Sensor for Classification of Material Type and Its Surface Properties Using Radial Basis Networks

Design of Near-Optimal Classifier Using Multi-Layer Perceptron Neural Networks for Intelligent Sensors

“…The generalization performance of the networks are validated meticulously on the basis of important parameters [12], [13] such as, mean square error (MSE), and percent classification accuracy (PCLA) on the testing instances, while attempting different input data partitions. Input data used in this paper belong to neural networks based intelligent sensor developed by the author [4]. A rigorous computer simulation has been carried out to design the near-optimal parameters of the MLP NN based classifiers.…”

“…Kato and Mukai [3] developed an intelligent gas sensor system for discrimination and quantification of gases by a single semiconductor gas sensor in real-time. Charniya and Dudul [4] developed neural networks-based intelligent sensor system for the classification of material type even with the variation in the sensor parameter. For classification, near-optimal classifier models were designed to maximize accuracy under the constraints of minimum network dimension.…”

A Radial Basis Function Neural Network Classifier for the Discrimination of Individual Odor Using Responses of Thick-Film Tin-Oxide Sensors

“…In the present work, the classification task of seven different types of alcoholic beverages is accomplished using already reported responses of thick film tin oxide sensor array [22]. Here, RBFNN was chosen as a classifier because RBFNNs are reported to outperform several times the multilayer perceptron neural networks (MLPNNs) with incredible speed and accuracy even with the smaller data sets [23]. RBFNNs have often being found to be better memorizers than generalizers.…”

“…Many new strategies have been devised recently to enhance the generalization capability of RBFNNs [24] so that they can be made more suitable for classification tasks. Recent studies have been able to establish the efficacy of RBFNNs in handling the classification tasks like classification of material types [23] and JPEG images [25]. Further investigation into data preprocessing techniques is the need of the hour in order to enhance the performance of RBFNNs as classifiers without altering their basic architecture.…”

Development of a spherical motor with a 3-DOF sensing system

“…Unlike the forward modeling-to solve MFD with given orientation, which can be solved under superposition principle once MFD of one or one pair of PMs is computed by analysis or FEM, the inverse modeling-to obtain three Euler angles from measured MFD is much more challenging, and it is very difficult to acquire a closed-loop equation unless high-order terms of the magnetic field are ignored. Neglecting high-order terms results in low accuracy, so neural networks[105][106][107][108][109][110][111][112] which can fit any practical function have been adopted here to approximate the relationship between inputs-measured MFD and outputsorientation angles.3-axis Hall Effect sensors as presented inFigure 4.1 are applied in the sensing system.A 3-axis Hall Effect sensor can output voltages in response to the magnetic field in 3 mutually perpendicular directions. The output voltages are proportional to magnetic flux density components in each direction over the working range of the sensors.…”