融合区域和全局特征提取的医学图像检索技术

针对胸部CT扫描图像库,提出了一种融合区域和全局特征提取的医学图像检索方法。为了提取局部感兴趣区域,给出了一种基于灰度层共现矩阵的区域增长算法,分割出病灶区域,再结合迭代阈值算法进行病灶边界的磨合。为了避免身体姿势问题造成的图像角度差异,利用具有旋转不变性的Zernike矩提取图像的全局特征,融合感兴趣区域的形状和分布特性以及整幅图像Zernike矩全局特征作为图像匹配准则的客观依据。实验结果表明,该算法能够比较有效地应用于基于内容的医学图像检索系统中。     

一种动态场景多运动目标的综合检测方法

提出一种动态场景下多运动目标检测的方法。该方法融合基于帧间图像差值的运动分割技术以及区域生长法来获得各运动目标的初始轮廓。再利用主动轮廓线模型进行优化,从而得到各运动目标的最优轮廓,该方法具有以下明显特点：允许背景任意复杂;在无补偿情况下仍能得到良好结果;目标大小不影响算法的鲁棒性．实验证明了该方法的有效性、实用性和鲁棒性。     

基于最大熵原则和灰度变换的图像增强

提出了一种利用最大熵原则和灰度变换进行图像对比度增强的方法.在最大熵原则基础上利用条件迭代算法对图像灰度级进行最佳分类,对各分类区域进行相应的灰度变换,根据不同需要选取变换参数,在图像对比度增强同时各区域均衡性也得到很大改善.将利用条件迭代算法计算最大熵多阈值的方法与最小均方误差(LMSE)计算多阈值的方法进行比较,实验结果表明,文中所用方法在迭代次数上大大低于基于最小均方误差算法所需迭代次数,节省了图像处理时间,图像均衡化效果也相对提高.     

基于分级位平面偏移的多感兴趣区图像编码

被JPEG2000推荐的Maxshift方式能够高效完成单一感兴趣区(ROI)图像编码,但无法支持不同兴趣度下的多ROI编码.本文提出一种采用分级位平面偏移框架的多ROI编码方式-CBShift.新方法将所有ROI与背景位平面按重要性不同分为重要ROI与背号位平面区(MSRB)、一般重要ROI与背景位平面区(GSRB)及不重要背景区(LSB).当图像中含有多个ROI时,通过调节每个ROI在GSRB区的位平面偏移量支持不同兴趣度下的多ROI编码.实验显示,新方式不仅具有Maxshift方式的优点,而且支持不同兴趣度下无需ROI形状信息的多ROI编码.     

Blood smear analyzer for white blood cell counting: A hybrid microscopic image analyzing technique

Total count and differential count of leukocytes or white blood cells (WBC) in blood samples are very important pathological factors for diagnosing a disease. There are not enough pathological infrastructures in the remote places of India and other developing countries. The objective of this work is to design a system, compatible with telemedicine, for automatic calculation of the total count and differential count of WBC from the blood smear slides. Hemocytometer based WBC counting provides more accurate result than manual counting, but hemocytometer preparation process needs expertise. As this device is targeted for remote places, blood smear technique is adopted to reduce the overhead of the operator. In the proposed system, microscopic images of blood smear sample are processed to highlight the WBC for segmentation. Region segmentation procedure involves background scaling and redundant region elimination from the region set. After segmentation, the more accurate region boundary is restored by using gradient based region growing with neighbourhood influence. Individual regions are separately classified on the basis of shape, size, color and texture features independently using different fuzzy and non-fuzzy techniques. A final decision is taken by combining these classification results, which is a kind of hybridization. A set of rules has been generated for making final classification decision based on outputs from various classifiers. The sensitivity and specificity of the system are found to be 96.4% and 79.6%, respectively on a database of 150 blood smear slides collected from different health centres of Kolkata Municipal Corporation, Kolkata, India.     

The Power Function,of the Exact Test for the 2 × 3 Contingency Table

In continuation of the authors' tables of significance tests (1963), the power function of this exact test for the 2 × 3 contingency table using the Freeman & Halton principle is derived and evaluated numerically in the form of contours of constant power for the cases: A = 5(5) 15, where A is the total number in each of the three groups observed according to the numbers of ‘Successes’ (= a _i ) in each. Comparisons are also made with the exact significance points of the competing X ² and (?2 log λ) tests under the null hypothesis H ₀ and also alternatives to H ₀.     

从科技发展看建筑的地域性

建筑的地域性不仅是特色．更是一种永恒的思潮和没有终点的过程。论文从科技与人文的互动作用于建筑的发展历程来考察建筑的地域性．说明地域建筑表征人文精神．并常常作为科技发展的伴生物而存在．是一种能够矫正科技偏离人性的力量。论文在上述认识的基础上进一步归纳了地域建筑在创作层面的四个基本属性有机性．相对性、开放性与现实性。     

共建共享地区特色数据库的思考与对策

信息时代的到来，数字资源建设已经成为图书馆建设的核心。本文通过对潮汕地区特色数据库建设的调查，分析了广东省潮汕地区共建特色数据库的必要性与可行性，提出了共建共享地区特色数据库的对策。     

Automatic segmentation of color images with transitive closure

In this paper, a mask based automatic segmentation algorithm for color images which uses pixel similarity has been presented. Main concept of the algorithm relies on spatial mask for course segmentation and the Warshall's transitive closure (TC) computation algorithm for region merging. Although the proposed spatial mask approach reduces the computational burden required for segmentation or clustering techniques such as seeded region growing (SRG) or fuzzy c-means (FCM) in which user supplied parameters are essential, it has over segmentation drawback. Therefore, the transitive closure algorithm, which uses adjacency and similarity matrix associated to undirected graph of the over segmented image, has been employed to merge the regions. After comparing to existing methods, the obtained experimental results confirmed that the color images as well as gray level images could be segmented with considerable accuracy. Also computational complexity of image segmentation is significantly reduced. Furthermore, there is no need any user supplied parameter such as the number of clusters or seed points.     

二维网格中基于最小通路区的自适应和最小容错路由算法

网格结构是并行与分布式处理中最流行的一种网络拓扑结构。在存在故障的情况下，如何设计具有最优性的容错路由算法一直是研究的热，点问题。本文研究了采用故障块模型的二维网格的最小路由问题，提出存在最小通路的一个充分必要条件。基于最小通路区（RMP）的概念，提出一种自适应的最小容错路由算法。如果源节点和目的节点之间存在最小通路区，则在最小通路区中进行自适应最小容错路由；反之，则采用多阶段最小容错路由。主要思想就是在存在故障的情况下，尽量保证路由算法能走最短路径。因为只要求知道每个节点的局部信息，故算法是分布式的。