Segmentation Of Image Sequences Using Self-Organising Feature Maps

Segmentation of gray level images into regions of uniform texture is investigated. An unsupervised approach through the use of Kohonen's self organizing map (SOM) and a multilayer version of it, the hierarchical self organizing map (HSOM), is employed to find the regions in an image composed of textures from different classes. For testing, gray level artificial textured images modeled as Markov random fields are used as the input. No parameter estimation is done. The size and the topology of SOM and HSOM are independent from the size of the input image. The segmentation results are very promising.

Summary Well-test data have been used traditionally for determining a variety of reservoir parameters, such as average permeability, storage capacity, reservoir damage, presence of faults and fractures, and reservoir mechanism. A number of techniques, both conventional methods, such as type-curve matching and numerical simulation, and artificial intelligence (AI) methods, have been used for identifying well-test models. These methods are laborious and time-consuming and at times give incorrect results. Artificial neural networks (ANN's) are recent developments in computer vision and image analysis. These networks are specialized computer software that generate a strategy to produce nonlinear mapping functions for complex problems. ANN's are commonly used as a tool for recognizing an object or predicting an event given an associated pattern. Only a limited number of applications of ANN for analyzing well-test data have been reported. These applications are mostly model-specific (developed for specific reservoir models) and, hence, are not general enough. This paper presents a new method based on ANN's that uses Ko-honen's self-organizing feature (SOF) mapping technique to identify well-test interpretation models. By grouping well-test data into distinct categories, the SOF algorithm produces a general mapping function. This method can help analyze well-test data from a large variety of reservoirs (including reservoirs with faults, fractures, boundaries, etc.) more efficiently and inexpensively than was feasible previously.

Spiking Self-organizing Maps for Classification Problem

Enhancing the content and structure of a web site is a very important task which can help to maintain people visiting a web site and gain new visits (or customers). Web mining area helps to enhance a web site organization and contents using data mining algorithms. In particular we may perform Web Mining using a Self Organizing Feature Map (SOFM or SOM) it is always needed an analysis phase by experts. To help analysts to perform this phase after SOFMs' training, many post-processing techniques have been developed (component planes, labels, etc.); however, none of these techniques are useful when working in web mining for off-line enhancements of a web site. In this paper an algorithm called Reverse Cluster Analysis (RCA) will be provided. It aims to identify important web pages based on a self organizing feature map (SOFM) when performing web text mining (WTM) and web usage mining (WUM). We successfully applied this technique in a real web site to show its effectiveness. We have extended previous work performing a comparison with another unsupervised technique, administrators survey and an extended survey.

The Self Organizing Map (SOM) involves neural networks, that learns the features of input data thorough unsupervised, competitive neighborhood learning. In the SOM learning algorithm, connection weights in a SOM feature map are initialized at random values, which also sets nodes at random locations in the feature map independent of input data space. The move distance of output nodes increases, slowing learning convergence. As precedence research, we proposed the method to improve this problem, initial node exchange by using a part of feature map. In this paper, we propose new exchange method, node exchange with fixed neighbor area. The method uses fixed position of winner node and fixed initial size of neighbor area that sets to cover whole feature map. We investigate how average move distance of all nodes and average deviation of move distance would change with the differences by type of fixed neighbor area in node exchange process. As a result of experiments, sufficient effect is acquired by fixed 1 neighbor area in the viewpoint of both average move distance of all nodes and average deviation of move distance.

Kohonen's self-organizing map (SOM), which performs topology-preserving transformation from a high dimensional data vector space to a low-dimensional map space, provides a powerful tool for data analysis, classification and visualization in many application fields. Despite its power, SOM can only deal with vectorized data, although many expansions have been proposed for various data-type cases. This study aims to develop a novel generalization of SOM called modular network SOM (mnSOM), which enables users to deal with general data classes in a consistent manner. mnSOM has an array structure consisting of function modules that are trainable neural networks, e.g. multi-layer perceptrons (MLPs), instead of the vector units of the conventional SOM family. In the case of MLP-modules, mnSOM learns a group of systems or functions in terms of the input-output relationships, and at the same time, mnSOM generates a feature map that shows distances between the learned systems. Thus, mnSOM with MLP modules is an SOM in function space rather than in vector space. From this point of view, the conventional SOM of Kohonen's can be regarded as a special case of mnSOM, the modules consisting of fixed-value bias units. In this paper, mnSOM with MLP modules is described along with some application examples.

An unsupervised learning network, such as the self organizing feature map (SOFM), has been applied successfully to color classification for image compression and pattern recognition. Like other vector quantization algorithms, the reconstruction quality and adaptation rate of the SOFM are sensitive to the neuron initialization. We propose an efficient new initialization method, whereby an excess number of neurons is defined and the neurons are adaptively pruned, merged and split within their lattice according to the spatial distribution of the input color pixels. Comparisons with conventional gray scale initialization using subsampling and butterfly jumping sequences show that the proposed method obtains good initial code vectors that can accelerate the convergence of the SOFM and improve the reconstructed image quality significantly.

We have been working on the application of Machining Learning in Metabolomics, Drug-Drug Interaction Discovery and Human Gait Recognition [1-5], profiling large data sets. Extraction of vital information about 1) plant metabolomics that can improve the environment and food quality, 2) in vitro neuronal network behavior patterns for various drugs that can be used to characterize drugs for brain deceases and 3) human gait patterns that can reveal various diseases were among these applications. We have demonstrated with considerable success in using unsupervised clustering techniques to analyze genetic and metabolomic data. This includes analysis of drought resistance in wheat [4] and microbial metagenomes [5].We introduced two methods: Near Unsupervised Learning (NUL) and Sub-sample Error Graphs (SEGs) [5] to analyze large amount of data. Self Organizing Map (SOM), which is one of the widely used Unsupervised Neural Networks, has been used as a data-mining tool due to its ability to map high dimensional data into a two dimensional feature map, which is expected to be topology preserving allowing users to visually identify clusters by their topological relationships in terms of their proximity on the map. The Growing Self Organizing Maps (GSOM) further allows the map size to be determined by the algorithm, which relies upon a user set parameter called Spread Factor (SF) [6]. The wide availability of GPUs for affordable prices allows faster comparison of various SOMs with different maps sizes taking away some of the advantages GSOM claimed. Further development of GSOM into a Dynamic SOM Tree exploits the possibility of varying SF to obtain multiple GSOMs from a small number of compact clusters to a large number of sparse clusters [7]. NUL methods can be applied on GSOM using a small number of labels that should be available for every class. This is however not realistic in some applications where the number of classes cannot be explicitly known. We propose a new method call Deep Near Unsupervised Learning (D-NUL), where Dynamic SOM tree is used instead of GSOM and the number of classes are not assumed to be known. The implementation of Dynamic SOM tree methods with varying SF on GPUs will make the computation with D-NUL possible for many problems in big data analytics.

Artificial neural network analysis of noisy visual field data in glaucoma

Unsupervised knowledge discovery using Self Organising Maps (SOM) has been successfully used in obtaining unbiased and visualisable results. A Growing (or Dynamic) Self Organising Maps (GSOM) is an extended version of the original SOM with adaptive map size and controllable spread. In experiments a GSOM usually has considerably higher topographic error than SOM with similar quantisation error. This can be undesirable in cases where, topology preservation is important, therefore in this paper the authors proposed an algorithm to assist the growing of the dynamic self-organising map in achieving better topographic quality whilst maintaining or even improving level of quantisation error. Results have shown improvement of topographic error when comparing to GSOM, and have better topology preservation than non-topologically optimised SOM with similar map size.

Regional flood frequency analysis by combining self-organizing feature map and fuzzy clustering

Background: A Liver tumor is a dangerous disease that may leads to death, the chances of survival will increase when it is detected in early time. Materials and Methods: In this study, clustering Fuzzy c-mean and ''Self-Organization Feature Map'', HSOFM, which is an unsupervised artificial neural network, based on the histogram of the image, are presented for segmenting, isolating, and then extracting tumors and other abnormal regions in liver images of MRI and CT imaging. In an additional processing, morphological operations were used to achieve the complete final extraction of the isolated regions without any extra pixels that do not belong to the abnormal regions. These two methods were applied on five MRI and four CT scan images. The extracted regions surface areas were calculated and compared with the groundtruth, manually extracted mass regions, to check the goodness adequacy of the adopted methods. The work was achieved by Mat Lab programing environment. Results: The percent relative differences of the extracted abnormal regions by implementing the adopted methods with the groundtruth were ranged from 1.108 % to 3.861 % for the images of MRI, while for CT scan images, the percent relative differences were in the range from 0.732 % to 3.456%. By implementing FCM clustering, the percent relative difference ranged from 0.724% to 4.370 % for three MRI images, while for CT scan images, the percent relative difference was 4.327 % for one of the adopted CT images. Results of implementing HSOFM indicate the high-quality performance of this method with 95 % accuracy. Conclusion: From the results we can conclude that there is an appropriate number of nodes and clusters that is a more adequate choice than others depending on the properties’ intensity variance in each processed image. The comparison between the two implemented methods figures out the superiority of the HSOFM method over FCM. There are some limitations, such as the small size of the abnormal regions as well as the interference of the abnormal regions with the normal regions of the similar intensity that require applying advanced enhancement methods as an additional preprocessing step.

This study focused on demonstrating the use of a self-organizing map (SOM) algorithm to elucidate patterns among variables in simulated syngas combustion. The work was implemented in two stages: (1) modelling and simulation of syngas combustion under various feed composition and reactor temperature implemented in AspenPlusTM chemical process simulation software, and (2) pattern recognition among variables using SOM algorithm implemented in MATLAB. The varied levels of feed syngas composition and reactor temperature was randomly sampled from uniform distributions using the Morris screening technique creating four thousand eight hundred simulation conditions implemented in the process simulation which consequently produced a multivariate dataset used in the SOM analysis. Results show that cylindrical SOM topology models the dataset at lower quantization error and topographic error as compared to the rectangular SOM topology indicating suitability of the former for variables pattern elucidation for the simulated combustion. Nonetheless, the variables pattern between component planes from rectangular SOM (9 × 28 grid) and those from cylindrical SOM (9 × 28 grid) are almost similar, indicating that either rectangular or cylindrical architectures may be used for variables pattern analysis. The component planes of process variables from trained SOM are a convenient visualization of the trends across all process variables.

Self Organizing Maps (SOM) are widely used in data mining and high-dimensional data visualization due to its unsupervised nature and robustness. Growing Self Organizing Maps (GSOM) is a variant of SOM algorithm which allows nodes to be grown so that it can represent the input space better. Without using a fixed 2D grid like SOM, GSOM starts with four nodes and keeps track of the quantization error in each node. New nodes are grown from an existing node if its error value exceeds a pre-defined threshold. Ability of the GSOM algorithm to represent input space accurately is vital to extend its applicability to a wider spectrum of problems. This ability can be improved by identifying nodes that represent low probability regions in the input space and removing them periodically from the map. This will improve the homogeneity and completeness of the final clustering result. A new extension to GSOM algorithm based on node deletion is proposed in this paper as a solution to this problem. Furthermore, two new algorithms inspired by cache replacement policies are presented. First algorithm is based on Adaptive Replacement Cache (ARC) and maintains two separate Least Recently Used (LRU) lists of the nodes. Second algorithm is built on Frequency Based Replacement policy (FBR) and maintains a single LRU list. These algorithms consider both recent and frequent trends in the GSOM grid before deciding on the nodes to be deleted. The experiments conducted suggest that the FBR based method for node deletion outperforms the standard algorithm and other existing node deletion methods.

Random forest and Self Organizing Maps application for analysis of pediatric fracture healing time of the lower limb