Global feature space neural network for active computer vision

We advance new active computer vision algorithms based on the Feature space Trajectory (FST) representations of objects and a neural network processor for computation of distances in global feature space. Our algorithms classify rigid objects and estimate their pose from intensity images. They also indicate how to automatically reposition the sensor if the class or pose of an object is ambiguous from a given viewpoint and they incorporate data from multiple object views in the final object classification. An FST in a global eigenfeature space is used to represent 3D distorted views of an object. Assuming that an observed feature vector consists of Gaussian noise added to a point on the FST, we derive a probability density function for the observation conditioned on the class and pose of the object. Bayesian estimation and hypothesis testing theory are then used to derive approximations to the maximum a posterioriprobability pose estimate and the minimum probability of error classifier. Confidence measures for the class and pose estimates, derived using Bayes theory, determine when additional observations are required, as well as where the sensor should be positioned to provide the most useful information.     

Neural networks discover a near-identity relation to distinguish simple syntactic forms

Computer simulations show that an unstructured neural-network model [Shultz, T. R., & Bale, A. C. (2001). Infancy, 2, 501–536] covers the essential features␣of infant learning of simple grammars in an artificial language [Marcus, G. F., Vijayan, S., Bandi Rao, S., & Vishton, P. M. (1999). Science, 283, 77–80], and generalizes to examples both outside and inside of the range of training sentences. Knowledge-representation analyses confirm that these networks discover that duplicate words in the sentences are nearly identical and that they use this near-identity relation to distinguish sentences that are consistent or inconsistent with a familiar grammar. Recent simulations that were claimed to show that this model did not really learn these grammars [Vilcu, M., & Hadley, R. F. (2005). Minds and Machines, 15, 359–382] confounded syntactic types with speech sounds and did not perform standard statistical tests of results.     

First and second order dynamics in a hierarchical SOM system for action recognition

Human recognition of the actions of other humans is very efficient and is based on patterns of movements. Our theoretical starting point is that the dynamics of the joint movements is important to action categorization. On the basis of this theory, we present a novel action recognition system that employs a hierarchy of Self-Organizing Maps together with a custom supervised neural network that learns to categorize actions. The system preprocesses the input from a Kinect like 3D camera to exploit the information not only about joint positions, but also their first and second order dynamics. We evaluate our system in two experiments with publicly available datasets, and compare its performance to the performance with less sophisticated preprocessing of the input. The results show that including the dynamics of the actions improves the performance. We also apply an attention mechanism that focuses on the parts of the body that are the most involved in performing the actions.     

一种基于神经网络的可调整模糊系统模型

对基于人工神经网络的模糊系统进行研究,建立了一种可调整的模糊系统模型,并且在智能汽车控制器中得到了运用。通过神经网络对输入的经验值进行学习、调整来获得模糊系统控制的新参数。经计算机仿真,并与传统模糊控制模型相比较,小车绕开障碍的性能得到了明显的改善,是一种比较理想的模糊系统模型。     

Adaptive noise filtering based on artificial hydrocarbon networks: An application to audio signals

Many audio signal applications are corrupted by noise. In particular, adaptive filters are frequently applied to white noise reduction in audio. Recent work provides that there exist some insights on using an artificial intelligence method called artificial hydrocarbon networks (AHNs) for filtering audio signals. Thus, the scope of this paper is to design and implement a novel approach of artificial hydrocarbon networks on adaptive filtering for audio signals. Three experiments were developed. Results demonstrate that AHNs can reduce noise from audio signals. A comparison between the proposed algorithm and a FIR-filter is also provided. The short-time objective intelligibility value (STOI) and the signal-to-noise ratio (SNR) were used for evaluation. At last, the proposed training method for finding the parameters involved in the AHN-filter can also be used in other fields of application.     

An investigation of the modified direction feature for cursive character recognition

This paper describes and analyses the performance of a novel feature extraction technique for the recognition of segmented/cursive characters that may be used in the context of a segmentation-based handwritten word recognition system. The modified direction feature (MDF) extraction technique builds upon the direction feature (DF) technique proposed previously that extracts direction information from the structure of character contours. This principal was extended so that the direction information is integrated with a technique for detecting transitions between background and foreground pixels in the character image.In order to improve on the DF extraction technique, a number of modifications were undertaken. With a view to describe the character contour more effectively, a re-design of the direction number determination technique was performed. Also, an additional global feature was introduced to improve the recognition accuracy for those characters that were most frequently confused with patterns of similar appearance. MDF was tested using a neural network-based classifier and compared to the DF and transition feature (TF) extraction techniques. MDF outperformed both DF and TF techniques using a benchmark dataset and compared favourably with the top results in the literature. A recognition accuracy of above 89% is reported on characters from the CEDAR dataset.     

神经网络在机器人视觉图像命令识别中的应用

在智能机器人技术中,视觉识别是关键.在智能机器人视觉系统获得的图像中,由于图像倾斜而造成的识别错误是视觉识别难以解决的问题.针对机器人所要完成的具体任务,对机器人的视觉识别问题进行探讨,为实现机器人对图像命令的识别,首先对机器人视觉系统获得的倾斜图像,采用Hough变换进行倾斜度检测并进行校正,然后采用人工神经网络法进行识别,根据识别结果对机器人的下一步运动进行决策与控制,达到了预期的目的.实验结果表明,该方法具有较高的识别率.     

《VFP数据库程序设计》网络个性化教学研究

用好网络教学平台是提高教学质量的重要途径之一.但现有的大多数网络课程在个性化教学方面存在诸多问题.而课程领域知识的有效组织和智能导航是实现个性化教学的关键.提出用本体概念网组织课程领域知识,并在此基础上提出了一种智能导航算法.为基于概念网的个性化网络教学系统的设计与开发提供基础.     

概念获取的框架表征重述方法*

基于知识的问题求解需要一个丰富而相对完备的信念系统,尤其是当任务具有领域非限定特征时。经典知识工程关于知识获取、表示与使用的方法只能适应领域受限问题,因为它不是从概念发展的角度来建构概念系统的,面临着Framework Problem。针对认知心理学对概念系统发展与表征问题研究有一定的认知深度同时又缺乏系统构造与实现机制研究的情况,提出了一种基于框架表征的概念系统表征与发展方法,详细研究了概念在内的隐水平I、外显水平E1、外显水平E2和外显水平E3上的表征和发展过程。这一研究对于提高基于知识的系统推理能力和问题求解能力具有显著意义。     

A general framework for reasoning about change

The capability to represent and use concepts like time and events in computer science is essential to solve a wide class of problems characterized by the notion of change. Real-time, databases and multimedia are just a few of several areas which needs good tools to deal with time. Another area where this concepts are essential is artificial intelligence because an agent must be able to reason about a dynamic environment. In this work a formalism is proposed which allows the representation and use of several features that had been recognized as useful in the attempts to solve such class of problems. A general framework based on a many-sorted logic is proposed centering our attention in issues such as the representation of time, actions, properties, events and causality. The proposal is compared with related work from the temporal logic and artificial intelligence areas. This work complements and enhances previously related efforts on formalizing temporal concepts with the same purpose. Juan Carlos Augusto, Ph.D.: He is a Lecturer in the Department of Computer Science at Universidad Nacional del Sur (Argentina), where he graduated as Licenciado en Ciencias de la Computacion and Doctor en Ciencias de la Computacion. Currently on leave in the Department of Electronics and Computer Science, University of Southampton (United Kingdom). His research interests are focused in the dynamic aspects of computing systems. This involves solving conceptual problems related to the specification of time and change and designing tools to improve systems in several areas of computer science, such as artificial intelligence, databases, multimedia, software verification and real-time systems. He has been conducting research on temporal representation and reasoning since 1993. Throughout these years he had the opportunity to contribute to several research projects as a researcher and has head or co-head of research groups. Other activities and contributions to highlight are the organization of international events, editorial work and supervision of postgraduate students, all of which contributes to the generation and dissemination of knowledge about the dynamic aspects of computing systems.     

A modified model for the Lobula Giant Movement Detector and its FPGA implementation

Bio-inspired vision sensors are particularly appropriate candidates for navigation of vehicles or mobile robots due to their computational simplicity, allowing compact hardware implementations with low power dissipation. The Lobula Giant Movement Detector (LGMD) is a wide-field visual neuron located in the Lobula layer of the Locust nervous system. The LGMD increases its firing rate in response to both the velocity of an approaching object and the proximity of this object. It has been found that it can respond to looming stimuli very quickly and trigger avoidance reactions. It has been successfully applied in visual collision avoidance systems for vehicles and robots. This paper introduces a modified neural model for LGMD that provides additional depth direction information for the movement. The proposed model retains the simplicity of the previous model by adding only a few new cells. It has been simplified and implemented on a Field Programmable Gate Array (FPGA), taking advantage of the inherent parallelism exhibited by the LGMD, and tested on real-time video streams. Experimental results demonstrate the effectiveness as a fast motion detector.     

Spectral recognition using a modified Eckhorn neural network model

Studies of the visual cortex of the cat highlight the role of temporal processing using synchronous oscillations for object identification. In this paper, the original neural network model of Eckhorn has been modified according to the proposal of Johnson and others and used for spectral recognition. The method developed turns out to be a much simpler, faster and elegant way of spectral recognition than reported elsewhere.     

用细胞神经网络提取二值与灰度图象边缘

边缘是图象的重要特征,采用细胞神经网络提取图象边缘时,网络参数的选择是一个重要问题。为了能够有效地提取图象边缘,基于高通滤波模板,选择了细胞神经网络的一组简单易行的参数,首先将其用于检测二值图象边缘,再在此基础上,通过综合灰度值各位面边缘检测的结果提取出灰度图象的边缘。与传统边缘提取方法Sobel和Log方法的比较可见,该方法是有效的,并且由于细胞神经网络具有高速并行运算、便于硬件实现等特点,因此使其在图象实时处理中具有更大的潜力。     

Edge detection using a neural network

Artificial neural networks have been shown to perform well in many image processing applications such as coding, pattern recognition and texture segmentation. In a typical multi-layer model of this class, neurons in each layer are linked by synaptic weights to a receptive field region in the layer below it. The input image itself is linked to the lowest layer. We propose here a two stage encoder-detector network for edge detection. The single layer encoder stage, trained in a competitive mode, compresses data from an input receptive field and drives a back-propagation-trained detector network whose two outputs represent components of an edge vector. Experimental results show that for the case of step edges in noisy images, the performance of the neural edge detector is comparable to that of the Canny detector.     

A PDP constructive algorithm for system modelling

A new constructive algorithm is presented for building neural networks that learn to reproduce output temporal sequences based on one or several input sequences. This algorithm builds a network for the task of system modelling, dealing with continuous variables in the discrete time domain. The constructive scheme makes it user independent. The network's structure consists of an ordinary set and a classification set, so it is a hybrid network like that of Stokbro et al. [6], but with a binary classification. The networks can easily be interpreted, so the learned representation can be transferred to a human engineer, unlike many other network models. This allows for a better understanding of the system structure than just its simulation. This constructive algorithm limits the network complexity automatically, hence preserving extrapolation capabilities. Examples with real data from three totally different sources show good performance and allow for a promising line of research.     

基于人工神经网络的负相关学习研究

人工神经网络集成技术是神经计算技术的一个研究热点,在许多领域中已经有了成熟的应用.神经网络集成是用有限个神经网络对同一个问题进行学习,集成在某输入示例下的输出由构成集成的各神经网络在该示例下的输出共同决定.负相关学习法是一种神经网络集成的训练方法,它鼓励集成中的不同个体网络学习训练集的不同部分,以使整个集成能更好地学习整个训练数据.改进的负相关学习法是在误差函数中使用一个带冲量的BP算法,给合了原始负相关学习法和带冲量的BP算法的优点,使改进的算法成为泛化能力强、学习速度快的批量学习算法.     

背景格的若干性质及其完备性证明*

背景在概念图理论中至关重要,其作用尤其体现在信息组织方面。借鉴形式概念分析的理论,用二元组表示形式背景,能够体现出形式背景的内涵与外延的统一。背景格是形式背景的一种组织形式,反映了形式背景之间的蕴涵关系,完备的背景格有着广阔的应用前景。介绍了背景格的定义及其构造方法,并给出了背景格的若干性质以及完备性证明。     

Fast evaluation of neural networks via confidence rating

Neural networks have become very useful tools for input–output knowledge discovery. However, some of the most powerful schemes require very complex machines and, thus, a large amount of calculation. This paper presents a general technique to reduce the computational burden associated with the operational phase of most neural networks that calculate their output as a weighted sum of terms, which comprises a wide variety of schemes, such as Multi-Net or Radial Basis Function networks. Basically, the idea consists on sequentially evaluating the sum terms, using a series of thresholds which are associated with the confidence that a partial output will coincide with the overall network classification criterion. Furthermore, we design some procedures for conveniently sorting out the network units, so that the most important ones are evaluated first. The possibilities of this strategy are illustrated with some experiments on a benchmark of binary classification problems, using RealAdaboost and RBF networks, which show that important computational savings can be achieved without significant degradation in terms of recognition accuracy.