Locally Excitatory Globally Inhibitory Oscillator Network Research Articles

LEGION-based image segmentation by means of spiking neural networks using normalized synaptic weights implemented on a compact scalable neuromorphic architecture

LEGION (Locally Excitatory, Globally Inhibitory Oscillator Network) topology has demonstrated good capabilities in scene segmentation applications. However, the implementation of LEGION algorithm requires machines with high performance to process a set of complex differential equations limiting its use in practical real-time applications. Recently, several authors have proposed alternative methods based on spiking neural networks (SNN) to create oscillatory neural networks with low computational complexity and highly feasible to be implemented on digital hardware to perform adaptive segmentation of images. Nevertheless, existing SNN with LEGION configuration focus on the membrane model leaving aside the behavior of the synapses although they play an important role in the synchronization of several segments by self-adapting their weights. In this work, we propose a SNN-LEGION configuration along with normalized weight of the synapses to self-adapt the SNN network to synchronize several segments of any size and shape at the same time. The proposed SNN-LEGION method involves a global inhibitor, which is in charge of performing the segmentation process between different objects with different sizes and shapes on time. To validate the proposal, the SNN-LEGION method is implemented on an optimized scalable neuromorphic architecture. Our preliminary results demonstrate that the proposed normalization process of the synaptic weights along with the SNN-LEGION configuration keep the capacity of the LEGION network to separate the segments on time, which can be useful in video processing applications such as vision processing systems for mobile robots, offering lower computational complexity and area consumption compared with previously reported solutions.

Automatic Buildings Extraction From LiDAR Data in Urban Area by Neural Oscillator Network of Visual Cortex

Researchers have extensively applied Locally Excitatory Globally Inhibitory Oscillator Networks (LEGION) for segmentation. These networks are neural oscillator networks based on biological frameworks, in which each oscillator has excitatory lateral connections to the oscillators in its local neighbourhood, as well as a connection to a global inhibitor. In this paper, we develop a modified LEGION segmentation to extract buildings from high-quality digital surface models (DSMs). The extraction is implemented without assumptions on the underlying structures in the DSM data and without prior knowledge of the number of regions. For complex information hidden in the generated DSM of an urban area, grey level co-occurrence matrix homogeneity is used to measure DSM height texture. We then use this homogeneity to distinguish buildings from trees and identify major oscillator blocks in target buildings, instead of using lateral potential. To segment pixels into different groups, we calculate the weight of the global inhibitor (Wz) from DSM complexity. Building boundaries are traced and regularised after extraction from the segmented DSM. A least squares solution with perpendicular constraints for determining regularised rectilinear building boundaries is proposed, and arc line fitting is performed. This paper presents the concept, algorithms, and procedures of the proposed approach. Experimental results on the Vaihingen region studied in the ISPRS test project are also discussed.

LEGION SEMENTATION FOR BUILDING EXTRACTION FROM LIDAR BASED DSM DATA

Abstract. Recently, a neural oscillator network based on biologically framework named LEGION (Locally Excitatory Globally Inhibitory Oscillator Network), which each oscillator has excitatory lateral connections to the oscillators in its local neighbourhood as well as a connection with a global inhibitor, has been applied to segmentation field. The extended LEGION approach is constructed to extract buildings digital surface model (DSM) generated from LiDAR data. This approach is with no assumption about the underlying structures in DSM data and no prior knowledge regarding the number of regions. Instead of using lateral potential to find a major oscillator block in original way, Gray Level Co-occurence Matrix (GLCM) homogeneity measuring DSM height texture is applied to distinguish buildings from trees and assist to find LEGION leaders in building targets. Alongside the DSM height texture attribure, extended LEGION can extract buildings close to trees automatically. Then a solution of least squares with perpendicularity constraints is put forward to determine regularized rectilinear building boundaries, after tracing and connection the rough building boundaries. In general, the paper presents the concept, algorithms and procedures of the approach. It also gives experimental result of Vaihingen A2 region by the ISPRS text project and another result based on a DSM data of suburban area. The experiment result showed that the proposed method can effectively produce more accurate buildings boundary extraction.

A Systematic Approach toward Detection of Seagrass Patches from Hyperspectral Imagery

Changes in the coverage of seagrass populations are considered to be a key indicator of the health and biodiversity of coastal ecosystems. The overall extent of seagrass meadows is declining worldwide, primarily due to human-induced disturbances. In Tampa Bay, Florida, a nearly 35% loss of seagrass coverage occurred from the 1950s to the 2000s. This decline was primarily due to the effects of human population growth. To examine closely the continuing declining trend of this major indicator of the health of coastal ecosystems, a systematic approach for extracting seagrass patches using EO-1 Hyperion hyperspectral imagery has been developed. In our previous work, a method based on Locally Excitatory Globally Inhibitory Oscillator Networks (LEGION) was developed and successfully applied to military object recognition using hyperspectral and multispectral imagery. It showed great potential in target detection of hyperspectral imagery. In this work, it is extend and applied in seagrass extraction. This study includes (a) dimensionality reduction of the hyperspectral data, (b) seagrass extraction using LEGION and four other methods, and (c) analysis and evaluation of the results in an experiment involving two test sites at Tampa Bay, Florida. The results demonstrated that the methodology has the potential to provide timely seagrass coverage information for coastal zone management at greatly increased efficiency.

LEGION-Based Automatic Road Extraction From Satellite Imagery

An automatic method for road extraction from satellite imagery is presented. The core of the proposed method is locally excitatory globally inhibitory oscillator networks (LEGION). The road extraction task is decomposed into three stages. The first stage is image segmentation by LEGION. In the second stage, the medial axis of each segment is computed, and the medial axis points corresponding to narrow regions are selected. The third is the road grouping stage. Alignment-dependent connections between selected points are established, and LEGION is utilized to group well-aligned points, which represent the extracted roads. Due to the selective gating mechanism of LEGION, different roads in an image are grouped separately. Road extraction results on synthetic and real images are presented. A comparison with other methods shows that the proposed method produces very competitive extraction results.

An Extraction method for Water Body of Remote Sensing Image Based on Oscillatory Network

How to avoiding the disadvantage of the pixel-wise image process methods has been a open problem. An object-wise method was proposed to extract the water body of remote sensing image, which is locally excitatory globally inhibitory oscillator networks (LEGION). This oscillation network could bind an object based on the similar property. The oscillation network in this paper used the normalized difference water index (NDWI) to encode the binding of pixels of water body. Because of the good coherence of spectrum of water body on the remote sensing image, the pixels of the water surface have strong couple weight and can overcome the globally inhibition in the network. Additionally the special spectrum property also was considered, it could reduce the disturber of the other ground object which has the similar spectra property. The proposed method combined both pixel-wise and object-wise scale. Thus, the proposed extraction method was robust to noise, and could keep the detail information of water body. The outputs were according to the human perception mechanism and can be used in the higher image analysis and reorganization. The results of experiment validate the effective of the proposed method.

An Approach for Target Detection and Extraction Based on Biological Vision

Abstract Inspired by the mechanism of multi-scale image fusion of insect compound eye, this paper proposed a target detection and extraction method based on insect compound eye and human visual attention mechanism. The main feature of this method is that multi-scale visual attention mechanism is designed for improving the detection accuracy of interested target, meanwhile image is pre-segment based on it, then the target is extracted based on the LEGION (Locally Excitatory Globally Inhibitory Oscillator Networks) selection network model. For multi-scale visual attention mechanism, on intensity channel, this approach adopts multi-scale analysis to get the intensity saliency map on both global-level and local-level; on orientation channel, according to the mechanism of the large vision field of compound eye, directional information is added to improve the orientation sensitivity; on color channel, according to the characteristic of color sensitivity of human vision system, HSI color space is used instead of...

An oscillatory correlation model of visual motion analysis.

We describe and evaluate a model of motion perception based on the integration of information from two parallel pathways: a motion pathway and a luminance pathway. The motion pathway has two stages. The first stage measures and pools local motion across the input animation sequence and assigns reliability indices to these pooled measurements. The second stage groups locations on the basis of these measurements. In the luminance pathway, the input scene is segmented into regions on the basis of similarities in luminance. In a subsequent integration stage, motion and luminance segments are combined to obtain the final estimates of object motion. The neural network architecture we employ is based on LEGION (locally excitatory globally inhibitory oscillator networks), a scheme for feature binding and region labeling based on oscillatory correlation. Many aspects of the model are implemented at the neural network level, whereas others are implemented at a more abstract level. We apply this model to the computation of moving, uniformly illuminated, two-dimensional surfaces that are either opaque or transparent. Model performance replicates a number of distinctive features of human motion perception.

Scene analysis by integrating primitive segmentation and associative memory

Scene analysis is a major aspect of perception and continues to challenge machine perception. This paper addresses the scene-analysis problem by integrating a primitive segmentation stage with a model of associative memory. The model is a multistage system that consists of an initial primitive segmentation stage, a multimodule associative memory, and a short-term memory (STM) layer. Primitive segmentation is performed by a locally excitatory globally inhibitory oscillator network (LEGION), which segments the input scene into multiple parts that correspond to groups of synchronous oscillations. Each segment triggers memory recall and multiple recalled patterns then interact with one another in the STM layer. The STM layer projects to the LEGION network, giving rise to memory-based grouping and segmentation. The system achieves scene analysis entirely in phase space, which provides a unifying mechanism for both bottom-up analysis and top-down analysis. The model is evaluated with a systematic set of three-dimensional (3-D) line drawing objects, which are arranged in an arbitrary fashion to compose input scenes that allow object occlusion. Memory-based organization is responsible for a significant improvement in performance. A number of issues are discussed, including input-anchored alignment, top-down organization, and the role of STM in producing context sensitivity of memory recall.

LEGION—Based Segmentation of Magnetic Resonance Images of the Brain

In this paper a new biologically inspired oscillator network, called Locally Excitatory Globally Inhibitory Oscillator Network (LEGION), is used to perform segmentation on sampled medical image datasets. An image segmentation algorithm that uses LEGION components and an adaptive scheme to choose tolerances is described. The results of the algorithm on MRI medical image datasets, with manual segmentation are obtained and compared. LEGION'S computational and architectural properties make it a promising approach for real-time medical image segmentation using multiple features.

Texture segmentation using Gaussian-Markov random fields and neural oscillator networks

We propose an image segmentation method based on texture analysis. Our method is composed of two parts. The first part determines a novel set of texture features derived from a Gaussian-Markov random fields (GMRF) model. Unlike a GMRF-based approach, our method does not employ model parameters as features or require the extraction of features for a fixed set of texture types a priori. The second part is a 2D array of locally excitatory globally inhibitory oscillator networks (LEGION). After being filtered for noise suppression, features are used to determine the local couplings in the network. When LEGION runs, the oscillators corresponding to the same texture tend to synchronize, whereas different texture regions tend to correspond to distinct phases. In simulations, a large system of differential equations is solved for the first time using a recently proposed method for integrating relaxation oscillator networks. We provide results on real texture images to demonstrate the performance of our method.

Motion segmentation based on motion/brightness integration and oscillatory correlation

A segmentation method based on the integration of motion and brightness is proposed for image sequences. The method is composed of two parallel pathways that process motion and brightness, respectively. Inspired by the visual system, the motion pathway has two stages. The first stage estimates local motion at locations with reliable information. The second stage performs segmentation based on local motion estimates. In the brightness pathway, the input scene is segmented into regions based on brightness distribution. Subsequently, segmentation results from the two pathways are integrated to refine motion estimates. The final segmentation is performed in the motion network based on refined estimates. For segmentation, locally excitatory globally inhibitory oscillator network (LEGION) architecture is employed whereby the oscillators corresponding to a region of similar motion/brightness oscillate in synchrony and different regions attain different phases. Results on synthetic and real image sequences are provided, and comparisons with other methods are made.

On connectedness: a solution based on oscillatory correlation.

A long-standing problem in Neural Comp has been the problem of connectedness, first identified by Minsky and Papert (1969). This problem served as the cornerstone for them to establish analytically that perceptrons are fundamentally limited in computing geometrical (topological) properties. A solution to this problem is offered by a different class of neural networks: oscillator networks. To solve the problem, the representation of oscillatory correlation is employed, whereby one pattern is represented as a synchronized block of oscillators and different patterns are represented by distinct blocks that desynchronize from each other. Oscillatory correlation emerges from LEGION (locally excitatory globally inhibitory oscillator network), whose architecture consists of local excitation and global inhibition among neural oscillators. It is further shown that these oscillator networks exhibit sensitivity to topological structure, which may lay a neurocomputational foundation for explaining the psychophysical phenomenon of topological perception.

Synchrony and desynchrony in integrate-and-fire oscillators.

Due to many experimental reports of synchronous neural activity in the brain, there is much interest in understanding synchronization in networks of neural oscillators and its potential for computing perceptual organization. Contrary to Hopfield and Herz (1995), we find that networks of locally coupled integrate-and-fire oscillators can quickly synchronize. Furthermore, we examine the time needed to synchronize such networks. We observe that these networks synchronize at times proportional to the logarithm of their size, and we give the parameters used to control the rate of synchronization. Inspired by locally excitatory globally inhibitory oscillator network (LEGION) dynamics with relaxation oscillators (Terman & Wang, 1995), we find that global inhibition can play a similar role of desynchronization in a network of integrate-and-fire oscillators. We illustrate that a LEGION architecture with integrate-and-fire oscillators can be similarly used to address image analysis.

Object selection based on oscillatory correlation

One of the classical topics in neural networks is winner-take-all (WTA), which has been widely used in unsupervised (competitive) learning, cortical processing, and attentional control. Owing to global connectivity, WTA networks, however, do not encode spatial relations in the input, and thus cannot support sensory and perceptual processing where spatial relations are important. We propose a new architecture that maintains spatial relations between input features. This selection network builds on Locally Excitatory Globally Inhibitory Oscillator Networks (LEGION) dynamics and slow inhibition. In an input scene with many objects (patterns), the network selects the largest object. This system can be easily adjusted to select several largest objects, which then alternate in time. We analyze the speed of selection, and further show that a two-stage selection network gains efficiency by combining selection with parallel removal of noisy regions. The network is applied to select the most salient object in gray-level images. As a special case, the selection network without local excitation gives rise to a new form of oscillatory WTA.

Range image segmentation using a relaxation oscillator network

A locally excitatory globally inhibitory oscillator network (LEGION) is constructed and applied to range image segmentation, where each oscillator has excitatory lateral connections to the oscillators in its local neighborhood as well as a connection with a global inhibitor. A feature vector, consisting of depth, surface normal, and mean and Gaussian curvatures, is associated with each oscillator and is estimated from local windows at its corresponding pixel location. A context-sensitive method is applied in order to obtain more reliable and accurate estimations. The lateral connection between two oscillators is established based on a similarity measure of their feature vectors. The emergent behavior of the LEGION network gives rise to segmentation. Due to the flexible representation through phases, our method needs no assumption about the underlying structures in image data and no prior knowledge regarding the number of regions. More importantly, the network is guaranteed to converge rapidly under general conditions. These unique properties may lead to a real-time approach for range image segmentation in machine perception.

Segmentation of medical images using LEGION.

Advances in visualization technology and specialized graphic workstations allow clinicians to virtually interact with anatomical structures contained within sampled medical-image datasets. A hindrance to the effective use of this technology is the difficult problem of image segmentation. In this paper, we utilize a recently proposed oscillator network called the locally excitatory globally inhibitory oscillator network (LEGION) whose ability to achieve fast synchrony with local excitation and desynchrony with global inhibition makes it an effective computational framework for grouping similar features and segregating dissimilar ones in an image. We extract an algorithm from LEGION dynamics and propose an adaptive scheme for grouping. We show results of the algorithm to two-dimensional (2-D) and three-dimensional (3-D) (volume) computerized topography (CT) and magnetic resonance imaging (MRI) medical-image datasets. In addition, we compare our algorithm with other algorithms for medical-image segmentation, as well as with manual segmentation. LEGION's computational and architectural properties make it a promising approach for real-time medical-image segmentation.

Fast numerical integration of relaxation oscillator networks based on singular limit solutions

Abstract-Relaxation oscillations exhibiting more than one time scale arise naturally from many physical systems. When relaxation oscillators are coupled in a way that resembles chemical synapses, we propose a fast method to numerically integrate such networks. The numerical technique, called the singular limit method, is derived from analysis of relaxation oscillations in the singular limit. In such limit, system evolution gives rise to time instants at which fast dynamics takes place and intervals between them during which slow dynamics takes place. A full description of the method is given for a locally excitatory globally inhibitory oscillator network (LEGION), where fast dynamics, characterized by jumping which leads to dramatic phase shifts, is captured in this method by iterative operation and slow dynamics is entirely solved. The singular limit method is evaluated by computer experiments, and it produces remarkable speedup compared to other methods of integrating these systems. The speedup makes it possible to simulate large-scale oscillator networks.

Image segmentation based on oscillatory correlation.

We study the image segmentation on the basis of locally excitatory, globally inhibitory oscillator networks (LEGION), whereby the phases of oscillators encode the binding of pixels. We introduce a lateral potential for each oscillators so that only oscillators with strong connections from their neighborhood can develop high potentials. Based on the concept of the lateral potential, a solution to remove noisy regions in an image is proposed for LEGION, so that it suppresses the oscillators corresponding to noisy regions but without affecting those corresponding to major regions. We show that the resulting oscillator network separates an image into several major regions, plus a background consisting of all noisy regions, and we illustrate network properties by computer stimulation. The network exhibits a natural capacity in segmenting images. The oscillatory dynamics leads to a computer algorithm, which is applied successfully to segmenting real gray-level images. A number of issues regarding biological plausibility and perceptual organization are discussed. We argue that LEGION provides a novel and effective framework for image segmentation and figure-ground segregation.

Locally excitatory globally inhibitory oscillator networks

A novel class of locally excitatory, globally inhibitory oscillator networks (LEGION) is proposed and investigated. The model of each oscillator corresponds to a standard relaxation oscillator with two time scales. In the network, an oscillator jumping up to its active phase rapidly recruits the oscillators stimulated by the same pattern, while preventing other oscillators from jumping up. Computer simulations demonstrate that the network rapidly achieves both synchronization within blocks of oscillators that are stimulated by connected regions and desynchronization between different blocks. This model lays a physical foundation for the oscillatory correlation theory of feature binding and may provide an effective computational framework for scene segmentation and figure/ground segregation in real time.