基于线元几何的旋转面点云数据旋转轴提取算法

当前逆向工程CAD建模技术中,由于基于特征的曲面重构技术在精确表达原始模型、还原设计意图以及快速建模中所具备的优势,使其逐渐成为逆向工程领域新的研究热点.为实现海量数据快速特征提取,提出了一种基于线元几何、线性丛的旋转面点云数据旋转轴提取算法.算法首先将三维空间中的点投影到线元空间中,构建线性丛,而后在线性丛上拟合运动方程,通过特征参数计算旋转轴位置.此方法无须精确估算曲面法矢,有效地提高了特征提取速度.在算法中设计并使用了K-Local-RANSAC算法进行快速离散区域扩张并排除外点,保证了算法的健壮性.实验证明,本算法对于包括圆环面在内的旋转面都能得到较好的效果,而且能较好地适用于碎片数据及含有噪声的情况.     

Reverse innovative design — an integrated product design methodology

Today’s product designer is being asked to develop high quality, innovative products at an ever increasing pace. To meet this need, an intensive search is underway for advanced design methodologies that facilitate the acquisition of design knowledge and creative ideas for later reuse. Additionally, designers are embracing a wide range of 3D digital design applications, such as 3D digitization, 3D CAD and CAID, reverse engineering (RE), CAE analysis and rapid prototyping (RP). In this paper, we propose a reverse engineering innovative design methodology called Reverse Innovative Design (RID). The RID methodology facilitates design and knowledge reuse by leveraging 3D digital design applications. The core of our RID methodology is the definition and construction of feature-based parametric solid models from scanned data. The solid model is constructed with feature data to allow for design modification and iteration. Such a construction is well suited for downstream analysis and rapid prototyping. In this paper, we will review the commercial availability and technological developments of some relevant 3D digital design applications. We will then introduce three RE modelling strategies: an autosurfacing strategy for organic shapes; a solid modelling strategy with feature recognition and surface fitting for analytical models; and a curve-based modelling strategy for accurate reverse modelling. Freeform shapes are appearing with more frequency in product development. Since their “natural” parameters are hard to define and extract, we propose construction of a feature skeleton based upon industrial or regional standards or by user interaction. Global and local product definition parameters are then linked to the feature skeleton. Design modification is performed by solving a constrained optimization problem. A RID platform has been developed and the main RE strategies and core algorithms have been integrated into SolidWorks as an add-in product called ScanTo3D. We will use this system to demonstrate our RID methodology on a collection of innovative consumer product design examples.     

精神疲劳识别的可拓模型与策略生成

精神疲劳识别中普遍存在着方法的侵扰性、实时性与识别准确率之间相矛盾的问题.为此,引入可拓理论和方法来建立问题的可拓模型,针对矛盾主体建立关联函数和策略优度函数.结合领域知识,通过拓展分析、可拓变换对矛盾进行转化,生成多种同时满足非侵扰性、实时性和识别准确率的特征和识别策略,并对策略优度进行计算和分析.实验研究验证了本方法的有效性.本研究为计算机模拟人类思维进行算法研究和创新奠定了基础.     

三维形状特征提取技术研究进展

三维形状模型广泛应用于工业设计、教育、生物医药、动画娱乐、文物保护等多个领域中。三维形状模型的特征提取是计算机图形学和模式识别领域的重要问题,近年来受到学者的广泛关注。尤其是具有铰链、关节等的非刚性三维形状通常会发生变形,进一步增加了形状特征提取的难度。主要研究、分析、总结了近几年出现的刚性三维形状和非刚性三维形状的特征提取算法,分析了三维形状特征提取的难点,给出了三维形状特征提取的发展进程。介绍了近年来三维形状特征匹配研究领域中常用的一些测试数据库,重点探讨了非刚性三维形状的特征匹配方法,并展望了三维形状特征提取、特征匹配的未来发展方向。     

A framework for 3D model reconstruction in reverse engineering

We present a framework for 3D model reconstruction, which has potential applications to a spectrum of engineering problems with impacts on rapid design and prototyping, shape analysis, and virtual reality. The framework, composed of four main components, provides a systematic solution to reconstruct geometric model from the surface mesh of an existing object. First, the input mesh is pre-processed to filter out noise. Second, the mesh is partitioned into segments to obtain individual geometric feature patches. Then, two integrated solutions, namely solid feature based strategy and surface feature based strategy, are exploited to reconstruct primitive features from the segmented feature patches. Finally, the modeling operations, such as solid boolean and surface trimming operations, are performed to “assemble” the primitive features into the final model. The concepts of “feature”, “constraint” and “modeling history” are introduced into the entire reconstruction process so that the design intents are retrieved and exhibited in the final model with geometrical accuracy, topological consistency and flexible editability. A variety of industrial parts have been tested to illustrate the effectiveness and robustness of our framework.     

性别分类中头发特征提取方法的研究

针对人脸性别分类问题,提出基于头发特征的分类方法。对人脸重要外部特征之一的头发特征属性的提取与表示以及参数的选取进行分析与研究。在1 680张AR人脸图片上,利用头发特征模型对性别进行分类,将实验结果与基于人脸内部特征的分类结果进行比较,结果表明,采用头发特征的性别分类,其准确度获得平均7.5%的提升,最高准确率达96%。     

性别分类中头发特征提取方法的研究

针对人脸性别分类问题,提出基于头发特征的分类方法。对人脸重要外部特征之一的头发特征属性的提取与表示以及参数的选取进行分析与研究。在1 680张AR人脸图片上,利用头发特征模型对性别进行分类,将实验结果与基于人脸内部特征的分类结果进行比较,结果表明,采用头发特征的性别分类,其准确度获得平均7.5%的提升,最高准确率达96%。     

距离图像局部特征提取方法综述

基于距离图像的三维目标识别是计算机视觉领域的研究热点,而局部特征提取则是实现遮挡和复杂场景下三维目标识别的关键。文中首先介绍距离图像及其表示形式,详细分析法向量、曲率和形状索引等微分几何属性。进而将局部特征检测方法分类为固定尺度和自适应尺度方法,将局部特征描述方法分类为基于深度信息、基于点云空间分布和基于几何属性分布的方法,并对各种具体算法进行阐述、分析和定性评价。最后对现有方法进行归纳总结,并指出所面临的挑战及进一步研究的方向。     

基于双重优化的卷积神经网络图像识别算法*

为了进一步提高卷积神经网络算法的收敛速度和识别精度,提出基于双重优化的卷积神经网络图像识别算法。在构建卷积神经网络的过程中,针对特征提取和回归分类建立双重优化模型,实现对卷积与全连接过程的集成优化,并与局部优化算法对比,分析各算法的识别率和收敛速度的差异。在手写数字集和人脸数据集上的实验表明,双重优化模型可以在较大程度上提高卷积神经网络的收敛速度和识别精度,并且这种优化策略可以进一步拓展到其它与卷积神经网络相关的深度学习算法中。     

A cascade face recognition system using hybrid feature extraction

A novel cascade face recognition system using hybrid feature extraction is proposed. Three sets of face features are extracted. The merits of Two-Dimensional Complex Wavelet Transform (2D-CWT) are analyzed. For face recognition feature extraction, it has proved that 2D-CWT compares favorably with the traditionally used 2D Gabor transform in terms of the computational complexity and features? stability. The proposed recognition system congregates three Artificial Neural Network classifiers (ANNs) and a gating network trained by the three feature sets. A computationally efficient fitness function of the genetic algorithms is proposed to evolve the best weights of the ensemble classifier. Experiments demonstrated that the overall recognition rate and reliability have been significantly improved in both still face recognition and video-based face recognition.     

Towards automatic generation of as-built BIM: 3D building facade modeling and material recognition from images

As-built building information model(BIM) is an urgent need of the architecture, engineering, construction and facilities management(AEC/FM) community. However, its creation procedure is still labor-intensive and far from maturity. Taking advantage of prevalence of digital cameras and the development of advanced computer vision technology, the paper proposes to reconstruct a building facade and recognize its surface materials from images taken from various points of view. These can serve as initial steps towards automatic generation of as-built BIM. Specifically, 3D point clouds are generated from multiple images using structure from motion method and then segmented into planar components, which are further recognized as different structural components through knowledge based reasoning. Windows are detected through a multilayered complementary strategy by combining detection results from every semantic layer. A novel machine learning based 3D material recognition strategy is presented. Binary classifiers are trained through support vector machines. Material type at a given 3D location is predicted by all its corresponding 2D feature points.Experimental results from three existing buildings validate the proposed system.     

大口径光学组件面形检测系统关键算法研究

大口径光学组件面形检测系统中,激光光斑的分割、质心提取和背表面光斑的自动剔除算法关系到系统的检测精度。针对系统中多路光斑光强差异比较大的现象,提出了两次分割法,有效定位各光斑区域;通过比较不同质心提取算法的稳定性,选取灰度重心法提取激光光斑质心;采取基于特征的分类技术识别元件前表面反射光斑。实验结果表明,算法能精确稳定地提取光斑质心和有效识别前表面光斑,该方法已商用于光学组件面形检测系统。     

Topology based 2D engineering drawing and 3D model matching for process plant

Process plant models mainly include 3D models and 2D engineering drawings. Matching calculation between these CAD models has wide applicability in model consistency check and retrieval. In process plant, engineering design standards make 2D engineering drawing and 3D model differ in geometry, proportion and structure, leading to the inapplicability of current shape-feature based matching approaches. Since connection relationships between components are the core of a process plant, a topology based algorithm is proposed. Firstly, by exploiting components as vertices and relationships as edges, both 2D engineering drawing and 3D model are preprocessed into graph structures. Then each model’s relationship types are extracted from the graph. Finally, regarding the extracted relationship types as primary feature, feature similarity is calculated to measure the matching degree between their corresponding models. The proposed algorithm is geometric deformation invariant. Experiments with industrial applications are presented, which demonstrates the effectiveness and feasibility of the proposed algorithm.     

基于近邻三值模式和协作表示的三维掌纹识别

针对二维掌纹图像存在易伪造、抗噪能力差的问题，提出一种基于近邻三值模式（NTP）和协作表示的三维掌纹识别方法。首先，利用形状指数把三维掌纹的表面几何信息映射成二维数据，以弥补常用均值或高斯曲率映射无法精确描述三维掌纹特征的缺陷；其次，对形状指数图作分块处理，利用近邻三值模式提取分块形状指数图的纹理特征；最后，利用协作表示的方法进行特征分类。在三维掌纹库上和经典算法进行的对比实验中，该方法的识别率为99.52%，识别时长为0.6738 s，优于其他算法；在识别率方面，与经典的局部二值模式（LBP）、局部三值模式（LLTP）、CompCode、均值曲率图（MCI）法相比分别提高了7.77%、6.02%、5.12%和3.97%；在识别时间方面，与Homotopy、对偶增广拉格朗日法（DALM）、SpaRSA方法相比分别降低了6.7 s、15.9 s和61 s。实验结果表明，所提算法具有良好的特征提取和分类能力，能够有效地提高识别精度并减少识别时间。     

A reverse engineering system for rapid manufacturing of complex objects

This paper presents a reverse engineering system for rapid modeling and manufacturing of products with complex surfaces. The system consists of three main components: a 3D optical digitizing system, a surface reconstruction software and a rapid prototyping machine. The unique features of the 3D optical digitizing system include the use of white-light source, and cost-effective and quick image acquisition. The surface reconstruction process consists of three major steps: (1) range view registration by an iterative closed-form solution, (2) range surface integration by reconstructing an implicit function to update the volumetric grid, and (3) iso-surface extraction by the Marching Cubes algorithm. The modeling software exports models in STL format, which are used as input to an FDM 2000 machine to manufacture products. The examples are included to illustrate the systems and the methods.     

Volumetric feature recognition for machining components with freeform surfaces

This paper describes a procedure for the extraction of features of a part containing a combination of 2.5D features and freeform surfaces. This work invokes a previous algorithm that was designed to recognize machining features from 2.5D parts destined to be machined on a 3-axis milling machine. The essence of that algorithm was a volume decomposition based on a recursive descent into the part, yielding a feature graph that captured both the geometry and the spatial relationships of the features. This work augments the previous algorithm with the ability to handle a limited class of components having freeform surfaces. Freeform features are defined similar to the 2.5D features as comprising a planar contour, but substituting a bottom freeform surface for the depth. Covering faces, defined as projection of the freeform surface on the faces of the bounding box of the surface, are used as equivalent planar faces for performing the recursive descent. Inter-feature open edges are used to signal the relationship between the freeform feature and other neighboring features. Examples of molds and components that were machined using the proposed algorithms are also presented.     

Structural optimization of as-built parts using reverse engineering and evolution strategies

In industry, some parts are prone to failures or their design is simply sub-optimal. In those critical situations, one would like to be able to make changes to the part, making it lighter or improving its mechanical resistance. The problem of as-built parts is that the original computer-aided design (CAD) model is not available or is lost. To optimize them, a reverse engineering process is necessary to capture the shape and topology of the original design. This paper describes how to capture the original design geometry using a semi-automated reverse engineering process based on measurement provided by an optical 3D sensor. Following this reverse engineering process, a Fixed Grid Finite Element method and evolutionary algorithms are used to find the optimum shape that will minimize stress and weight. Several examples of industrial parts are presented. These examples show the advantages and disadvantages of the proposed method in an industrial scenario. Presented at the 7th World Congress on Computational Mechanics, Los Angeles 2006.     

Rough set-based heuristic hybrid recognizer and its application in fault diagnosis

Rough set theory (RS) has been a topic of general interest in the field of knowledge discovery and pattern recognition. Machine learning algorithms are known to degrade in performance when faced with many features (sometimes attributes) that are not necessary for rule discovery. Many methods for selecting a subset of features have been proposed. However, only one method cannot handle the complex system with many attributes or features, so a hybrid mechanism is proposed based on rough set integrating artificial neural network (Rough-ANN) for feature selection in pattern recognition. RS-based attributes reduction as the preprocessor can decrease the inputs of the NN and improve the speed of training. So the sensitivity of rough set to noise can be avoided and the system’s robustness is to be improved. A RS-based heuristic algorithm is proposed for feature selection. The approach can select an optimal subset of features quickly and effectively from a large database with a lot of features. Moreover, the validity of the proposed hybrid recognizer and solution is verified by the application of practical experiments and fault diagnosis in industrial process.     

融合注意力与时域多尺度卷积的手势识别算法

针对基于深度学习的动态手势识别方法网络规模庞大、提取的时空特征尺度单一、有效特征的提取能力不足等问题,提出了一种深度网络框架。首先,使用时域多尺度信息融合的时空卷积模块改进3D残差网络结构,大幅度缩小网络规模,获得多种尺度的时空感受野特征;其次,引入一种全局信息同步的时空特征通道注意力机制,使用极少量参数构建特征图通道之间的全局依赖关系,使模型更加高效地获取动态手势的关键特征。在自建手势数据集SHC和公开手势数据集SKIG上的实验结果显示,提出的手势识别方法参数量更少且多尺度时空特征提取能力强,识别率高于目前的主流算法。