逼近散乱点云数据的三角形网格精确剖分

基于自组织特征映射神经网络构建的三角形网格模型可以实现测量点云 压缩后的Delaunay 三角逼近剖分,但该模型存在逼近误差和边缘误差。为减小三角形网格 的逼近误差和边缘误差,构建了精确逼近的三角形网格模型。首先采用整个测量点云,对三 角形网格模型中的所有神经元进行整体训练;然后对三角形网格中的网格神经元的位置权 重,沿网格顶点法矢方向进行修正;最后采用测量点云中的边界点集,对三角形网格模型中 的网格边界神经元进行训练。算例表明,应用该模型,可以有效减小三角形网格的边缘误差, 三角形网格逼近散乱点云的逼近精度得到大幅提高并覆盖散乱点云整体分布范围。     

Neurocomputing Model for Computation of an Approximate Convex Hull of a Set of Points and Spheres

In this letter, a two-layer neural network is proposed for computation of an approximate convex hull of a set of given points in 3-D or a set of spheres of different sizes. The algorithm is designed based on an elegant concept-shrinking of a spherical rubber balloon surrounding the set of objects in 3-D. Logically, a set of neurons is orderly placed on a spherical mesh i.e., on a rubber balloon surrounding the objects. Each neuron has a parameter vector associated with its current position. The resultant force of attraction between a neuron and each of the given points/objects, determines the direction of a movement of the neuron lying on the rubber balloon. As the network evolves, the neurons (parameter vectors) approximate the convex hull more and more accurately     

A Simple Neural Network Pruning Algorithm with Application to Filter Synthesis

This paper describes an approach to synthesizing desired filters using a multilayer neural network (NN). In order to acquire the right function of the object filter, a simple method for reducing the structures of both the input and the hidden layers of the NN is proposed. In the proposed method, the units are removed from the NN on the basis of the influence of removing each unit on the error, and the NN is retrained to recover the damage of the removal. Each process is performed alternately, and then the structure is reduced. Experiments to synthesize a known filter were performed. By the analysis of the NN obtained by the proposed method, it has been shown that it acquires the right function of the object filter. By the experiment to synthesize the filter for solving real signal processing tasks, it has been shown that the NN obtained by the proposed method is superior to that obtained by the conventional method in terms of the filter performance and the computational cost.     

逼近点云的三角形网格的边缘误差减小研究

基于SOFM神经网络构建的三角形网格模型可以实现测量点云压缩后的Delaunay三角逼近剖分,但该模型存在边缘误差。为减小三角形网格的边缘误差,改进了三角形网格模型的训练模式,提出了3步训练模式。第1步采用整个测量点云,对三角形网格模型中的所有神经元进行整体训练;第2步采用测量点云中的边界点集,对三角形网格模型中的网格边界神经元进行训练;第3步采用边界点集中的角点点集,对与边界角点匹配最佳的网格边界神经元进行训练。算例表明,应用该训练模式,可以有效减小三角形网格的边缘误差,三角形网格逼近散乱点云的逼近精度得到提高并覆盖散乱点云整体分布范围。     

基于SOM的散乱数据点集的B样条曲面重建

利用自组织映射神经网络(SOM)技术对散乱数据点集进行B样条曲面重建时,往往存在网络学习时间过长和学习效果不理想等问题。提出了一种新的神经元初始化方法和分块学习算法,该算法首先运用主元素分析方法(PCA)对散乱数据进行分块,将拓扑结构为四边形的输出层神经元初始化在每块散乱数据的最小二乘平面上进行网络学习和训练,将分块学习得到的各网格曲面拼接成一个整体;然后对该整体网格曲面的边界和内部单独学习,得到一张逼近待重建曲面的双线性B样条曲面;最后对该B样条曲面误差进行了修正。实例证明,该算法可以明显地减少SOM网络学习时间,并改善网络学习效果。     

A Neural Network with Evolutionary Neurons

A neural network, combining evolution and learning is introduced. The novel feature of the proposed network is the evolutionary character of its neurons. The argument of the transfer function performed by the neurons in the network is neither a linear nor polynomial function of the inputs to the neuron, but an unknown general function P(·). The adequate functional form P(·) for each neuron, is achieved during the learning period by means of genetic programming. The proposed neural network is applied to the problem domain of time series prediction of the Mackey-Glass delay differential equation. Simulation results indicate that the new neural network is effective.     

Fast Mesh Reconstruction from Single View Based on GCN and Topology Modification

3D reconstruction based on single view aims to reconstruct the entire 3D shape of an object from one perspective. When existing methods reconstruct the mesh surface of complex objects, the surface details are difficult to predict and the reconstruction visual effect is poor because the mesh representation is not easily integrated into the deep learning framework; the 3D topology is easily limited by predefined templates and inflexible, and unnecessary mesh self-intersections and connections will be generated when reconstructing complex topology, thus destroying the surface details; the training of the reconstruction network is limited by the large amount of information attached to the mesh vertices, and the training time of the reconstructed network is too long. In this paper, we propose a method for fast mesh reconstruction from single view based on Graph Convolutional Network (GCN) and topology modification. We use GCN to ensure the generation of high-quality mesh surfaces and use topology modification to improve the flexibility of the topology. Meanwhile, a feature fusion method is proposed to make full use of the features of each stage of the image hierarchically. We use 3D open dataset ShapeNet to train our network and add a new weight parameter to speed up the training process. Extensive experiments demonstrate that our method can not only reconstruct object meshes on complex topological surfaces, but also has better qualitative and quantitative results.     

基于广义多项式神经网络的点云数据隐式曲面重构方法

针对点云数据的三维重建问题,提出了一种隐曲面重构的广义多项式神经网络新方法.该广义多项式神经网络隐层各神经元激励函数互不相同且线性无关,能够对应地学习点云数据样本中不同的模式,因此,具有较好的学习能力.基于梯度下降法原理,推导了其学习算法.仿真实验尝试将该方法应用于一些简单封闭物体的带噪点云数据隐式曲面重建,取得了较理想的重建质量和去噪效果.     

新型神经元模型及生物神经网络模拟实现算法

根据突触、胞内动作电位和轴突的主要特征,提出了一种新的神经元数学模型,讨论了它是如何来模拟不同类型的神经元,模型分为突触、胞体和轴突三个部分。作为举例,根据同突触可塑性的主要特征,对模型的突触部分进行形式化,根据动作电位发生后胞内电位随时间的变化规律和阈值随时间的变化规律,以及各种内源性和具备抑制后反弹神经元的持征,对模型的胞体部分进行形式化,因此根据这一生物神经元数学模型框架建立的网络可以是异构网络,并且给出了网络模拟实现算法。最后对算法进行复杂度分析,说明算法有较好的可计算性。     

Hybrid neural network model for reconstruction of occluded regions in multi-gait scenario

In real-time situations such as airports, railway stations, and shopping complexes, etc. people walk in a group, and such a group of walking persons termed as multi-gait (MG). In these situations, occlusion is a serious issue that affects gait recognition performance. This issue of occlusion of body regions affects the extraction of gait features for the correct recognition of an object. The objective of this article is to reconstruct occluded regions at the preprocessing stage, which can be used for human recognition in the MG scenario. The article is divided into two folds. Firstly, we segment five regions of interest such as ankle, knee, wrist, elbow, and shoulder. We propose a particle swarm optimization (PSO) based neural network (NN) called hybrid NN to solve this problem. The performance of the proposed model is validated on our constructed dataset (SMVDU-MG), considering two view directions i.e. lateral (left to right) and oblique (left to right diagonal). Experimental results show that the proposed model gives better performance compared to an artificial neural network and alternating least square (ALS) method based on mean square error (MSE) and mean absolute percentage error (MAPE) as a performance measure function.

     

Modular neuron comprises of memristor-based synapse

Design of analog modular neuron based on memristor is proposed here. Since neural networks are built by repetition of basic blocks that are called neurons, using modular neurons is essential for the neural network hardware. In this work modularity of the neuron is achieved through distributed neurons structure. Some major challenges in implementation of synaptic operation are weight programmability, weight multiplication by input signal and nonvolatile weight storage. Introduction of memristor bridge synapse addresses all of these challenges. The proposed neuron is a modular neuron based on distributed neuron structure which it uses the benefits of the memristor bridge synapse for synaptic operations. In order to test appropriate operation of the proposed neuron, it is used in a real-world application of neural network. Off-chip method is used to train the neural network. The results show 86.7 % correct classification and about 0.0695 mean square error for 4-5-3 neural network based on proposed modular neuron.

     

一种以环路作为存储单元的联想记忆神经网络

文章提出了一种新型联想记忆神经网络,每个模式被存储在一个通过网络中所有神经元的环路中,连接包括逻辑状态和一组神经元编号,网络中处理和传递的信号为神经元编号组成的序列,神经元执行一组处理这种序列的符号和逻辑运算;网络记忆容量为2N-2N、完全消除了假模式、同时具有更高的记忆效率和可靠性。     

Generating subdivision surfaces from profile curves

The construction of freeform models has always been a challenging task. A popular approach is to edit a primitive object such that its projections conform to a set of given planar curves. This process is tedious and relies very much on the skill and experience of the designer in editing 3D shapes. This paper describes an intuitive approach for the modeling of freeform objects based on planar profile curves. A freeform surface defined by a set of orthogonal planar curves is created by blending a corresponding set of sweep surfaces. Each of the sweep surfaces is obtained by sweeping a planar curve about a computed axis. A Catmull-Clark subdivision surface interpolating a set of data points on the object surface is then constructed. Since the curve points lying on the computed axis of the sweep will become extraordinary vertices of the subdivision surface, a mesh refinement process is applied to adjust the mesh topology of the surface around the axis points. In order to maintain characteristic features of the surface defined with the planar curves, sharp features on the surface are located and are retained in the mesh refinement process. This provides an intuitive approach for constructing freeform objects with regular mesh topology using planar profile curves.     

On surface reconstruction: A priority driven approach

To reconstruct an object surface from a set of surface points, a fast, practical, and efficient priority driven algorithm is presented. The key idea of the method is to consider the shape changes of an object at the boundary of the mesh growing area and to create a priority queue to the advancing front of the mesh area according to the changes. The mesh growing process is then driven by the priority queue for efficient surface reconstruction. New and practical triangulation criteria are also developed to support the priority driven strategy and to construct a new triangle at each step of mesh growing in real time. The quality and correctness of the created triangles will be guaranteed by the triangulation criteria and topological operations. The algorithm can reconstruct an object surface from unorganized surface points in a fast and reliable manner. Moreover, it can successfully construct the surface of the objects with complex geometry or topology. The efficiency and robustness of the proposed algorithm is validated by extensive experiments.     

基于鸽视顶盖神经元动作电位的图像重建

针对从神经元响应信号中解码视觉输入的问题,提出了一种利用神经元动作电位（Spike）信号重建视觉输入的方法。首先,记录鸽视顶盖（OT）神经元的Spike信号,提取Spike发放率特征;然后,构建线性逆滤波器和卷积神经网络重建模型,实现视觉输入的重建;最后,对通道数量、时间窗口、数据时间长度、延迟时间等参数进行优化。在相同参数条件下,利用线性逆滤波器重建图像的互相关系数达到0.910 7±0.021 9,利用卷积神经网络模型重建图像的互相关系数达到0.927 1±0.017 6。重建结果表明,提取神经元Spike发放率特征并运用线性逆滤波器和卷积神经网络重建模型可以有效重建视觉输入。     

A new approach for epileptic seizure detection using adaptive neural network

This paper presents new neural network models with adaptive activation function (NNAAF) to detect epileptic seizure. Our NNAAF models included three types named as NNAAF-1, NNAAF-2 and NNAAF-3. The activation function of hidden neuron in the model of NNAAF-1 is sigmoid function with free parameters. In the second model, NNAAF-2, activation function of hidden neuron is sum of sigmoid function with free parameters and sinusoidal function with free parameters. In the third model, NNAAF-3, hidden neurons’ activation function is Morlet Wavelet function with free parameters. In addition, we implemented traditional multilayer perceptron (MLP) neural network (NN) model with fixed sigmoid activation function in the hidden layer to compare NNAAF models. The proposed models were trained and tested using 5-fold cross-validation to prove robustness of these models and to find the best model. We achieved 100% average sensitivity, average specificity, and approximately 100% average classification rate in all the models. It was seen that their speeds and the number of maximum iteration were changed for each model. The training time and the number of maximum iteration were reduced on about 50% using NNAAF-3 model. Hence it can be remarkable that NNAAF-3 is more suitable than the other models for real-time application.     

Revised GMDH-type neural network algorithm self-selecting optimum neural network architecture

In this study, the revised group method of data handling (GMDH)-type neural network (NN) algorithm self-selecting the optimum neural network architecture is applied to the identification of a nonlinear system. In this algorithm, the optimum neural network architecture is automatically organized using two kinds of neuron architecture, such as the polynomial- and sigmoid function-type neurons. Many combinations of the input variables, in which the high order effects of the input variables are contained, are generated using the polynomial-type neurons, and useful combinations are selected using the prediction sum of squares (PSS) criterion. These calculations are iterated, and the multilayered architecture is organized. Furthermore, the structural parameters, such as the number of layers, the number of neurons in the hidden layers, and the useful input variables, are automatically selected in order to minimize the prediction error criterion defined as PSS.     

Improved Izhikevich neurons for spiking neural networks

Spiking neural networks constitute a modern neural network paradigm that overlaps machine learning and computational neurosciences. Spiking neural networks use neuron models that possess a great degree of biological realism. The most realistic model of the neuron is the one created by Alan Lloyd Hodgkin and Andrew Huxley. However, the Hodgkin–Huxley model, while accurate, is computationally very inefficient. Eugene Izhikevich created a simplified neuron model based on the Hodgkin–Huxley equations. This model has better computational efficiency than the original proposed by Hodgkin and Huxley, and yet it can successfully reproduce all known firing patterns. However, there are not many articles dealing with implementations of this model for a functional neural network. This study presents a spiking neural network architecture that utilizes improved Izhikevich neurons with the purpose of evaluating its speed and efficiency. Since the field of spiking neural networks has reinvigorated the interest in biological plausibility, biological realism was an additional goal. The network is tested on the correct classification of logic gates (including XOR) and on the iris dataset. Results and possible improvements are also discussed.     

Evolutionary fuzzy hybrid neural network for project cash flow control

This paper develops an evolutionary fuzzy hybrid neural network (EFHNN) to enhance project cash flow management. The developed EFHNN combines neural networks (NN) and high order neural networks (HONN) into a hybrid neural network (HNN), which acts as the major inference engine and operates with alternating linear and nonlinear NN layer connections. Fuzzy logic is employed to sandwich the HNN between a fuzzification and defuzzification layer. The authors developed and applied the EFHNN to sequential cash flow trend problems by fusing HNN, FL, and GA. Results show that the proposed EFHNN can be deployed effectively to sequential cash flow estimation. The performance of linear and nonlinear (high order) neuron layer connectors in the EFHNN was significantly better than the performance achieved by previous models that used singular linear NN. Trained results were used for the prediction and strategic management of project cash flow. The proposed strategy can assist project managers to control project cash flows within the banana envelope of the S-curve to enhance project success.     

确定RBF神经网络参数的新方法

提出一种确定RBF网络隐含层神经元和权值的有效方法。该方法将自动聚类算法与对称距离相结合优化每个隐含层神经元的中心向量；利用伪逆方法确定隐层神经元到输出神经元的权值。实验结果表明：该方法比自动聚类算法有更好的分类能力。