Neural network approach to land cover mapping

A pattern classification method is proposed for remote sensing data using neural networks. First, the authors apply the error backpropagation (BP) algorithm to classify the remote sensing data. In this case, the classification performance depends on a training data set. In order to get stable and precise classification results, the training data set is selected based on geographical information and Kohonen's self-organizing feature map. Using the training data set and the error backpropagation algorithm, a layered neural network is trained such that the training patterns are classified with a specified accuracy. After training the neural network, some pixels are deleted from the original training data set if they are incorrectly classified and a new training data set is built up. Once training is complete, a testing data set is classified by using the trained neural network. The classification results of LANDSAT TM data show that this approach produces excellent results which are more realistic and noiseless compared with a conventional Bayesian method     

A multi-objective strategy in genetic algorithms for gene selection of gene expression data

A microarray machine offers the capacity to measure the expression levels of thousands of genes simultaneously. It is used to collect information from tissue and cell samples regarding gene expression differences that could be useful for cancer classification. However, the urgent problems in the use of gene expression data are the availability of a huge number of genes relative to the small number of available samples, and the fact that many of the genes are not relevant to the classification. It has been shown that selecting a small subset of genes can lead to improved accuracy in the classification. Hence, this paper proposes a solution to the problems by using a multiobjective strategy in a genetic algorithm. This approach was tried on two benchmark gene expression data sets. It obtained encouraging results on those data sets as compared with an approach that used a single-objective strategy in a genetic algorithm. This work was presented in part at the 13th International Symposium on Artificial Life and Robotics, Oita, Japan, January 31–February 2, 2008     

Selecting informative genes from microarray data by using hybrid methods for cancer classification

Gene expression technology, namely microarrays, offers the ability to measure the expression levels of thousands of genes simultaneously in biological organisms. Microarray data are expected to be of significant help in the development of an efficient cancer diagnosis and classification platform. A major problem in these data is that the number of genes greatly exceeds the number of tissue samples. These data also have noisy genes. It has been shown in literature reviews that selecting a small subset of informative genes can lead to improved classification accuracy. Therefore, this paper aims to select a small subset of informative genes that are most relevant for cancer classification. To achieve this aim, an approach using two hybrid methods has been proposed. This approach is assessed and evaluated on two well-known microarray data sets, showing competitive results. This work was presented in part at the 13th International Symposium on Artificial Life and Robotics, Oita, Japan, January 31–February 2, 2008     

Selection of parameters in the design of real-time spectral analysis

A new analytical method with high speed processing in the time-frequency domain is presented. In this method, sine and cosine waves with an established frequency and multiple periods are used, and we call these waves “cutting-out waves.” We all the frequency the “established frequency,” and we call the number of periods of the cutting-out wave the “number of periods.” The inner product of the cutting-out wave and the signal are calculated, and a signal element with a frequency near the established frequency is detected. We call the unit that detects the signal element an “auditory cell.” There are many auditory cells, and they have an established frequency which differs very little. The design of this method is the arrangement of the auditory cells. There are three parameters in the design, and these parameters are a sampling frequency, the number of periods, and the increasing rate of the established frequencies. In this article, we show the selection of these parameters.     

Fuzzy ARTMAP classification of invariant features derived using angle of rotation from a neural network

Conventional regular moment functions have been proposed as pattern sensitive features in image classification and recognition applications. But conventional regular moments are only invariant to translation, rotation and equal scaling. It is shown that the conventional regular moment invariants remain no longer invariant when the image is scaled unequally in the x- and y-axis directions. We address this problem by presenting a technique to make the regular moment functions invariant to unequal scaling. However, the technique produces a set of features that are only invariant to translation, unequal/equal scaling and reflection. They are not invariant to rotation. To make them invariant to rotation, moments are calculated with respect to the principal axis of the image. To perform this, the exact angle of rotation must be known. But the method of using the second-order moments to determine this angle will also be inclusive of an undesired tilt angle. Therefore, in order to correctly determine the amount of rotation, the tilt angle which differs for different scaling factors in the x- and y-axis directions for the particular image must be obtained. In order to solve this problem, a neural network using the back-propagation learning algorithm is trained to estimate the tilt angle of the image and from this the amount of rotation for the image can be determined. Next, the new moments are derived and a Fuzzy ARTMAP network is used to classify these images into their respective classes. Sets of experiments involving images rotated and scaled unequally in the x- and y-axis directions are carried out to demonstrate the validity of the proposed technique.     

An evaluation of an olfactory sensor and recognition system using metal-oxide semiconductor gas sensor with variable heater controller

An artificial olfactory, which is called an electronic nose system (e-nose), is studied for realizing new human–machine interface. The system consists of sensor unit and a signal processing unit. There are some types of sensors for the sensor unit, metal oxide semiconductor gas sensors (MOGS) and quartz crystal microbalance gas sensors are useful in our study. Our system in this paper has MOGS. Many of MOGS utilize an effect of an oxidation–reduction reaction on the surface of the sensors. One of the features of the sensor, the character of the sensor is changed by temperature of the sensors. In this paper, we build a extend output sensor unit using this feature, and show experimental result of classification applying multilayer perceptron. In the experiment, we choose soy sauce as classification targets because we are considering applying the system for management of cooked foods.     

Graph extension with constant connectivity

In the field of the scaling-up of communication networks, numbers of communication stations (nodes) and the corresponding communication links (edges) are increasing rapidly. The reliability of the networks then becomes important. To keep the reliability of the networks, the connectivity (invulnerability) of the networks should not decrease. In this article, a method of increasing the nodes of a graph with a constant connectivity is proposed, and some examples of graph extension are shown to realize the extended networks. This work was presented in part at the 11th International Symposium on Artificial Life and Robotics, Oita, Japan, January 23–25, 2006     

Extraction rice-planted areas by RADARSAT data using neural networks

A classification technique using the neural networks has recently been developed. We apply a neural network of learning vector quantization (LVQ) to classify remote-sensing data, including microwave and optical sensors, for the estimation of a rice-planted area. The method has the capability of nonlinear discrimination, and the classification function is determined by learning. The satellite data were observed before and after planting rice in 1999. Three sets of RADARSAT and one set of SPOT/HRV data were used in Higashi–Hiroshima, Japan. Three RADARSAT images from April to June were used for this study. The LVQ classification was applied the RADARSAT and SPOT to evaluate the estimate of the area of planted-rice. The results show that the true production rate of the rice-planted area estimation of RADASAT by LVQ was approximately 60% compared with that of SPOT by LVQ. It is shown that the present method is much better than the SAR image classification by the maximum likelihood method.     

Gene subset selection using an iterative approach based on genetic algorithms

Microarray data are expected to be useful for cancer classification. However, the process of gene selection for the classification contains a major problem due to properties of the data such as the small number of samples compared with the huge number of genes (higher-dimensional data), irrelevant genes, and noisy data. Hence, this article aims to select a near-optimal (small) subset of informative genes that is most relevant for the cancer classification. To achieve this aim, an iterative approach based on genetic algorithms has been proposed. Experimental results show that the performance of the proposed approach is superior to other previous related work, as well as to four methods tried in this work. In addition, a list of informative genes in the best gene subsets is also presented for biological usage.