低质量图像基于散射图SEM估计的MAP像素聚类方法

直方图是最常用的图像灰度统计分布的表示方法,二维直方图包括共生矩阵,灰度－边界值散射图,灰度平均灰度散射图等,本文提出了一种采用ＳＥＭ算法完成对灰－平均灰度的二维正态分布假设的参数估计,然后采用最大后验概率（ＭＰＡ）准则进行像素无监督聚类的图像分割算法,测试结果显示,我们的算法性能良好,尤其是对低对比度,有阴影和重噪声的低质量图像的分效果远优于其他基于散射图的阈值化方法。     

基于图象方向性信息测度的图象象素分类

针对以往象象图处理方法缺乏通用性的问题,提出了一种新的基于图象方向性信息测度的图象素数分方法,即利用不同类型的图象象素的方向性信息测度,以及方向性信息测试随观观察尺度变化时的改变不同,来将图象象素分成边缘点,平滑点和纹理点以及用于图象分析,该方法不仅可用于任意图象的前期分析,而且具有良好的抗噪能力,实验证明,该方法具有令人满意的效果。     

用于视频镜头突变切换检测的二次差分法和像素点匹配二次差分法

镜头切换检测是视频检索研究领域中一项基本的、重要的技术．对于数字视频镜头突变切换的检测，一般有模板匹配法、直方图法等基本算法，但这些算法在实际检测中通常达不到较高的检测精度．为此，提出了一种新的用以消除视频运动对镜头切换检测影响的检测算法，即二次差分法，和一种新的计算两帧图象帧间差的算法，即像素点匹配法，实验结果表明，二次差分法较之直接帧间差算法，明显地提高了镜头突变切换检测的检测精度，像素点匹配法较之现有基本帧间差算法总体地提高了检测精度，而像素点匹配二次差分法达到了最好的检测精度，它能够实现在无虚检或只有很小虚检的条件下，对镜头突变切换的无漏检检测．     

反走样技术在计算机图形仿真中的运用

阐述了反走样技术的基本原理和实现方法,重点描述了计算机图形中的直线反走样算法。该文采用的算法是基于三像素线宽的,通过调整像素的亮度来消除直线绘制中的锯齿现象。并把该算法推广到圆和椭圆的反走样。最后,采用改进的直线扫描转换算法与反走样算法相结合,进一步提高了反走样直线的绘制速度。该算法利用了直线的多段特性,以及空间对称特性。仿真表明,相比于Bresenham的算法,该算法的速度大为提高。     

Image analysis system for quantitative metallography

This paper describes a Z-80 microprocessor-based image analyzer developed for global parameter evaluation of images over a 256 × 256 pixel frame. It consists of a microscope,ccd scanner, 6-bit videoadc, Z-80 computer and an image display monitor. Facilities are provided for feature erosion/dilation and halo correction. The paper also presents the details of another more powerful user microprogrammable HP1000 minicomputer-based image analysis system under development. This system consists of an optical microscope/epidiascope coupled to a chalnicon scanner. Here the 512 × 512 pixel image is acquired with 8-bit resolution. It provides for shading correction, auto-delineation, image processing and image analysis functions for evaluation of various basic and derived parameters. Both the systems are software intensive and are realised according to requirements of quantitative metallography. They can also be used for analysis of images obtained in the fields of biology, medicine, geological survey, photography and space.     

Parallel,overlapped, and intersected digital straight lines

A digital straight line is represented by a chain code of directions which satisfy Freeman's three criteria. Given two chain codes of digital straight lines and their starting pixels where their equations are not known, the paper will describe how to determine if they are parallel, overlapped, or intersected. An algorithm with only integer arithmetic is also implemented to find the intersection.     

屏幕坐标系中求任意曲线交点的一种方法

本文讨论了用光栅化比较法在屏幕坐标系中求任意曲线交点的一种方法，与传统的解析法相比，具有简单，准确等特点，且节省了大量的求交运算。     

参数曲线的分段步长生成算法*

介绍了参数曲线的现有逐点生成算法,并在此基础上提出了一个新的分段步长生成参数曲线的算法,对所绘制的曲线的每段分别使用不同的步长。验证了新算法的步长值优于现有算法的步长值,从而进一步解决了生成参数曲线的取点过密问题。实验结果表明,新算法所绘制的点数及运行时间都小于现有算法。     

三维图形显示系统的图形消除器和象素处理器技术实现

图形消隐器和象素处理器是三维图形显示系统的重要组成部分。讨论了图形消除器和象素处理器的算法,结合具体的硬件电路提出了实现设计的思路,实践应用表明了这种设计的可行性。     

Shared Sampling for Real‐Time Alpha Matting

Image matting aims at extracting foreground elements from an image by means of color and opacity (alpha) estimation. While a lot of progress has been made in recent years on improving the accuracy of matting techniques, one common problem persisted: the low speed of matte computation. We present the first real‐time matting technique for natural images and videos. Our technique is based on the observation that, for small neighborhoods, pixels tend to share similar attributes. Therefore, independently treating each pixel in the unknown regions of a trimap results in a lot of redundant work. We show how this computation can be significantly and safely reduced by means of a careful selection of pairs of background and foreground samples. Our technique achieves speedups of up to two orders of magnitude compared to previous ones, while producing high‐quality alpha mattes. The quality of our results has been verified through an independent benchmark. The speed of our technique enables, for the first time, real‐time alpha matting of videos, and has the potential to enable a new class of exciting applications.