高分辨率遥感图像中机场目标的并行快速提取

针对高分辨遥感图像机场目标检测存在漏检和误检,鲁棒性低,且处理多幅图片速度慢的特点,提出一种基于批量化并行处理模式的遥感图像机场目标提取方法。利用模糊增强方法对图像进行预处理,利用基于像素梯度和方差信息的模糊边缘检测算法对图像进行边缘检测,从中筛选出长直线,利用Hough变换提取其中平行的直线作为机场跑道特征。在得到的特征点中选取种子点进行区域生长,从而提取出完整的机场目标。利用MPI多进程并行处理的编程方法来实现对多幅图片中机场目标的批量化并行快速提取。实验结果表明该算法具有很好的鲁棒性,能准确地检测并提取出完整的机场目标,能够大幅度提高程序处理多幅图片的速度。     

A GPU-based matting Laplacian solver for high resolution image matting

The recently proposed matting Laplacian (Levin et al., IEEE Trans. Pattern Anal. Mach. Intell. 30(2):228–242, 2008) has been proven to be a state-of-the-art method for solving the image matting problem. Using this method, matting is formulated as solving a high-order linear system which is hard-constrained by the input trimap. The main drawback of this method, however, is the high computational cost. As the size of the input image increases, the matting Laplacian becomes expensive to solve in terms of both memory and computational time. In this paper we propose a GPU-based matting Laplacian solution which is dramatically faster than a conventional CPU solution, and at the same time largely reduces the memory consumption, making this method practical for the first time for high resolution image matting. To achieve this end, we employ a novel hierarchical windowing scheme to approximate the global optimal solution by solving a serial of local regions at multiple scales. We further employ a GPU-based local solver which can efficiently evaluate local solutions under various boundary conditions. Experimental results show that our system in general is more than two orders of magnitude faster than traditional CPU-based solvers, with about 80% less memory footprint.     

一种机载高分辨率图像实时压缩系统的设计

为解决机载高分辨率图像的实时压缩问题,提出了一个基于FPGA+PowerPC的高分辨率图像实时压缩系统的设计方案。本系统主控采用PowerPC处理器,压缩芯片采用ADV202,用FPGA实现图像数据流程中各个环节控制,最终输出数据为高性能的静止图像压缩标准JPEG2000格式。     

A system organization for parallel image processing

One of the problems in the development of multiprocessor systems for image analysis is the selection and efficient utilization of an interconnection network between the multiple processing units. This paper proposes a system organization centered around a class of interconnection networks and a global bus. Control schemes are developed for realizing the intertask communication requirements typically encountered in the parallel formulation of problems for image analysis. These schemes are simple, distributed and efficient. The utility of this organization is demonstrated by evaluating the performance of two applications.     

A parallel algorithm for stochastic image segmentation

A parallel algorithm for syntactic image segmentation is introduced. Stochastic tree grammar is used as a context-generating model. It is shown that when this context-generating process is in the equilibrium state, a matched filter can be designed and applied in parallel to the image. This process can be used for image segmentation in a syntactic pattern recognition system to enhance the performance of the succeeding recognition process.     

Models for low resolution slit scan measurements based on high resolution laser scanning image analysis: DNA and nuclear dimensions

A laser scanning microscope has been developed and employed to make high precision measurements of cells from clinical cervical samples. These specimens have been stained with the DNA specific fluorochrome mithramycin and the fluorescent images analyzed to give information on nuclear size, DNA content and nuclear shape. This information has been used to develop low resolution parameters as measured with flow cytometers, thus permitting some comparisons between high resolution and low resolution features as measured from the same cells. The results indicate that both low resolution flow slit scan and high resolution pattern recognition techniques yield the same average cellular information. However, low resolution methods are insensitive to shape and the influence of shape on the broadening of low resolution features is now under investigation.     

面向对象的高分辨率SAR图像处理及应用

目的随着合成孔径雷达(SAR)技术和分辨率的不断提高,越来越多的空间细节呈现在高分辨率SAR影像上。与此同时,SAR图像的数据量越来越大,人们对其应用需求也越来越高,这使得传统的基于像素的SAR处理方法不再适用。面向对象分析技术以像元集合——"对象"为分析单元,为高分辨率遥感图像处理提供了有效的思路,并日渐成为遥感、摄影测量以及GIS等领域所关注的对象和研究热点之一。目前该技术在光学遥感中已经得到了广泛的应用,但在SAR图像处理中的应用还处于起步阶段。方法本文在简要阐述面向对象分析技术起源和特点的基础上,对SAR图像面向对象技术中常用的多尺度分割算法进行了分类分析,接着对面向对象技术在SAR遥感的应用方向进行全面介绍,最后对面向对象技术在SAR上的应用进行了总结与展望。结果面向对象分析技术在SAR图像处理中的应用主要分为以下五个方面:地物分类、城市信息提取、变化检测、海洋应用、森林应用。结论面向对象分析技术在解决高分辨率SAR图像尺度效应、抑制噪声等方面有着重要作用。目前,国外学者在基于SAR的面向对象分析技术研究上已经取得了一定的进展,但总体上该技术仍面临诸多问题,需要进一步的研究和完善。     

Software package for the quantitative image analysis of histological sections

A software package of general interest, and adapted to the analysis of histological sections has been written. The different programs of this package have been classified into four groups: macroanalysis, selection of fields, analysis, and data processing. The macroanalysis allows the acquisition of topographical information that will be used during the selection of fields and the analysis. Mapping programs have been added to the classical data processing of the measured parameters (statistical calculations, edition of values, histograms, graphs, .. .). They allow the generation of parametric images representing the distribution of particular elements in the section, or the topographical variations of calculated parameters. The recombination of information obtained at different magnifications, and the correlation of information of different types is thus possible and of great interest in the study of complex histological structures.     

高分辨率航空图像压缩系统设计

针对高分辨率航空图像存储和传输的要求,提出了基于POWERPC、FPGA和ADV212的嵌入式图像压缩存储系统设计方法,实现了大小为4008x5366x12bit图像的有损和无损压缩存储。系统设计目标为每2秒存储一幅无损压缩图像,压缩比为2:1;并在限定传输速率为1.44Mbps的条件下,每8秒向地面传送一幅有损压缩图像。系统设计中使用ADV212完成JPEG2000标准压缩算法,FPGA实现图像数据的接收和分块,POWRE PC实现对ADV212的配置管理及图像文件的存储。实验结果表明,该系统设计能够满足预定高分辨率航空图像的实时压缩与存储要求。     

基于形状的高分辨率遥感图像特征提取方法

基于内容的图像检索的关键在于对图像进行特征提取。提出一种基于形状的高分辨率遥感图像特征提取方法。首先使用最小吸收同值核区SUSAN（Smallest Univalue Segment Assimilating Nucleus）算子对高分辨率遥感图像进行边缘检测,生成边缘图像。之后,对边缘图像计算其不变矩,作为该遥感图像形状特征的描述向量。试验结果说明,所使用的方法计算简便,速度快,而且该描述向量能够很好地代表图像的特征,具有较高的应用价值。     

一种Linux网络硬件加密高性能并发调度方法

结合生产者—消费者调度原理进行改进,针对Linux系统的网络数据硬件加密调度提出了一种将Linux网络协议栈与硬件加密调度紧密结合的并发调度方法——HENC。在生产者—消费者调度队列中增加了传送者指针,实现并通过测试数据验证了该高性能调度算法。通过与其他常用网络数据硬件加密调度方法的测试对比,说明了该算法的先进性,并进一步讨论了改进思路。     

A novel method for parallel image processing applications

It is the aim of this paper to introduce a novel method for the calculation of 2-D convolution/correlation for image processing application. The technique combines a recently developed 2-D transform based on the Mersenne numbers with the 2-D Fermat number transform, using the 2-D mixed radix conversion. The resulting combination uses fast two dimensional residue transforms which can be implemented in parallel for high speed and high throughput rate. The technique is suitable for parallel calculation of 2-D convolution/correlation for digital image processing applications purposes.     

Very high resolution remote sensing image classification with SEEDS-CNN and scale effect analysis for superpixel CNN classification

Pixel-based convolutional neural network (CNN) has demonstrated good performance in the classification of very high resolution images (VHRI) from which abstract deep features are extracted. However, conventional pixel-based CNN demands large resources in terms of processing time and disk space. Therefore, superpixel CNN classification has recently become a focus of attention. We therefore propose a CNN based deep learning method combining superpixels extracted via energy-driven sampling (SEEDS) for VHRI classification. The approach consists of three main steps. First, based on the concept of geographic object-based image analysis (GEOBIA), the image is segmented into homogeneous superpixels using the SEEDS based superpixel segmentation method thereby decreasing the number of processing units. Second, the training data and testing data are extracted from the image and concatenated on a superpixel level at a variety of scales for CNN. Third, the training data are input to train the parameters of CNN and abstract deep features are extracted from the VHRI. Using these extracted deep features, we classify two VHRI data sets at single scales and multiple scales. To verify the effectiveness of SEEDS based CNN classification, the performance of SEEDS and three others superpixel segmentation algorithms are compared, and the superpixel extraction via SEEDS method was found to be the optimal superpixel segmentation approach for CNN classification. The scale effect on CNN classification accuracy was investigated by comparing the four superpixel segmentation methods. We found that (1) There is no strong evidence that using scales combinations is better than a single scale in some specific situations; (2) Natural objects with low complexity are not as sensitive to scale as artificial objects; (3) For a simple VHRI that contains clear artificial objects and simple texture, the classification result with multiple scales performs better a the single scale; (4) In contrast, for the complex VHRI containing a large number of complex objects, the classification result with a single small-scale best.     

一种单时相高分辨率遥感影像时空融合算法

受制于传感器本身材料及卫星轨道参数,空间分辨率和时间分辨率是卫星遥感传感器固有的性能指标且难以兼备,使得高空间分辨率卫星的多时相数据合成问题至今仍是制约其广泛应用的关键问题之一。由于可有效综合空间-光谱-时间维的影像信息,多源遥感影像时空融合技术在近十年间得到迅速发展并已成为解决多时相数据合成问题的有力手段,其中基于学习的时空融合策略在合成精度上具有显著优势且应用潜力较高,但因其对字典训练过程的依赖程度较高而在融合过程中存在一定的不确定性。为提高基于学习的时空融合策略的预测精度、运算效率及鲁棒性,通过综合基于辐射归化的大气校正方法、基于误差约束的数据标准化转换机制、自适应多层递进融合策略以及高效的稀疏求解函数库,设计了一种适用于单时相高分辨率遥感影像的时空融合框架,并以国产高分二号卫星与Landsat-8卫星遥感影像为实验数据对该方法进行充分的对比性分析。实验结果表明,该融合框架不仅提升了运算效率,还在影像保真度、纹理特征描述以及光谱一致性等方面比当前的单数据对融合方法具有更好的重构质量。     

一种新的高分辨率SAR图像道路提取算法

针对高分辨率SAR图像中道路目标难以有效提取的问题,提出一种新的高分辨率SAR图像道路提取算法,它结合了参数化内核图割和数学形态学算法。利用参数化内核图割对高分辨率SAR图像中的道路目标进行初级分割,用数学形态学填充空洞,平滑道路边缘;基于道路的几何特征,使用矩阵度、改进的长宽比、复杂度等因子去除虚警;针对处理过程中出现的道路断裂情况,利用数学形态学提取道路目标的中心线,同时根据线段邻近性、方向一致性准则对其断裂部分进行连接,用数学形态学还原道路宽度,得到道路提取结果。实验结果表明该算法不用进行SAR图像预处理,也可以有效抑制相干斑噪声,并且能准确、较为完整地提取道路目标。     

A high quality color imaging system for computerized tongue image analysis

In order to improve the quality and consistency of tongue images acquired by current imaging devices, this research aims to develop a novel imaging system which records human tongue information faithfully and precisely for medical analysis. A thorough demand analysis is firstly conducted to summarize requirements for reliable rendering of all possible medical clues, i.e., color, texture and geometric features. Then a series of system design criteria are illustrated accordingly, and by following them, three hardware modules of the imaging system, including illuminant, lighting path and imaging camera, are optimally proposed. Moreover, one built-in software module, the color correction process, is also provided to compensate color variations caused by system components. Finally, several important performance indicators, including illumination uniformity, system reproducibility and accuracy, are elaborately tested. Experimental results show that captured images are in high quality and keep stable when acquisitions are repeated. The largest color difference between any two acquired images is 1.6532, which is hardly to be distinguished by human observation. Compared to existing devices, the proposed system could provide much more accurate and stable solution for tongue image acquisition. Besides, this developed imaging system has been evaluated by doctors of Traditional Chinese Medicine for almost three years and over 9,000 tongue images have been collected, analysis results based these data also validate the effectiveness of the proposed system.     

高辐射分辨率遥感影像可逆信息隐藏算法

高辐射分辨率（≥10bit）航空航天数字遥感影像已经逐渐成为空间地理信息获取的主要数据源。针对高辐射分辨率遥感影像的特性,提出了一种基于整数DCT变换的可逆信息隐藏算法,通过对整数DCT变换后的高频系数进行修改来实现待隐藏信息的调制嵌入,并提出了相应的容量提升和失真控制方案,且能够在精确提取隐藏信息的同时实现原始数据的完全无损恢复。实验结果表明,该算法具有较好的不可感知性,可以有效抵抗线性拉伸攻击,并且对于剪切、旋转、噪声,滤波等攻击都具有较强的鲁棒性。     

一种基于Hadoop的高分辨率遥感图像处理方法

随着海量、多源的高分辨率遥感数据的获取,耗时较多、效率低下的传统处理方式已经不能满足用户需求。针对上述问题,提出了一种基于云计算的高分遥感数据处理框架,利用Hadoop技术设计和改进了Meanshift图像边缘分割算法,并在Hadoop环境下进行了仿真实验。实验结果表明,在Hadoop环境下的高分辨率卫星图像数据处理速度有了明显的改善。     

A software architecture for user transparent parallel image processing

This paper describes a software architecture that allows image processing researchers to develop parallel applications in a transparent manner. The architecture's main component is an extensive library of data parallel low level image operations capable of running on homogeneous distributed memory MIMD-style multicomputers. Since the library has an application programming interface identical to that of an existing sequential library, all parallelism is completely hidden from the user.The first part of the paper discusses implementation aspects of the parallel library, and shows how sequential as well as parallel operations are implemented on the basis of so-called parallelizable patterns. A library built in this manner is easily maintainable, as extensive code redundancy is avoided. The second part of the paper describes the application of performance models to ensure efficiency of execution on all target platforms. Experiments show that for a realistic application performance predictions are highly accurate. These results indicate that the core of the architecture forms a powerful basis for automatic parallelization and optimization of a wide range of imaging software.