光纤近红外漫反射光谱技术在前胡族植物分类中的应用探讨

探讨了近红外光谱技术在前胡族植物分类中的可能性。用系统聚类分析获得的分类结果与传统的植物学分类比较一致 ,说明本方法应用于前胡族植物中三亚族的分类是基本可行的。     

一种混合策略的家族遗传分类算法

本文在以往研究工作的基础上,对遗传分类算法进行了改进。算法在种群中引入家族概念,采用族间交叉算子对算法进行综合调控,对族内交叉操作进行微调,针对分类问题的特点采用相应的编码方式和适应度函数。实验结果表明了分类算法的有效性。     

基于卷积神经网络的植物图像分类方法研究

近年来,卷积神经网络已经成为图像分类领域的应用研究热点,其对图像特征进行自提取、自学习,解决了以往图像分类方法的图像低层特征到高层概念之间存在的语义鸿沟。为了解决植物图像的自动分类问题,该文提出一种基于卷积神经网络(CNN)的植物图像分类方法,以植物图像为研究对象,将经典卷积神经网络VGG16与全卷积网络(FCN)相结合,把VGG16中两个通道数为4096的全连接层改为卷积层,构造一个新的VGG16模型为植物图像分类模型。文中制作了一个由43类每类500张总共21500张植物图像组成的图像数据集,作为植物图像分类模型的训练数据集。实验结果表明,所提方法在植物的图像分类上的准确率达到97.23%。应用文中提出的卷积神经网络对植物图像进行分类可以取得目前最好的植物图像分类效果。     

基于产生式规则的植物分类专家系统的设计

本文首先阐述了植物分类的方法和专家系统的组成，在植物性状与种属之间的对应关系的层次模型基础上．采用了基于产生式规则的植物分类知识表示，提出了植物分类专家系统的设计方案。该系统的使用将使得植物分类结果更加准确、高效，具有一定的工程应用价值。     

物候和波谱——位置分析在城镇绿化植物群分类中的应用

遥感图像植物群分类已被证明是植物群分布自动制图快速有效的方法。然而,场景噪声和植物群之间光谱可分性差等形成的负面影响,使传统的分类方法无法满足必要的精度要求。为了解决这个问题,提出了一种称为SLPA的遥感图像植物群分类方法。它由波谱—位置联合分析(S-L分析)和植物物候遥感分析(PA)两部分组成。通过向特征空间添加密度描述符以及在特征空间叠加冬、夏季图像特征数据,可以将这两类分析嵌入分类过程。这种改进增加了可用描述符的数量,使分类特征空间足够丰富,以适应复杂分类;同时又降低了分类不确定性,使分类精度获得显著改善。精度测试显示,增加S-L分析和物候分析,将使植物群分类的全局精度分别平均提高15.0%和29.3%。另外,由于采用二值邻域均值替代灰度邻域密度,使得加入S-L分析几乎不引起运算复杂性增大。Matlab测试结果显示,SLPA在城镇植物群遥感自动分类方面具有鲁棒和普适性。     

一种基于模糊聚类和粗糙集的产品族设计知识约简方法*

针对产品族设计中产品族共性与产品个性化之间相互冲突的问题,提出了一种基于模糊聚类和粗糙集的产品族设计知识约简方法。通过模糊聚类对设计参数进行离散化和分类,以粗糙集进行知识约简区分出产品族设计参数和个性化产品设计参数,实现产品族综合性能损失最小。通过对通用电机产品族的实例分析,并经与PPCEM方法比较,验证了该方法的高效性和可行性。     

基于多特征融合与极限学习机的植物叶片分类方法

对植物的分类多通过对植物叶片的分类来实现,为提高植物叶片分类的准确率提出了一种基于多特征融合与极限学习机的植物叶片分类方法.首先对植物叶片彩色图像进行预处理,得到去除叶片颜色与背景的二值图像和灰度图像;然后从二值图像中提取植物叶片的形状特征和不变矩特征,利用灰度图像提取灰度共生矩阵参数作为叶片图像的纹理特征,共得到28...     

基于TensorRT的植物叶片病害实时检测分类模型优化

为了提高边缘计算设备对植物叶片病害检测的识别速率,本研究采用卷积神经网络搭建了植物叶片目标识别模型和植物叶片病害分类模型,并且使用OpenCV将两个模型整合成植物叶片病害检测系统.通过SSD (single shot multibox detector)算法对植物叶片的目标区域进行定位并裁剪,再利用植物叶片病害分类模型对裁剪的植物叶片区域进行病害分类.同时,通过TensorRT加速推理对分类模型进行优化处理,以及在同一台主机设备和Jetson Nano计算平台上,对优化前后的模型进行了对比实验.实验表明,在同一主机设备上优化后的植物分类模型识别速率提升22倍.同时,优化后的分类模型使植物叶片病害检测系统识别速率提升7倍.而将优化后的系统部署在Jetson Nano计算平台上,对比优化前的植物叶片病害检测速率提升10倍,实现了实时的植物叶片病害检测.     

植物志数据库的信息组织

本文以中国植物志为研究对象，根据植物分类的特点和植物地检索，鉴定，统计等要求，以索引文件的形式来表示各分类群的特征信息和它们之间的分类关系，设计并实现了中国植物志的数据库模型和信息组织，用计算机来进行植物的检索鉴定和统计。本课题是中国科学院科学基金资助的课题，已经顺利通过鉴定。     

植物图像识别理论研究

为实现对植物的高效识别与分类,现拟设计一种改进型卷积网络对大规模植物进行提取分类。利用图像算法对提取的植物图像进行增强,提高识别的灵敏度;然后以现有深度神经网络为基础,设计一个门网络,实现数千维输出的深度混合网络;再通过机器学习快速识别并提取探测目标的特征来实现植物的有效识别及分类。根据结果显示,本设计对植物的识别精确度能达到83%。     

A method for taxonomy development and its application in information systems

A fundamental problem in many disciplines is the classification of objects in a domain of interest into a taxonomy. Developing a taxonomy, however, is a complex process that has not been adequately addressed in the information systems (IS) literature. The purpose of this paper is to present a method for taxonomy development that can be used in IS. First, this paper demonstrates through a comprehensive literature survey that taxonomy development in IS has largely been ad hoc. Then the paper defines the problem of taxonomy development. Next, the paper presents a method for taxonomy development that is based on taxonomy development literature in other disciplines and shows that the method has certain desirable qualities. Finally, the paper demonstrates the efficacy of the method by developing a taxonomy in a domain in IS.     

Building Fuzzy Thematic Clusters and Mapping Them to Higher Ranks in a Taxonomy

We present a novel methodology for the analysis of activities engaged in an organization such as the research conducted in a University department by mapping them to a related hierarchical taxonomy such as Classification of Computer Subjects by ACM (ACM-CCS). We start by collecting data of activities of the individual components of the organization and present them as the components fuzzy membership profiles over the subjects of the taxonomy. Our method generalizes the profiles in two steps. First step finds fuzzy clusters of taxonomy subjects according to the working of the organization. Second, each cluster is mapped to higher ranks of the taxonomy in a parsimonious way. Each of the steps is formalized and solved in a novel way. We build fuzzy clusters of the taxonomy leaves according to the similarity between individual profiles by using a novel, additive spectral, fuzzy clustering method that involves a number of model-based stopping conditions, in contrast to other methods. As the found clusters are not necessarily consistent with the taxonomy, each is considered as a query set. To lift a query set to higher ranks of the taxonomy, we develop an original recursive algorithm for minimizing a penalty function that involves ''head subjects'' on the higher ranks of the taxonomy together with their ''gaps'' and ''offshoots''. The method is illustrated by applying it to real-world data.     

基于神经网络的植物分类专家系统

本文介绍了植物分类专家系统（ＰＴＥＳ）的设计与实现。该系统采用神经网络知识表示方法表示知识，是一个基于神经网络（Ｎｅｕｒａｌ Ｎ ｅｔｗｏｒｋｓ－ＮＮ）的植物分类专家系统。     

面向可生存性的网络攻击分类方法

网络攻击分类是研究网络攻击特点及其防护方法的前提,介绍了网络攻击分类的研究现状,对已有网络攻击分类方法进行了分析比较,提出了一种面向网络可生存性研究的网络攻击分类方法,并用实例说明了该分类方法的分类过程。该方法对网络防护人员进行网络可生存性提升很有帮助,为他们有针对性地研究网络可生存性技术提供了依据。     

A Taxonomy of Error Types for Failure Analysis and Risk Assessment

We describe a taxonomy of error types that builds on work by Reason (1990) and Hollnagel (1993). The taxonomy uses a faceted schema subdivided into layers. Each layer targets a particular analysis task, and a method is proposed leading the assessor through sets of error types at different levels (e.g. cognitive, social, organizational). This allows the causes of failure to be attributed to one or more error types. The method is illustrated with two case studies of diverse failures. The utility of taxonomy analysis in understanding the reasons for failure and risk assessment is discussed.     

Face anti-spoofing with generated near-infrared images

Reflection differences between live faces and spoof faces under near-infrared spectrum make near-infrared image based methods obtain superior performance for face anti-spoofing. However, for conventional face recognition systems, near-infrared image based methods need additional near-infrared equipment to capture the input near-infrared images. In this paper, we propose a novel face anti-spoofing method which exploits the clues in both visible light (VIS) images and near-infrared (NIR) images without utilizing any near-infrared equipment during testing. Specifically, we first propose a novel multiple categories image translation generative adversarial network (MCT-GAN) which generates corresponding NIR images for VIS live and spoof face images. Then we utilize convolution neural network to learn fusing features from both VIS images and corresponding generated NIR images for the goal of live and spoof face classification. Qualitative and quantitative experiments demonstrate that our method obtains excellent results compared to the state-of-the-art methods.

     

A particle swarm optimization-aided fuzzy cloud classifier applied for plant numerical taxonomy based on attribute similarity

Data mining techniques are widely used in many fields. One application of data mining in the field of the botany is numerical taxonomy. In the present work, a particle swarm optimization-aided fuzzy cloud classifier based on attribute similarity (PSOCCAS) is used for plant taxonomy by two datasets. Firstly, the proposed classifier is been tested by employing it for the benchmark classification data sets, Fisher’s iris data. The testing accuracy is found very encouraging. The performance of our proposed system is only bettered by some genetic algorithm (GA) or evolutionary algorithm (EA)-based fuzzy systems which showed fantastic results. Then for further validation and broadening application, the PSOCCAS has been presented for quantitative features evaluation, ‘expected species’ selection and successful classification of three sections in genus Camellia (belongs to the family Theaceae). The selected quantitative features are almost those selected in previous works. The method is able to produce 100% accurate classification results in genus Camellia. It is a very simple and robust method to divergences in plant taxonomy. No extensive preprocessing is required. The classification is performed with comparatively comprehensive features than those used in our previous work. The method utilizes the inherent nature of the data in performing various tasks. Consequently, the method can be used for plant numerical taxonomy.     

Critical success factors taxonomy for software process deployment

Many organizations have adopted methods, models, and standards to improve their software processes. However, despite these efforts, they can still find it difficult to deploy processes throughout the organization because most of them focus more on the technical rather than human aspects. This paper proposes a taxonomy of critical success factors for software process deployment. A method to create this taxonomy was developed and applied based on a systematic review of existing literature and is complemented with industry experiences where software processes have been deployed or implemented. Finally, the categories, subcategories, and items of this taxonomy are presented.     

基于RGB色彩空间的四带树木图像颜色校正方法

针对图像受近红外光线参与的影响而出现偏色现象,介绍了一种基于RGB色彩空间的四带（RGB带和一个附加的近红外带）树木图像颜色校正方法. 该方法首先根据R-G差值图像,通过实验获取合适的阈值将四带树木图像二值化. 然后,利用中值滤波器和数学形态学对二值图像进行滤噪处理,提取出偏色较严重区域. 最后分区域进行颜色校正,达到消除近红外干扰的目的. 实验结果表明,该方法对部分常见含植物的场景图像有比较好的校正效果.