Multi-stream pose convolutional neural networks for human interaction recognition in images

Recognizing human interactions in still images is quite a challenging task since compared to videos, there is only a glimpse of interaction in a single image. This work investigates the role of human poses in recognizing human–human interactions in still images. To this end, a multi-stream convolutional neural network architecture is proposed, which fuses different levels of human pose information to recognize human interactions better. In this context, several pose-based representations are explored. Experimental evaluations in an extended benchmark dataset show that the proposed multi-stream pose Convolutional Neural Network is successful in discriminating a wide range of human–human interactions and human poses when used in conjunction with the overall context provides discriminative cues about human–human interactions.     

Application of neural networks for sea ice classification inpolarimetric SAR images

Several automatic methods have been developed to classify sea ice types from fully polarimetric synthetic aperture radar (SAR) images, and these techniques are generally grouped into supervised and unsupervised approaches. In previous work, supervised methods have been shown to yield higher accuracy than unsupervised techniques, but suffer from the need for human interaction to determine classes and training regions. In contrast, unsupervised methods determine classes automatically, but generally show limited ability to accurately divide terrain into natural classes. In this paper, a new classification technique is applied to determine sea ice types in polarimetric and multifrequency SAR images, utilizing an unsupervised neural network to provide automatic classification, and employing an iterative algorithm to improve the performance. The learning vector quantization (LVQ) is first applied to the unsupervised classification of SAR images, and the results are compared with those of a conventional technique, the migrating means method. Results show that LVQ outperforms the migrating means method, but performance is still poor. An iterative algorithm is then applied where the SAR image is reclassified using the maximum likelihood (ML) classifier. It is shown that this algorithm converges, and significantly improves classification accuracy. The new algorithm successfully identifies first-year and multiyear sea ice regions in the images at three frequencies. The results show that L- and P-band images have similar characteristics, while the C-band image is substantially different. Classification based on single features is also carried out using LVQ and the iterative ML method. It is found that the fully polarimetric classification provides a higher accuracy than those based on a single feature. The significance of multilook classification is demonstrated by comparing the results obtained using four-look and single-look classifications     

Minimal radial basis function neural networks for nonlinear channelequalisation

The results of linear and nonlinear channel equalisation in data communications are presented, using a previously developed minimal radial basis function neural network structure, referred to as the minimal resource allocation network (MRAN). The MRAN algorithm uses online learning, and has the capability to grow and prune the RBF network's hidden neurons ensuring a parsimonious network structure. Compared to earlier methods, the proposed scheme does not have to estimate the channel order first, and fix the model parameters. Results showing the superior performance of the MRAN algorithm for two linear channels (minimum and non-minimum phase) for 2PAM signalling, and three nonlinear channels for 2PAM and 4QAM signalling, are presented     

Flood forecasting using radial basis function neural networks

A radial basis function (RBF) neural network (NN) is proposed to develop a rainfall-runoff model for three-hour-ahead flood forecasting. For faster training speed, the RBF NN employs a hybrid two-stage learning scheme. During the first stage, unsupervised learning, fuzzy min-max clustering is introduced to determine the characteristics of the nonlinear RBFs. In the second stage, supervised learning, multivariate linear regression is used to determine the weights between the hidden and output layers. The rainfall-runoff relation can be considered as a linear combination of some nonlinear RBFs. Rainfall and runoff events of the Lanyoung River collected during typhoons are used to train, validate,and test the network. The results show that the RBF NN can be considered a suitable technique for predicting flood flow     

Terrain classification in SAR images using principal componentsanalysis and neural networks

The development of a neural-network-based classifier for classifying three distinct scenes (urban, park, and water) from several polarized SAR images of the San Francisco Bay area is discussed. The principal components (PC) scheme or Karhunen-Loeve transform is used to extract the salient features of the input data, and to reduce the dimensionality of the feature space prior to the application to the neural networks. Using the PC scheme along with the polarized images used in the present study led to substantial improvements in the classification rates when compared with previous studies. When a combined polarization architecture was used, the classification rate for water, urban, and park areas improved to 100%, 98.7%, and 96.1%, respectively     

基于RBF神经网络的人类活动识别

为了实现对智能家居中人类日常生活活动（ADLS）的识别,将使用径向基函数RBF神经网络来进行人类活动的识别.并使用志愿者在智能家居试验台执行活动搜集到的数据对算法的准确率进行评估.实验结果表明,选择合适的特征量和参数,相比于隐含马尔科夫模型径向基函数神经网络人类活动的识别方面显示了较高的准确率.     

Environment-adaptation mobile radio propagation prediction usingradial basis function neural networks

This paper investigates the application of a radial basis function (RBF) neural network to the prediction of field strength based on topographical and morphographical data. The RBF neural network is a two-layer localized receptive field network whose output nodes from a combination of radial activation functions computed by the hidden layer nodes. Appropriate centers and connection weights in the RBF network lead to a network that is capable of forming the best approximation to any continuous nonlinear mapping up to an arbitrary resolution. Such an approximation introduces best nonlinear approximation capability into the prediction model in order to accurately predict propagation loss over an arbitrary environment based on adaptive learning from measurement data. The adaptive learning employs hybrid competitive and recursive least squares algorithms. The unsupervised competitive algorithm adjusts the centers while the recursive least squares (RLS) algorithm estimates the connection weights. Because these two learning rules are both linear, rapid convergence is guaranteed. This hybrid algorithm significantly enhances the real-time or adaptive capability of the RBF-based prediction model. The applications to Okumura's (1968) data are included to demonstrate the effectiveness of the RBF neural network approach     

Parallel design of convolutional neural networks for remotesensing images object recognition based ondata-driven array processor

Object recognition in very high-resolution remote sensing images is a basic problem in the field of aerial and satellite image analysis. With the development of sensor technology and aerospace remote sensing technology, thequality and quantity of remote sensing images are improved. Traditional recognition methods have a certainlimitation in describing higher-level features, but object recognition method based on convolutional neural network(CNN) can not only deal with large scale images, but also train features automatically with high efficiency. It ismainly used on object recognition for remote sensing images. In this paper, an AlexNet CNN model is trained using2 100 remote sensing images, and correction rate can reach 97.6% after 2 000 iterations. Then based on trainedmodel, a parallel design of CNN for remote sensing images object recognition based on data-driven array processor(DDAP) is proposed. The consuming cycles are counted. Simultaneously, the proposed architecture is realized onXilinx V6 development board, and synthesized based on SMIC 130 nm complementary metal oxid semiconductor(CMOS) technology. The experimental results show that the proposed architecture has a certain degree ofparallelism to achieve the purpose of accelerating calculations.     

Application of wavelet transforms and neural networks to the recognition and classification of blemishes

A blemish is a stain or a damage mark on the surface of a product that is unsightly and which may thereby render the product unacceptable. Research is described which seeks to recognise the presence of blemishes and to classify them both in terms of the type of damage sustained and the extent of the damage. This will permit analysis of the cause of the blemishes and the development of ways of preventing marking in the future. It will also permit decisions to be made regarding the acceptability of the product and whether it should be withdrawn or, where appropriate, repaired. The research has led to a new approach to the derivation of shift–invariance by using an overcomplete wavelet transform. The classification performances of a complex orthogonal estimation algorithm, the Fourier transform and the wavelet transform are presented and compared.     

Optimization neural networks for the segmentation of magnetic resonance images

The application of the Hopfield neural network for the multispectral unsupervised classification of MR images is reported. Winner-take-all neurons were used to obtain a crisp classification map using proton density-weighted and T(2)-weighted images in the head. The preliminary studies indicate that the number of iterations needed to reach ;good' solutions was nearly constant with the number of clusters chosen for the problem.     

Design of neural networks for classification of remotely sensed imagery

As currently planned, future Earth remote sensing platforms (i.e., Earth Observing System [EOS]) will be capable of generating data at a rate of over 50 Megabits per second. To address this issue the Intelligent Data Management (IDM) project at NASA/GSFC has prototyped an Intelligent Information Fusion System (IIFS) that uses backpropagation neural networks for the classification of remotely sensed imagery. This is part of the IDM strategy of providing archived data to a researcher through a variety of discipline-specific indices. In this paper we discuss classification accuracies of a backpropagation neural network and compare it with a maximum likelihood classifier (MLC) with multivariate normal class models. We have found that, because of its nonparametric nature, the neural network outperforms the MLC in this area. In addition, we discuss techniques for constructing optimal neural nets on parallel hardware like the MasPar MP-1 currently at NASA/GSFC. Other important discussions are centered around training and classification times of the two methods, and sensitivity to the training data. Finally we discuss future work in the area of classification and neural nets.     

Obstacle detection in single images with deep neural networks

Obstacle detection in single images is a challenging problem in autonomous navigation on low-cost condition. In this paper, we introduce an approach for obstacle detection in single images with deep neural networks. We propose the followings: (1) a deep model combined with other deep neural network for obstacle detection; (2) a method to segment obstacles and infer their depths. Among others, both local and global information are generated in our method for better classification and portability. Experiments are performed on the open datasets and images captured by our autonomous vehicle. The results show that our method is effective in both obstacle detection and depth inference.     

Radial basis function network assisted space-time equalisation for dispersive fading environments

A novel radial basis function network assisted decision-feedback aided space-time equaliser designed for receivers employing multiple antennas is presented. The proposed receiver structure outperforms the linear minimum mean-squared error benchmarker and is less sensitive to both error propagation and channel estimation errors.     

Multi-level fusion with deep neural networks for multimodal sentiment classification

The task of multimodal sentiment classification aims to associate multimodal information, such as images and texts with appropriate sentiment polarities. There are various levels that can affect human sentiment in visual and textual modalities. However, most existing methods treat various levels of features independently without having effective method for feature fusion. In this paper, we propose a multi-level fusion classification (MFC) model to predict the sentiment polarity based on the fusing features from different levels by exploiting the dependency among them. The proposed architecture leverages convolutional neural networks ( CNNs) with multiple layers to extract levels of features in image and text modalities. Considering the dependencies within the low-level and high-level features, a bi-directional (Bi) recurrent neural network (RNN) is adopted to integrate the learned features from different layers in CNNs. In addition, a conflict detection module is incorporated to address the conflict between modalities. Experiments on the Flickr dataset demonstrate that the MFC method achieves comparable performance compared with strong baseline methods.     

Application of neural networks to radar image classification

A number of methods have been developed to classify ground terrain types from fully polarimetric synthetic aperture radar (SAR) images, and these techniques are often grouped into supervised and unsupervised approaches. Supervised methods have yielded higher accuracy than unsupervised techniques, but suffer from the need for human interaction to determine classes and training regions. In contrast, unsupervised methods determine classes automatically, but generally show limited ability to accurately divide terrain into natural classes. In this paper, a new terrain classification technique is introduced to determine terrain classes in polarimetric SAR images, utilizing unsupervised neural networks to provide automatic classification, and employing an iterative algorithm to improve the performance. Several types of unsupervised neural networks are first applied to the classification of SAR images, and the results are compared to those of more conventional unsupervised methods. Results show that one neural network method-Learning Vector Quantization (LVQ)-outperforms the conventional unsupervised classifiers, but is still inferior to supervised methods. To overcome this poor accuracy, an iterative algorithm is proposed where the SAR image is reclassified using a maximum likelihood (ML) classifier. It is shown that this algorithm converges, and significantly improves classification accuracy     

Digitally programmable nonlinear function generator for neural networks

The design of a new digitally programmable analogue circuit well suited for the implementation of several sets of nonlinear functions by approximating them by using a linear combination of sigmoidal terms is presented. The proposed circuit, allowing the building of several collections of nonlinear functions, would be useful in modelling artificial neural networks, fuzzy as well as partial differential equations based circuits     

CMOS current-mode implementation of spatiotemporal probabilistic neural networks for speech recognition

In this paper, a Spatiotemporal Probabilistic Neural Network (SPNN) is proposed for spatiotemporal pattern recognition. This new model is developed by applying the concept of Gaussian density function to the network structure of the SPR (Spatiotemporal Pattern Recognition). The main advantages of this model include faster training and recalling process for patterns. In addition, the overall architecture is also simple, modular, regular, locally connected, and suitable for VLSI implementation. One set of independent speaker isolated (Mandarin digit) speech database is used as an example to demonstrate the superiority of the neural networks for spatiotemporal pattern recognition. The testing result with a reduced error rate of 7% shows that the SPNN is very attractive and effective for practical applications. p ]The CMOS current-mode IC technology is used to implement the SPNN to achieve the objective of minimum classification error in a more direct manner. In this design, neural computation is performed in analog circuits while template information is stored in digital circuits. The prototyping speech recognition processor for the 12th LPC calculation is designed by 1.2μm CMOS technology. The HSPICE simulation results are also presented, which verifies the function of the designed neural system.     

基于两种神经网络的SAR图像多尺度分割与分析

基于多尺度回归技术和神经网络提出两种合成孔径雷达(SAR)图像分割的新方法.首先利用多尺度自回归模型(MAR)来描述SAR图像不同尺度间的统计相依性,以此提取SAR图像的多尺度统计特征;然后分别构造自组织特征映射网络和概率神经网络,并利用统计特征作为输入训练两种网络,实现SAR图像的分割.最后通过实验对这两种方法以及其他方法之间进行比较、分析,结果表明本文提出的两种方法的实验结果比较理想.     

基于径向基函数网络的模拟电路故障诊断设计

模拟电路的固有特点使其故障诊断较数字电路困难.相对于BP网络,RBF神经网络具有最佳逼近性能且收敛快、无局部极小,可引入解决上述困难.根据具体电路,定义故障,选定测试点,确定网络结构,用Pspice获得训练样本,经过训练得到RBF网络.网络的输入为从测试点得到的输入向量,输出为对应的故障.为了验证网络的泛化性能,对每种...