Special Issue "Neural Networks and Sensors"

Abstract: A new spectral processing technique designed for application in the on-line detection and classification of arc-welding defects is presented in this paper. A noninvasive fiber sensor embedded within a TIG torch collects the plasma radiation originated during the welding process. The spectral information is then processed in two consecutive stages. A compression algorithm is first applied to the data, allowing real-time analysis. The selected spectral bands are then used to feed a classification algorithm, which will be demonstrated to provide an efficient weld defect detection and classification. The results obtained with the proposed technique are compared to a similar processing scheme presented in previous works, giving rise to an improvement in the performance of the monitoring system.

Abstract: This work presents the design, manufacturing process, calibration and validation of a new microwave ten-port waveguide reflectometer based on the use of neural networks. This low-cost novel device solves some of the shortcomings of previous reflectometers such as non-linear behavior of power sensors, noise presence and the complexity of the calibration procedure, which is often based on complex mathematical equations. These problems, which imply the reduction of the reflection coefficient measurement accuracy, have been overcome by using a higher number of probes than usual six-port configurations and by means of the use of Radial Basis Function (RBF) neural networks in order to reduce the influence of noise and non-linear processes over the measurements. Additionally, this sensor can be reconfigured whenever some of the eight coaxial power detectors fail, still providing accurate values in real time. The ten-port performance has been compared against a high-cost measurement instrument such as a vector network analyzer and applied to the measurement and optimization of energy efficiency of microwave ovens, with good results.

Abstract: The presence of sea clutter in marine radar signals is sometimes not desired. So, efficient radar signal processing techniques are needed to reduce it. In this way, nonlinear signal processing techniques based on neural networks (NNs) are used in the proposed clutter reduction system. The developed experiments show promising results characterized by different subjective (visual analysis of the processed radar images) and objective (clutter reduction, target enhancement and signal-to-clutter ratio improvement) criteria. Moreover, a deep study of the NN structure is done, where the low computational cost and the high processing speed of the proposed NN structure are emphasized.

Abstract: Digital mobile mapping, which integrates digital imaging with direct geo-referencing, has developed rapidly over the past fifteen years. Direct geo-referencing is the determination of the time-variable position and orientation parameters for a mobile digital imager. The most common technologies used for this purpose today are satellite positioning using Global Positioning System (GPS) and Inertial Navigation System (INS) using an Inertial Measurement Unit (IMU). They are usually integrated in such a way that the GPS receiver is the main position sensor, while the IMU is the main orientation sensor. The Kalman Filter (KF) is considered as the optimal estimation tool for real-time INS/GPS integrated kinematic position and orientation determination. An intelligent hybrid scheme consisting of an Artificial Neural Network (ANN) and KF has been proposed to overcome the limitations of KF and to improve the performance of the INS/GPS integrated system in previous studies. However, the accuracy requirements of general mobile mapping applications can’t be achieved easily, even by the use of the ANN-KF scheme. Therefore, this study proposes an intelligent position and orientation determination scheme that embeds ANN with conventional Rauch-Tung-Striebel (RTS) smoother to improve the overall accuracy of a MEMS INS/GPS integrated system in post-mission mode. By combining the Micro Electro Mechanical Systems (MEMS) INS/GPS integrated system and the intelligent ANN-RTS smoother scheme proposed in this study, a cheaper but still reasonably accurate position and orientation determination scheme can be anticipated.

Abstract: An intelligent sensor for light wavelength readout, suitable for visible range optical applications, has been developed. Using buried triple photo-junction as basic pixel sensing element in combination with artificial neural network (ANN), the wavelength readout with a full-scale error of less than 1.5% over the range of 400 to 780 nm can be achieved. Through this work, the applicability of the ANN approach in optical sensing is investigated and compared with conventional methods, and a good compromise between accuracy and the possibility for on-chip implementation was thus found. Indeed, this technique can serve different purposes and may replace conventional methods.

Abstract: A new application of neurocomputing for data approximation and classification is introduced to process data in a wireless sensor network. For this purpose, a simplified dynamic sliding backpropagation algorithm is implemented on a wireless sensor network for transportation applications. It is able to approximate temperature and humidity in sensor nodes. In addition, two architectures of “radial basis function” (RBF) classifiers are introduced with probabilistic features for data classification in sensor nodes. The applied approximation and classification algorithms could be used in similar applications for data processing in embedded systems.

Abstract: This work presents two current-mode integrated circuits designed for sensor signal preprocessing in embedded systems. The proposed circuits have been designed to provide good signal transfer and fulfill their function, while minimizing the load effects due to building complex conditioning architectures. The processing architecture based on the proposed building blocks can be reconfigured through digital programmability. Thus, sensor useful range can be expanded, changes in the sensor operation can be compensated for and furthermore, undesirable effects such as device mismatching and undesired physical magnitudes sensor sensibilities are reduced. The circuits were integrated using a 0.35 mm standard CMOS process. Experimental measurements, load effects and a study of two different tuning strategies are presented. From these results, system performance is tested in an application which entails extending the linear range of a magneto-resistive sensor. Circuit area, average power consumption and programmability features allow these circuits to be included in embedded sensing systems as a part of the analogue conditioning components.

Abstract: Camera calibration is a crucial prerequisite for the retrieval of metric information from images. The problem of camera calibration is the computation of camera intrinsic parameters (i.e., coefficients of geometric distortions, principle distance and principle point) and extrinsic parameters (i.e., 3D spatial orientations: ω, φ, κ, and 3D spatial translations: tx, ty, tz). The intrinsic camera calibration (i.e., interior orientation) models the imaging system of camera optics, while the extrinsic camera calibration (i.e., exterior orientation) indicates the translation and the orientation of the camera with respect to the global coordinate system. Traditional camera calibration techniques require a predefined mathematical-camera model and they use prior knowledge of many parameters. Definition of a realistic camera model is quite difficult and computation of camera calibration parameters are error-prone. In this paper, a novel implicit camera calibration method based on Radial Basis Functions Neural Networks is proposed. The proposed method requires neither an exactly defined camera model nor any prior knowledge about the imaging-setup or classical camera calibration parameters. The proposed method uses a calibration grid-pattern rotated around a static-fixed axis. The rotations of the calibration grid-pattern have been acquired by using an Xsens MTi-9 inertial sensor and in order to evaluate the success of the proposed method, 3D reconstruction performance of the proposed method has been compared with the performance of a traditional camera calibration method, Modified Direct Linear Transformation (MDLT). Extensive simulation results show that the proposed method achieves a better performance than MDLT aspect of 3D reconstruction.

Abstract: The crowning objective of this research was to identify a better cloud classification method to upgrade the current window-based clustering algorithm used operationally for China’s first operational geostationary meteorological satellite FengYun-2C (FY-2C) data. First, the capabilities of six widely-used Artificial Neural Network (ANN) methods are analyzed, together with the comparison of two other methods: Principal Component Analysis (PCA) and a Support Vector Machine (SVM), using 2864 cloud samples manually collected by meteorologists in June, July, and August in 2007 from three FY-2C channel (IR1, 10.3-11.3 μm; IR2, 11.5-12.5 μm and WV 6.3-7.6 μm) imagery. The result shows that: (1) ANN approaches, in general, outperformed the PCA and the SVM given sufficient training samples and (2) among the six ANN networks, higher cloud classification accuracy was obtained with the Self-Organizing Map (SOM) and Probabilistic Neural Network (PNN). Second, to compare the ANN methods to the present FY-2C operational algorithm, this study implemented SOM, one of the best ANN network identified from this study, as an automated cloud classification system for the FY-2C multi-channel data. It shows that SOM method has improved the results greatly not only in pixel-level accuracy but also in cloud patch-level classification by more accurately identifying cloud types such as cumulonimbus, cirrus and clouds in high latitude. Findings of this study suggest that the ANN-based classifiers, in particular the SOM, can be potentially used as an improved Automated Cloud Classification Algorithm to upgrade the current window-based clustering method for the FY-2C operational products.

Abstract: This paper introduces a comparison of training algorithms of radial basis function (RBF) neural networks for classification purposes. RBF networks provide effective solutions in many science and engineering fields. They are especially popular in the pattern classification and signal processing areas. Several algorithms have been proposed for training RBF networks. The Artificial Bee Colony (ABC) algorithm is a new, very simple and robust population based optimization algorithm that is inspired by the intelligent behavior of honey bee swarms. The training performance of the ABC algorithm is compared with the Genetic algorithm, Kalman filtering algorithm and gradient descent algorithm. In the experiments, not only well known classification problems from the UCI repository such as the Iris, Wine and Glass datasets have been used, but also an experimental setup is designed and inertial sensor based terrain classification for autonomous ground vehicles was also achieved. Experimental results show that the use of the ABC algorithm results in better learning than those of others.

Abstract: The increasing technology of high-resolution image airborne sensors, including those on board Unmanned Aerial Vehicles, demands automatic solutions for processing, either on-line or off-line, the huge amountds of image data sensed during the flights. The classification of natural spectral signatures in images is one potential application. The actual tendency in classification is oriented towards the combination of simple classifiers. In this paper we propose a combined strategy based on the Deterministic Simulated Annealing (DSA) framework. The simple classifiers used are the well tested supervised parametric Bayesian estimator and the Fuzzy Clustering. The DSA is an optimization approach, which minimizes an energy function. The main contribution of DSA is its ability to avoid local minima during the optimization process thanks to the annealing scheme. It outperforms simple classifiers used for the combination and some combined strategies, including a scheme based on the fuzzy cognitive maps and an optimization approach based on the Hopfield neural network paradigm.

Abstract: Artificial neural network (ANN) based prediction of the response of a microbend fiber optic sensor is presented. To the best of our knowledge no similar work has been previously reported in the literature. Parallel corrugated plates with three deformation cycles, 6 mm thickness of the spacer material and 16 mm mechanical periodicity between deformations were used in the microbend sensor. Multilayer Perceptron (MLP) with different training algorithms, Radial Basis Function (RBF) network and General Regression Neural Network (GRNN) are used as ANN models in this work. All of these models can predict the sensor responses with considerable errors. RBF has the best performance with the smallest mean square error (MSE) values of training and test results. Among the MLP algorithms and GRNN the Levenberg-Marquardt algorithm has good results. These models successfully predict the sensor responses, hence ANNs can be used as useful tool in the design of more robust fiber optic sensors.

Abstract: In this study, a hybridized neuro-genetic optimization methodology realized by embedding numerical simulations trained artificial neural networks (ANN) into a genetic algorithm (GA) is used to optimize the flow rectification efficiency of the diffuser element for a valveless diaphragm micropump application. A higher efficiency ratio of the diffuser element consequently yields a higher flow rate for the micropump. For that purpose, optimization of the diffuser element is essential to determine the maximum pumping rate that the micropump is able to generate. Numerical simulations are initially carried out using CoventorWare® to analyze the effects of varying parameters such as diffuser angle, Reynolds number and aspect ratio on the volumetric flow rate of the micropump. A limited range of simulation results will then be used to train the neural network via back-propagation algorithm and optimization process commence subsequently by embedding the trained ANN results as a fitness function into GA. The objective of the optimization is to maximize the efficiency ratio of the diffuser element for the range of parameters investigated. The optimized efficiency ratio obtained from the neuro-genetic optimization is 1.38, which is higher than any of the maximum efficiency ratio attained from the overall parametric studies, establishing the superiority of the optimization method.

Abstract: This paper proposes a wavelet neural network (WNN) for SAR image segmentation by combining the wavelet transform and an artificial neural network. The WNN combines the multiscale analysis ability of the wavelet transform and the classification capability of the artificial neural network by setting the wavelet function as the transfer function of the neural network. Several SAR images are segmented by the network whose transfer functions are the Morlet and Mexihat functions, respectively. The experimental results show the proposed method is very effective and accurate.

Abstract: This paper proposes a hybrid classifier for polarimetric SAR images. The feature sets consist of span image, the H/A/α decomposition, and the GLCM-based texture features. Then, a probabilistic neural network (PNN) was adopted for classification, and a novel algorithm proposed to enhance its performance. Principle component analysis (PCA) was chosen to reduce feature dimensions, random division to reduce the number of neurons, and Brent’s search (BS) to find the optimal bias values. The results on San Francisco and Flevoland sites are compared to that using a 3-layer BPNN to demonstrate the validity of our algorithm in terms of confusion matrix and overall accuracy. In addition, the importance of each improvement of the algorithm was proven.

Abstract: This paper describes the procedures for development of signal analysis algorithms using artificial neural networks for Bridge Weigh-in-Motion (B-WIM) systems. Through the analysis procedure, the extraction of information concerning heavy traffic vehicles such as weight, speed, and number of axles from the time domain strain data of the B-WIM system was attempted. As one of the several possible pattern recognition techniques, an Artificial Neural Network (ANN) was employed since it could effectively include dynamic effects and bridge-vehicle interactions. A number of vehicle traveling experiments with sufficient load cases were executed on two different types of bridges, a simply supported pre-stressed concrete girder bridge and a cable-stayed bridge. Different types of WIM systems such as high-speed WIM or low-speed WIM were also utilized during the experiments for cross-checking and to validate the performance of the developed algorithms.

Abstract: The Integral Equation Model with multiple scattering (IEMM) represents a well-established method that provides a theoretical framework for the scattering of electromagnetic waves from rough surfaces. A critical aspect is the long computational time required to run such a complex model. To deal with this problem, a neural network technique is proposed in this work. In particular, we have adopted neural networks to reproduce the backscattering coefficients predicted by IEMM at L- and C-bands, thus making reference to presently operative satellite radar sensors, i.e., that aboard ERS-2, ASAR on board ENVISAT (C-band), and PALSAR aboard ALOS (L-band). The neural network-based model has been designed for radar observations of both flat and tilted surfaces, in order to make it applicable for hilly terrains too. The assessment of the proposed approach has been carried out by comparing neural network-derived backscattering coefficients with IEMM-derived ones. Different databases with respect to those employed to train the networks have been used for this purpose. The outcomes seem to prove the feasibility of relying on a neural network approach to efficiently and reliably approximate an electromagnetic model of surface scattering.

Abstract: I take this opportunity to discuss a few things that I have learned from being Guest Editor of this special issue of Sensors devoted to Neural Networks and Sensors. The advancement in artificial neural network (ANN) technology is very impressive. The wide variety of fields in which this technology applies in the form of practical applications to clearly identifiable real problems demonstrates that ANNs are being routinely used to solve non-trivial problems. I mention that because A. K. Dewdney wrote in 1997 that while ANNs have been used to solve a few toy problems, he was surprised that anyone takes them seriously as general problem-solving tools [1]. The ANN applications reported by Yu Liu et al. [2], Erkan Beşdok [3], Guillermo Zatorre et al. [4], Amir Jabbari et al. [5], Mohamed Lamine Hafiane et al. [6], Kai-Wei Chiang et al. [7], Raúl Vicen-Bueno et al. [8], Juan L. Pedreño-Molina et al. [9], and P. B. Garcia-Allende et al. [10] are far more than toy applications. The lesson to be learned here is that it is a bad idea to publically bet against technological progress in computer applications. [...]