IRS-P6卫星LISS3图像数据质量评价

Quality Evaluation of LISS3 Image from IRS-P6 Satellite

王钦军[1] 田庆久[1,2]

利用辐射精度、清晰度、信息量、信噪比、几何精度、地面分辨率6个数据质量评价指标，对LISS3与TM图像进行对比分析。研究结果表明，LISS3图像在地面分辨率和信噪比方面有显著提高，几何精度（均方根误差为0．28）小于1／2个像元，在绿（0．52～0．59μm）、红（0．62～0．68μm）和近红外（0．77～0．86μm）波段其辐射精度、清晰度、信息量均高于TM图像，在短波红外波段（1．55-1．70μm）相应指标略低于TM图像。分析LISS3在短波红外波段图像质量指标降低的原因，提出增加LISS3第5波段宽度、新增波长2．2～2．5μm的短波红外波段以改进LISS3传感器的波段参数。[著者文摘]

Results of comparing LISS3 and TM images＇ radiance accuracy, sharpness, amount of information, signal to noise ratio, geometric accuracy and ground resolution indicate that LISS3 image＇s ground resolution and signal to noise ratio improve significantly. Not only does LISS3 image＇s gecometric accuracy （0.28） which is less than 0.5 pixels fix the requirements of its use in civilian field, but also its radiance accuracy,sharpness and amount of information in green （0.52-0.59 μm）,red （0.62-0.68 μm） and near infrared （0.77-0.86μm） bands are superior to that of TM image. However, radiance accuracy, sharpness and amount of information in short infrared band are inferior to those of TM image. Based on the results, the authors point out that the narrow band width of LISS3＇ s short infrared band is the main reason that leads to the decrease of image quality in this band and put forward two advice on its improvement in future： increasing the 5th band width and adding a new band whose wave length is between 2.2 and 2.5 micrometer.[著者文摘]

用VB处理ACCESS数据库的图像数据

use VB to management picture data of ACCESS database

史胜利

用VB访问access数据库存储的图像信息时有其特定的方法。VB连接数据库的接口最好使用ADO对象,图像显示可使用image或picture控件,对数据库中存储的图像信息使用ADO的stream对象处理会很方便。[著者文摘]

The paper introduces detailed method how to access the picture information through VB that is stored in ACCESS Database ADO is better data access interface while using VB to connect Database. Showing picture may use image or picture controls. St is convenient that using stream object of ADO to manage picture information that is stored in the Database.[著者文摘]

基于DSP和binLBT的卫星遥感图像数据压缩系统设计

Design of Satellite Remote Sensing Imagery Data Compression System Based on DSP and binLBT

王继东 罗武胜

卫星数据压缩要求实时压缩、低比特率和良好的图像质量，系统压缩算法选择基于定点叠式变换的低复杂度图像压缩方法，给出了定点叠式变换的实现流程和一种改进的零数编码的思路；在综合考虑成本、功耗、数据接口和处理能力的基础上，提出了多DSP＋FPGA的硬件方案．主要由DSP完成压缩运算，型号选取TMS320C6416，对压缩算法的DSP实现中的一些关键问题，如数据溢出、代码优化、并行计算等技术进行了研究，以一组遥感灰度图像为样本，分4：1和8：1两种压缩比对数据压缩模块进行了测试；在CCD相机时钟频率为50MHz、帧频为1．3K、每帧4096字节的输出条件下，每个DSP可以处理两路图像数据，压缩图像质量良好。[著者文摘]

Satellite data compression requires real time processing, low bit rate and high fidelity of reconstruct image. The compression arithmetic is a low complexity compression method based on fixed point lapped biorthogonal transform. The flow chart of fixed point lapped biorthogonal transform and the thing of a modified zero tree coding algorithm are gived. Based on the analysis to cost. energy consume, interface and performance, the project of hardware based on multi-DSP and FPGA is put forward. The computation is mainly done by DSP whose type is TMS320C6416. Also, some key techniques on transplant of compression arithmetic from PC to DSP, such as data overflow, optimization methods for the code and parallel computing are studied. The data compression module is tested at the ratio of 4：1 and 8：1 based on a group of remote sensing images. Every DSP can deal with picture datas of two channel when the clock frequency of CCD camera is 50M, the frame frequency is 1. 3K and every frame has 4096 bytes, the reconstruct image quanlity is good.[著者文摘]

基于同步时分复用技术的数字CCD图像数据的光纤传输设计

The Design of Fiber Transmission System for Digital CCD Camra Image Data Based on Synchronous Time-Division Multiplexing

李岩[1,2] 司玉美[1,2] 耿爱辉[1] 郭丽红[1]

针对数字CCD图像数据传输的问题，提出了一种借助于光纤的串行传输方案．阐述了系统工作原理、硬件结构和软件仿真。以CPLD和FIFO为核心器件，基于同步时分复用原理，实现了2路同步控制信号和8路图像数据的不同速率信号的复接和分接。对以往的并行传输系统进行了改进，简化了系统结构，并实现了CCD相机图像数据高速率。远距离传输的要求。[著者文摘]

Aim at the problem of transmission for digital CCD image data ,this paper presents an implementation of serial fiber transmission system and its operating principle,hardware makeup and software design .with CPLD and FIFO used as the core devices in the system ,the multiplexing of two-route synchronus controlling signals and eight-route image data was realized on the basis of synchronous time-division multiplexing theory. This system simplify the structure of parallel tansmission system used in the past , and meet the need of high speed and long distance.[著者文摘]

机械手细胞微注射的深度提取及图像数据处理

Extraction of Depth Information and Image Processing in Manipulator Cell Injection

郭阿全 钟辉 洪小雅 席文明 孙道恒

在细胞注射时,显微镜上CCD获取的只是注射针的二维信息,但光轴方向的深度信息丢失,这会导致细胞注射的失败,必须用其他方法获得。利用激光三角测量法对注射针以及细胞载玻片的深度信息进行测量,通过纳米平台的移动来标定激光入射角参数,对获取的数据进行图形化处理,确定线激光条纹图像间的偏移距离,从而获得标准物体高度与激光条纹图像偏移量的像素比值,该比值可以用来计算实际物体的高度。另外,利用三自由度机械手,对测量的探针离细胞载玻片的距离进行验证,获得激光三角测量法的误差值。实验结果表明,在细胞注射中,利用激光三角测量法获得注射针的深度信息是可行的。[著者文摘]

Only two dimensions information of injection needle could be captured by CCD in microscope image, and the lack of the depth information through the optical axis of microscope system would result in the failure of cell injection. A new way called laser triangulation to exract depth information of the injection needle and the slide glass was presented. The measuring principle was calibrating the projection angle of laser, θ was callibrated by the remotion of nano platform, then ascertain the displacement of laser stripe, S by image processing, The height of actual objects will be decided by S. tgθ. In addition, the error of laser triangulation is demonstrated to be about from 5 to 10 μm by calibration using a 3-DOF manipulators. The result indicates that it is feasible to measure the depth information of injection needle by laser triangulation with the required precision in cell injection.[著者文摘]

基于对象的编码模式优化选择算法

张剑 郑心炜

各种视频压缩编码标准都是根据人们在不同领域中对声像数据的要求所制定的，并且随着人们的需求不断地发展。目前，视频压缩编码研究主要分为两个方向：一是基于传统的DCT混合编码方案；另一个是基于第二代图像编码技术而提出的基于对象的编码方案。其中，基于对象的编码方法不仅能满足进一步获得更大的图像数据压缩比的要求，而且能够实现人机对话的功能，所以，我们认为它将是未来视频压缩编码的发展方向。本文对基于对象的编码模式选择算法进行了研完，借用了MPEG-4中视频对象的概念，提出了一种视频分割的方法。

扩散张量成像图像数据采集技术的参数优化

Parameters Optimization of Data Acquisition Techniques in Diffusion Tensor Imaging

倪红艳[1,2] 祁吉[2] KAVCICVoyko[3] 钟建辉[4] 王明时[1]

为优化扩散张量成像图像采集技术,在1.5 T磁共振扫描仪上,使用3种具有不同扩散编码梯度方向数目（NDGD）的序列成像,采用基于感兴趣区和体素的2种方法,在6个部位比较3个序列的扩散系数（〈D〉）、各向异性分数（FA）和扩散张量的本征值.另外,采用具有不同扩散敏感因子b值的序列成像,测量并探讨b值与图像信噪比和表观扩散系数的关系.结果表明,在感兴趣区水平,所有序列的〈D〉和FA及其标准差没有统计学区别,但张量本征值却有差异.在体素水平,在多数部位21-NDGD和31-NDGD序列的相关系数较高.随着b值增高,信噪比降低.当b值在一定范围内时,表观扩散系数及其标准差有不同的变化.实际中应根据不同的应用（感兴趣区或体素分析以及扩散张量成像测量参数的不同）来选择NDGD,选用b值介于400～1 300 s/mm^2较为合适.[著者文摘]

To study and optimize diffusion tensor imaging （ DTI ）acquisition techniques,in a 1.5 T magnetic resonance imaging scanner, single-shot DTI images were acquired by using three protocols with different numbers of diffusion-encoding gradient directions（ NDGD ）. DTI indices, including diffusivity （〈D〉）, fractional anisotropy （ FA ） and diffusion tensor eigenvalues, were calculated and compared with regions of interest （ ROI ） level and voxel-level methods for three protocols in six locations. In addition, DTI images were acquired with different diffusion sensitized factors b. The relationships between signal-to-noise ratio, apparent diffusion coefficient and b were analyzed. At the ROI level, no significant differences were found for 〈D〉 and FA among protocols, however, not for eigenvalues. At the voxel level, the correlation coefficient between 21-NDGD and 31-NDGD protocols was higher in most locations. Signal-to-noise ratio reduced with increased b. Apparent diffusion coefficient and its standard deviation had different change in a certain range of b. It is concluded that the selection of NDGD should depend on the different diffusion tensor imaging applications to ROI or voxel-level analyses and different diffusion tensor imaging indices, and b in the range of 400-1 300 s/mm^2 is better.[著者文摘]

H.264/AVC帧内4×4块预测模式选择算法的研究

4 × 4 intra prediction mode selection algorithm in H. 264/AVC

丁媛媛[1] 司玉娟[1,2]

在分析H.264帧内预测技术的基础上,提出了一种快速的帧内4×4块预测模式选择算法。该算法计算当前4×4块与空间/时间相邻4×4块之间的相似程度,判断是否跳过其模式选择过程;对未跳过的4×4块,根据变换系数的绝对差值和（SATD）信息及空间/时间相关性,排除一些可能性小的预测模式。实验表明,与JM10.2相比,对于IIII及IPPP编码结构：编码时间平均减少了44.07%和20.49%,码率和Y分量的信号噪声比（SNRY）基本不变。[著者文摘]

Based on the analysis of intra-prediction technology, a fast 4 × 4 intra-prediction mode selection algorithm was proposed. The comparability between the current 4 × 4 block and the spatial/temporal 4 × 4 block was computed, and whether to skip the process of mode decision was decided. For other 4 × 4 blocks, some less probable modes were eliminated based on Sum of Absolute Transformed Difference （SATD） and temporal/spatial correlation. Experimental results show that compared to JM10.2, for IIII structure： timesaving is 44.07% ; for IPPP structure： timesaving is 20.49%. The bit rate and Signal to Noise Ratio of Y （SNRY） are almost invariable.[著者文摘]

动态联盟中伙伴选择策略的研究

陈翔宇 申德荣 于戈

提出了一种在动态联盟中进行伙伴选择的策略。首先给出了动态联盟中伙伴选择过程，接着分析了伙伴选择评价指标体系，然后考虑到在此类群体决策问题中盟主及各专家拥有不同决策权重的情况，基于Fuzzy AHP提出了一种加权群体决策方法，最后给出一个应用此伙伴选择策略的实例。

基于人工免疫的入侵检测系统负选择并行算法

葛丽娜[1] 钟诚[2]

基于人工免疫的入侵检测技术是网络安全的一个新兴研究领。运用人体免疫系统的出色功能，对负选择算法进行并行化设计，根据程序并行性的条件将串行的负选择算法中的任务进行划分，由多个处理机并行求解子任务，并将子解合并获得问题的解。分析表明，基于人工免疫的入侵检测系统负选择并行算法获得线性加速。