (e.g. modern vehicles) is costly both in terms of engineer time and hardware resources. With the availability of wireless communication in vehicles, information can be transmitted from vehicles to allow historical or fleet comparisons. New networked applications can be created that, e.g., monitor if the behavior of a certain system in a vehicle deviates compared to the system behavior observed in a fleet. This allows a new approach to fault detection that can help reduce development costs of fault detection software and create vehicle individual service planning. The COSMO (consensus self-organized modeling) methodology described in this paper creates a compact representation of the data observed for a subsystem or component in a vehicle. A representation that can be sent to a server in a backoffice and compared to similar representations for other vehicles. The backoffice server can collect representations from a single vehicle over time or from a fleet of vehicles to define a norm of the vehicle condition. The vehicle condition can then be monitored, looking for deviations from the norm. The method is demonstrated for measurements made on a real truck driven in varied conditions with ten different generated faults. The proposed method is able to detect all cases without prior information on what a fault looks like or which signals to use.</p> +
<p>We propose two virtual sensors which estimate the location of the pressure peak and the air-fuel ratio from measurements of the ionization current across the spark plug gap.</p><p>The location of pressure peak virtual sensor produces estimates on a cycle-by-cycle basis for each of the cylinders. These estimates are twice as good as estimates obtained from a linear model.</p><p>The air-fuel ratio virtual sensor uses the universal exhaust gas oxygen sensor as reference; it produces estimates that are ten times better than estimates obtained from a linear model.</p> +
<p>In this paper, we present an approach to neural modelling and control of the offset lithographic printing process. A committee of neural networks is trained to measure the printing process output - the observable variables. From only one measurement the trained committee is capable of estimating the actual relative amount of each cyan, magenta, yellow, and black inks dispersed on paper in the measuring area. The obtained measurements are then further used by a neural model predictive control unit for generating control signals to compensate for colour deviation in offset newspaper printing. The experimental investigations performed have shown that the system developed achieves a higher printing process control accuracy than that usually obtained by the press operator.</p> +
Neural networks based colour measuring for process monitoring and control in multicoloured newspaper printing +
<p>This paper presents a neural networks based method and a system for colour measurements on printed halftone multicoloured pictures and halftone multicoloured bars in newspapers. The measured values, called a colour vector, are used by the operator controlling the printing process to make appropriate ink feed adjustments to compensate for colour deviations of the picture being measured from the desired print. By the colour vector concept, we mean the CMY or CMYK (cyan, magenta, yellow and black) vector, which lives in the three- or four-dimensional space of printing inks. Two factors contribute to values of the vector components, namely the percentage of the area covered by cyan, magenta, yellow and black inks (tonal values) and ink densities. Values of the colour vector components increase if tonal values or ink densities rise, and vice versa. If some reference values of the colour vector components are set from a desired print, then after an appropriate calibration, the colour vector measured on an actual halftone multicoloured area directly shows how much the operator needs to raise or lower the cyan, magenta, yellow and black ink densities to compensate for colour deviation from the desired print. The 18 months experience of the use of the system in the printing shop witnesses its usefulness through the improved quality of multicoloured pictures, the reduced consumption of inks and, therefore, less severe problems of smearing and printing through.</p> +
<p>Aiming to hide the real money gains and to avoid taxes, fictive prices are sometimes recorded in the real estate transactions. This paper is concerned with artificial neural networks based screening of real estate transactions aiming to categorize them into "clear" and "fictitious" classes. The problem is treated as an outlier detection task. Both unsupervised and supervised approaches to outlier detection are studied here. The soft minimal hyper-sphere support vector machine (SVM) based novelty detector is employed to solve the task without the supervision. In the supervised case, the effectiveness of SVM, multilayer perceptron (MLP), and a committee based classification of the real estate transactions are studied. To give the user a deeper insight into the decisions provided by the models, the real estate transactions are not only categorized into "clear" and "fictitious" classes, but also mapped onto the self organizing map (SOM), where the regions of "clear", "doubtful" and "fictitious" transactions are identified. We demonstrate that the stability of the regions evolved in the SOM during training is rather high. The experimental investigations performed on two real data sets have shown that the categorization accuracy obtained from the supervised approaches is considerably higher than that obtained from the unsupervised one. The obtained accuracy is high enough for the technique to be used in practice. © ICS AS CR 2007.</p> +
<p>In recent years, there has been increased research interest in detecting anomalies in temporal streaming data. A variety of algorithms have been developed in the data mining community, which can be divided into two categories (i.e., general and ad hoc). In most cases, general approaches assume the one-size-fits-all solution model where a single anomaly detector can detect all anomalies in any domain. To date, there exists no single general method that has been shown to outperform the others across different anomaly types, use cases and datasets. On the other hand, ad hoc approaches that are designed for a specific application lack flexibility. Adapting an existing algorithm is not straightforward if the specific constraints or requirements for the existing task change. In this paper, we propose SAFARI, a general framework formulated by abstracting and unifying the fundamental tasks in streaming anomaly detection, which provides a flexible and extensible anomaly detection procedure. SAFARI helps to facilitate more elaborate algorithm comparisons by allowing us to isolate the effects of shared and unique characteristics of different algorithms on detection performance. Using SAFARI, we have implemented various anomaly detectors and identified a research gap that motivates us to propose a novel learning strategy in this work. We conducted an extensive evaluation study of 20 detectors that are composed using SAFARI and compared their performances using real-world benchmark datasets with different properties. The results indicate that there is no single superior detector that works well for every case, proving our hypothesis that "there is no free lunch" in the streaming anomaly detection world. Finally, we discuss the benefits and drawbacks of each method in-depth and draw a set of conclusions to guide future users of SAFARI.</p> +
<p>In recent years, there has been increased research interest in detecting anomalies in temporal streaming data. A variety of algorithms have been developed in the data mining community, which can be divided into two categories (i.e., general and ad hoc). In most cases, general approaches assume the one-size-fits-all solution model where a single anomaly detector can detect all anomalies in any domain. To date, there exists no single general method that has been shown to outperform the others across different anomaly types, use cases and datasets. On the other hand, ad hoc approaches that are designed for a specific application lack flexibility. Adapting an existing algorithm is not straightforward if the specific constraints or requirements for the existing task change. In this paper, we propose SAFARI, a general framework formulated by abstracting and unifying the fundamental tasks in streaming anomaly detection, which provides a flexible and extensible anomaly detection procedure. SAFARI helps to facilitate more elaborate algorithm comparisons by allowing us to isolate the effects of shared and unique characteristics of different algorithms on detection performance. Using SAFARI, we have implemented various anomaly detectors and identified a research gap that motivates us to propose a novel learning strategy in this work. We conducted an extensive evaluation study of 20 detectors that are composed using SAFARI and compared their performances using real-world benchmark datasets with different properties. The results indicate that there is no single superior detector that works well for every case, proving our hypothesis that "there is no free lunch" in the streaming anomaly detection world. Finally, we discuss the benefits and drawbacks of each method in-depth and draw a set of conclusions to guide future users of SAFARI.</p> +
<p>In this paper we present a novel method for remote breathing detection based on ultra-wide-band (UWB) radar. This is a method that does not require any wearable sensors, making it more comfortable and convenient for users. Furthermore, because of the wall penetrating characteristics of the transmitted signal, our system is useful in emergency situations such as monitoring people who may be trapped under earthquake rubble. For our investigation we used a Novelda UWB radar that provides high processing speed and low power consumption. In this paper we present two new convolution-based methods to extract breathing rate information from the received radar signal. This method was tested on several people who were monitored while laying down on a bed. The subject's position and breathing rate were calculated. Experimental results including 20 different subjects are provided, showing that this is a viable method for monitoring breathing rate using a low-power UWB radar.</p> +
<p>A technique evaluating liveness in face image sequences is presented. To ensure the actual presence of a live face in contrast to a photograph (playback attack), is a significant problem in face authentication to the extent that anti-spoofing measures are highly desirable. The purpose of the proposed system is to assist in a biometric authentication framework, by adding liveness awareness in a non-intrusive manner. Analyzing the trajectories of certain parts of a live face reveals valuable information to discriminate it against a spoofed one. The proposed system uses a lightweight novel optical flow, which is especially applicable in face motion estimation based on the structure tensor and inputs of a few frames. For reliable face part detection, the system utilizes a model-based local Gabor decomposition and SVM experts, where selected points from a retinotopic grid are used to form regional face models. Also the estimated optical flow is exploited to detect a face part. The whole procedure, starting with three images as input and finishing in a liveness score, is executed in near real-time without special purpose hardware. Experimental results on the proposed system are presented on both a public database and spoofing attack simulations.</p> +
Nonlinear Optimization of Multimodal Two-Dimensional Map Alignment With Application to Prior Knowledge Transfer +
<p>We propose a method based on a nonlinear transformation for nonrigid alignment of maps of different modalities, exemplified with matching partial and deformed two-dimensional maps to layout maps. For two types of indoor environments, over a dataset of 40 maps, we have compared the method to state-of-the-art map matching and nonrigid image registration methods and demonstrate a success rate of 80.41% and a mean point-to-point alignment error of 1.78 m, compared to 31.9% and 10.7 m for the best alternative method. We also propose a fitness measure that can quite reliably detect bad alignments. Finally, we show a use case of transferring prior knowledge (labels/segmentation), demonstrating that map segmentation is more consistent when transferred from an aligned layout map than when operating directly on partial maps (95.97% vs. 81.56%). © 2018 IEEE.</p> +
<p>This paper presents a method for mining nonlinear relationships in machine data with the purpose of using such relationships to detect faults, isolate faults and predict wear and maintenance needs. The method is based on the symmetrical uncertainty measure from information theory, hierarchical clustering and self-organizing maps. It is demonstrated on synthetic data sets where it is shown to be able to detect interesting signal relations and outperform linear methods. It is also demonstrated on real data sets where it is considerably harder to select small feature sets. It is also demonstrated on the real data sets that there is information about system wear and system faults in the detected relationships. The work is part of a long-term research project with the aim to construct a self-organizing autonomic computing system for self-monitoring of mechatronic systems.</p> +
<p>Although wearable devices can be used toperform continuous gait analysis in daily life, existing platforms only support short-term analysis in quasi-controlled environments. This paper proposes a novel system architecture that is designed for long-term, online gait analysis in free-living environments. Various aspects related to the feasibility and scalability of the proposed system are presented.</p> +
Novel methodology for estimating Initial Contact events from accelerometers positioned at different body locations +
<p>Identifying Initial Contact events (ICE) is essential in gait analysis as they segment the walking pattern into gait cycles and facilitate the computation of other gait parameters. As such, numerous algorithms have been developed to identify ICE by placing the accelerometer at a specific body location. Simultaneously, many researchers have studied the effects of device positioning for participant or patient compliance, which is an important factor to consider especially for long-term studies in real-life settings. With the adoption of accelerometery for long-term gait analysis in daily living, current and future applications will require robust algorithms that can either autonomously adapt to changes in sensor positioning or can detect ICE from multiple sensors locations.</p><p>This study presents a novel methodology that is capable of estimating ICE from accelerometers placed at different body locations. The proposed methodology, called DK-TiFA, is based on utilizing domain knowledge about the fundamental spectral relationships present between the movement of different body parts during gait to drive the time-frequency analysis of the acceleration signal. In order to assess the performance, DK-TiFA is benchmarked on four large publicly available gait databases, consisting of a total of 613 subjects and 7 unique body locations, namely, ankle, thigh, center waist, side waist, chest, upper arm and wrist. The DK-TiFA methodology is demonstrated to achieve high accuracy and robustness for estimating ICE from data consisting of different accelerometer specifications, varying gait speeds and different environments.</p> +
O
Object Recognition Based on Radial Basis Function Neural Networks: experiments with RGB-D camera embedded on mobile robots +
<p>An object recognition strategy based on artificial radial basis functions neural networks is presented in this paper. The general context of this work is to recognize object from captures made by a mobile robot. Unlike classical approaches which always select the closest object, our method outputs a set of potential candidates if the input information is not enough discriminant. There are three main steps in our approach: objects segmentation, signature extraction and classification. Segmentation is inspired from previous works and is shortly described. Signature extraction based on global geometric and color features is detailed. Classification based on artificial neural networks is also explained and architecture of the network is justified. Finally a real experiment made with a RGB-D camera mounted on a mobile robot is presented and classification results is criticized.</p> +
<p>Many vision-based approaches for obstacle detection often state that vertical thin structure is of importance, e.g. poles and trees. However, there are also problem in detecting thin horizontal structures. In an industrial case there are horizontal objects, e.g. cables and fork lifts, and slanting objects, e.g. ladders, that also has to be detected. This paper focuses on the problem to detect thin horizontal structures. The system uses three cameras, situated as a horizontal pair and a vertical pair, which makes it possible to also detect thin horizontal structures. A comparison between a sparse disparity map based on edges and a dense disparity map with a column and row filter is made. Both methods use the Sum of Absolute Difference to compute the disparity maps. Special interest has been in scenes with thin horizontal objects. Tests show that the sparse dense method based on the Canny edge detector works better for the environments we have tested.</p> +
<p>This paper describes two appraches for Amharic word recognition in unconstrained handwritten text using HMMs. The first approach builds word models from concatenated features of constituent characters and in the second method HMMs of constituent characters are concatenated to form word model. In both cases, the features used for training and recognition are primitive strokes and their spatial relationships. The recognition system does not require segmentation of characters but requires text line detection and extraction of structural features, which is done by making use of direction field tensor. The performance of the recognition system is tested by DEHR dataset of unconstrained handwritten documents collected from various sources.</p> +
<p>This paper describes two approaches for Amharic word recognition in unconstrained handwritten text using HMMs. The first approach builds word models from concatenated features of constituent characters and in the second method HMMs of constituent characters are concatenated to form word model. In both cases, the features used for training and recognition are a set of primitive strokes and their spatial relationships. The recognition system does not require segmentation of characters but requires text line detection and extraction of structural features, which is done by making use of direction field tensor. The performance of the recognition system is tested by a dataset of unconstrained handwritten documents collected from various sources, and promising results are obtained. (C) 2011 Elsevier B.V. All rights reserved.</p> +
Y. Assabie, J. Bigun +
