首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献,搜索用时 15 毫秒
1.
高光谱遥感影像中光谱向量编码方法的研究   总被引:2,自引:1,他引:2  
Taking into account the demands of hyperspectral remote sensing (RS) image retrieval and processing, some encoding methods of spectral vector including direct encoding, feature-based encoding and tree-based encoding methods are proposed and compared. In direct encoding, based on the analysis of binary encoding and quad-value encoding, decimal encoding is proposed. It is proved that quad-value encoding and decimal encoding are suitable to fast processing and retrieval. In absorption feature-based encoding method, five common metrics are compared. Because locations of reflection/absorption features are sensitive to noise, this method is not very effective in retrieval. In tree-based encoding methods, bitree, quadtree, octree and hextree are proposed and discussed. It is proved that 2-level octree and 2-level hextree are more effective than bitree and quadtree. Finally, quad-value encoding, decimal encoding, 2-level octree and 2-level hextree are proposed in spectral vectors encoding, similarity measure and hyperspectral RS image retrieval.  相似文献   

2.
图像特征提取作为图像处理中的一项基本技术,仅依靠单一的图像特征很难准确、可靠地描述图像信息.利用四叉树分裂合并和局部特征SIFT算法的特点,本文提出了一种基于四叉树和特征融合的图像特征提取方法,首先利用四叉树对图像进行区域分割,然后对分割图像提取颜色直方图信息,接着使用局部特征SIFT算法进行特征精提取,达到特征信息的融合,从而更好识别图像特征信息.最后对此方法进行实例仿真,所提特征能够较完整地表示图像的特征信息,冗余度比较小,具有良好的尺度不变性.  相似文献   

3.
通过对八叉树数据结构的分析,将闭孔材料的体数据转换成具有八叉树分层结点的体数据,详述了基于八叉树结点的孔泡逐层搜索算法的6步体积增长遍历和一步体积分解遍历步骤,给出了算法继续遍历的条件和终止遍历的3种可能,并从理论上证明了该遍历算法的完备性.在算法调用过程中,确定了孔泡基于体素为单位的体积和分布等参数的计算方法.利用仿真的柱状孔泡图像体数据,对八叉树遍历算法的提取效果进行了验证.实验结果表明,基于八叉树结构的遍历算法能准确提取闭孔材料中孔泡的体积和分布信息,为更有效地进行多孔材料性能分析奠定了基础.  相似文献   

4.
针对谱聚类算法稳定性较差的问题,提出了一种改进的半监督谱聚类算法。该算法依据图像的颜色、纹理和空间特征进行聚类,通过Bayes距离学习对相似度矩阵的内容进行修正;然后,使用半监督K—means聚类算法对调整后的特征向量进行聚类划分。仿真实验结果表明。较传统谱聚类而言该算法在准确率及稳定性上都有了显著提升。  相似文献   

5.
MRI and PET images fusion based on human retina model   总被引:1,自引:0,他引:1  
The diagnostic potential of brain positron emission tomography (PET) imaging is limited by low spatial resolution. For solving this problem we propose a technique for the fusion of PET and MRI images. This fusion is a trade-off between the spectral information extracted from PET images and the spatial information extracted from high spatial resolution MRI. The proposed method can control this trade-off. To achieve this goal, it is necessary to build a multiscale fusion model, based on the retinal cell photoreceptors model. This paper introduces general prospects of this model, and its application in multispectral medical image fusion. Results showed that the proposed method preserves more spectral features with less spatial distortion. Comparing with hue-intensity-saturation (HIS), discrete wavelet transform (DWT), wavelet-based sharpening and wavelet-à trous transform methods, the best spectral and spatial quality is only achieved simultaneously with the proposed feature-based data fusion method. This method does not require resampling images, which is an advantage over the other methods, and can perform in any aspect ratio between the pixels of MRI and PET images.  相似文献   

6.
The technique of adaptive tree mesh is an effective way to reduce computational cost through automatic adjustment of cell size according to necessity. In the present study, the 2D numerical N-S solver based on the adaptive quadtree mesh system was extended to a 3D one, in which a spatially adaptive octree mesh system and multiple particle level set method were adopted for the convenience to deal with the air-water-structure multiple-medium coexisting domain. The stretching process of a dumbbell was simulated and the results indicate that the meshes are well adaptable to the free surface. The collapsing process of water column impinging a circle cylinder was simulated and from the results, it can be seen that the processes of fluid splitting and merging are properly simulated. The interaction of second-order Stokes waves with a square cylinder was simulated and the obtained drag force is consistent with the result by the Morison’s wave force formula with the coefficient values of the stable drag component and the inertial force component being set as 2.54.  相似文献   

7.
针对少数民族服饰具有的典型纹样图案和款式特征,提出一种基于区域—图非对称匹配的少数民族服饰图像检索方法,首先利用 gPb-OWT-UCM 算法对检索图像进行区域分割,然后将分割后图像中的每个区域特征与未分割的待检索图像进行基于动态规划的非对称匹配,并通过评估匹配组的布局一致性计算少数民族服饰图像之间的相似度,最后通过排序实现图像检索任务。实验结果表明,与传统基于外观特征的方法相比,该方法不仅可以获得较好的图像检索效果,还可以通过相似区域匹配结果呈现不同少数民族服饰独有的风格特征,为少数民族服饰文化的数字化保护与传承提供有效途径。  相似文献   

8.
Within the context of classroom learning, strategic transfer can be viewed as a tool for academic problem solving. Strategic transfer is defined as the spontaneous access and retrieval (remembering) of previously learned formal procedures for the successful solution of a problem. The transfer-appropriate processing encoding model (Morris, Bransford, and Franks, 1977), and the transfer-appropriate procedures retrieval model (Roediger, Weldon, and Challis, 1989) are reviewed. An integration of the two models is proposed through the development of a training-for-transfer paradigm (Phye, 1990). By integrating encoding and retrieval processing in a single transfer paradigm the issue of accessing prior knowledge (Bransford, 1990) that is also referred to as the inert knowledge problem (Whitehead, 1929) can directly be addressed.  相似文献   

9.
The non-symmetry anti-packing image representation (NAIR) uses a sequence of the instances of some predefined prototypes to represent an image. While significantly reducing the instances required to represent an image in contrary to the quadtree and the linear quadtree, however, NAIR has lost the explicit space relationship among these instances and hence made some geometric operations such as perimeter computation hard to be implemented. In this paper, longitude and latitude grid (L^2G), a data structure which can restore lost space relationship from the NAIR is first presented, and then a novel algorithm to compute the perimeters of the images represented by the NAIR is presented. The experimental results show that the new algorithm has saved at least 90% of the running time comparing with that based on the quadtree.  相似文献   

10.
由于单一特征不足以准确地描述图像特征,提出了一种结合颜色特征和边界方向特征的图像检索方法.针对传统颜色直方图中图像对所有像素具有相同重要性的问题进行了改进,提出了像素加权的改进颜色直方图方法;然后采用非分割图像的边界方向直方图方法提取图像的形状特征,该方法相对分割方法具有简单、有效等特点,并对图像的缩放、旋转以及视角具有不变性.为进一步提高图像检索的质量引入相关反馈机制,动态调整两幅图像相似度中颜色特征和方向特征的权值系数,并给出了相应的权值调整算法.实验结果表明,上述方法明显地优于其它方法.  相似文献   

11.
计算机教学中应强调的几种图像处理方法   总被引:1,自引:0,他引:1  
图像处理方法有很多种,教学中应特别强调图像增强、图像编码、图像复原、图像分割、图像识别等方法的使用,重点把握各种方法的适用情况.  相似文献   

12.
为了有效地定位交通监控图像中的车辆区域,提出了一种基于车辆轮廓对称和车牌定位信息融合的车辆检测方法. 该方法首先检测图像中的车辆轮廓竖直对称轴,然后以车辆轮廓对称轴位置为基准检测车牌水平和竖直对称轴,最后根据车牌横纵对称轴和车辆轮廓图像的水平、竖直投影进行车辆区域定位. 以450 张 15 类车型的图片为测试集进行了基于对称特征融合的车辆区域检测,并与基于车辆边缘、车牌、车辆纹理特征和车辆图像 Gabor 特征的 4 种方法进行了对比,实验结果表明基于车辆轮廓对称与车牌对称特征融合的车辆区域检测方法最优,其检测率和检测时间分别为 90. 7%和 125 ms.  相似文献   

13.
根据加权多特征图像检索方法存在的不足之处,结合相关反馈技术中的权系数调整方法,给出一种根据用户反馈信息动态地调整多特征权值的检索方法,即动态地调整颜色、纹理和形状的权值以及其对应的各特征分量的权值,如颜色矩的9个分量等.实验证明此方法是有效的,提高了图像检索的有效性,减轻了用户设置权值的负担.  相似文献   

14.
随着网络技术的迅猛发展,图像被作为一种数据信息频繁传输于世界各地的网络中。为了保证某些保密图像的安全性,提出了一种改进基于均值运算的线性同余伪随机序列图像置乱算法,先对通过均值运算获取的图像象素进行矩阵转换,将每个象素值从十进制转换成二进制,随后实现按位倒序,将其逐个转换成十进制数,最终实现对于图像的置乱。实验结果表明,该算法更加适用于图像文件的加密传输和秘密图像信息在数字水印技术中的置乱预处理,而且对图像的处理有良好的鲁棒性。  相似文献   

15.
纠两错的RS码在FPGA上的实现   总被引:1,自引:0,他引:1  
介绍了RS码的结构和一般译码方法,重点讨论了RS(15,11)的译码方法。我们根据PS码的循环性采用了step-by-step算法,使占用硬件资源最少从而适合在VLSI实现。并用ALTERA公司的FLEX10K10器件实现了RS(15,11)纠错码。经试验测试,对码组中任意不多于两个的符号错误都可以给予纠正,且运行可靠。  相似文献   

16.
冒伟 《教育技术导刊》2020,19(3):248-251
为解决传统谱聚类算法在图像分割时计算量大、使用单一特征分割的局限性问题,设计一种融合谱聚类和多特征的图像分割算法。首先进行超像素分割以减少计算量,分别提取每个超像素的颜色特征和纹理特征,构建超像素相似度矩阵|然后采用特征加权方法线性融合颜色和纹理特征的超像素相似度矩阵|最后采用谱聚类算法进行聚类分割。在UCMerced_LandUse和Berkeley数据集上进行实验测试,并与现有方法进行比较。实验结果表明,大多数实验图像IOU指标均在90%以上,相比于传统方法有了显著提高。  相似文献   

17.
基于多级夹角函数的傅里叶形状描述子(英文)   总被引:1,自引:0,他引:1  
为了描述形状由全局信息到局部变化的层次信息,提出一种有效的形状签名,即多级夹角函数.多级夹角函数具有内在的旋转、平移和缩放不变性.对轮廓上每一点,其多级夹角函数通过轮廓的非等弧长分割所得的成对线段计算得到.然后利用多级夹角函数推导出傅里叶描述子,以进行高效的形状检索.使用标准的性能评价方法对所提出的描述子在3个形状图像库上进行了测试,包括MPEG-7图像库、Kimia-99图像库和Swedish树叶图像库.形状检索实验结果表明,基于多级夹角函数的傅里叶描述子优于已有的傅里叶描述子,且具有较低的计算复杂度.与其他类型的形状描述方法相比,所提出的描述子在相同查全率时具有最高的查准率,证明了该描述子的有效性.  相似文献   

18.
基于数字图像处理技术研究了数显仪表中的数字识别,首先对图像进行预处理和二值化,然后对图像连通域进行快速提取并标号,最后根据数字模板和区域结构特征对数字和小数点进行识别。实验结果表明,数显仪表中的数字能得到有效的识别。  相似文献   

19.
地球空间信息科学是一门研究地球空间系统中各种物质要素存在的空间分布和时序变化及其相互作用的综合性信息科学,多源信息的集成和融合是地球空间信息科学领域的一大热点问题.它是解决地球空间中海量信息的不确定性、多样性、和复杂性的有效手段之一.本文从地球空间信息科学与多源信息获取和处理的关系入手,介绍了多源信息融合的方式,遥感数据融合的方法以及遥感图像信息融合的应用实例,并对多源遥感数据融合的前沿问题和方向进行了分析和展望。  相似文献   

20.
面对耕地资源日益紧张的形势,快速准确地提取耕地信息已成为耕地动态监测的基础。该文以绵阳市游仙区QuickBird遥感影像为基础图件,借助于RS和GIS技术,分析影像本身特点与耕地光谱特征,通过非监督分类法、监督分类法、基于光谱知识的光谱阈值法分别提取耕地信息。试验结果表明非监督分类提取精度为88.4%,监督分类提取精度为91.53%,光谱阈值法提取精度为94.65%。光谱阈值法的提取效果最好。  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号