基于PSO和Level Set的快速曲线演化算法

提出了一种基于微粒群算法（PSO）、水平集（Level Set）的快速曲线演化算法．该算法将图像分割分为两阶段,利用PSO方法对Snake模型中的控制点进行寻优,使控制点能快速地收敛到图像的边缘附近;利用插值算法,得到目标较粗糙的大致轮廓,以此目标轮廓作为初始的零水平集曲线;使用Level Set窄带方法得到准确的轮廓．从而克服了参数主动轮廓线模型对初始曲线敏感、噪声敏感和不能收敛到凹陷的边缘问题和几何主动轮廓线模型水平集方法计算量大等问题．实例研究结果表明了该算法的正确性、有效性．     

基于HMM-SVM融合模型的鲁棒人脸识别算法

针对人脸识别的鲁棒性问题,鉴于HMM具有良好的时间序列建模能力和SVM在有限样本的分类方面具有优良性能,采用一种基于HMM-SVM融合模型的鲁棒人脸识别算法.首先将归一化人脸图像用采样窗从上到下进行采样,采用DCT争SVD提取各个采样窗图像的特征参数并串接成观察向量,然后由每个人的训练图像的观察向量训练得到每个人HMM模型,将测试图像的观察向量采用Viterbi算法求出对应于每个人HMM模型的输出概率,最后将输出概率送入支持向量机进行分类训练及识别测试,得到人脸识别结果.在ORL库和Yale库的实验表明该算法的识别率高于传统的单一HMM方法和SVM方法,鲁棒性有一定的提高.     

一种改进的C-V主动轮廓模型

本文对C-V主动轮廓模型进行改进.依据曲线演化理论对C-V模型中的图像数据力驱动项进行简化,提出一种常微分方程(ODE)类型的模型.理论分析验证了该模型的水平集函数可初始化为零.与传统C-V模型相比,不但具有其特点,如可以自动检测带孔目标的内轮廓等,而且具有以下优点:抗噪性能较优;水平集函数无需重新初始化,可快速计算出全局最优分割;远离轮廓的边界可以被准确检测;时间迭代步长不受限制.对合成和真实图像的分割结果证明了本文模型具有稳健、快速的优点.     

复杂防雷条件下保护范围及三维模型研究

研究了复杂防雷条件下防雷保护范围模型的重建方法,提出通过保护范围平面二维轮廓线重建保护范围三维模型的基本思路。通过运用滚球法原理,详细分析了利用包络原理生成保护范围平面二维轮廓线产生的方法,并给出了相应的数学模型;研究通过二维轮廓线重建三维模型的方法。由二维轮廓线重建保护范围模型的方案不仅适合于复杂条件的防雷范围设计,同样也适合简单模型条件下的防雷范围模型,为防雷保护范围设计提供了一条新的思路。     

应用优化的隐马尔可夫模型预测蛋白质二级结构

针对3-状态隐马尔可夫模型(HMM)预测蛋白质二级结构准确率不高的问题,提出了7-状态和15-状态HMM.研究对象为CB513数据集合中筛选出的492条蛋白质序列,将其随机均分7组.分别应用7-状态和15-状态HMM对以上数据集进行二级结构预测,对预测准确率进行了7-交叉验证,并将预测结果与应用3-状态HMM的预测结果进行了比较.结果表明,应用7-状态HMM,Q3准确率提高3.11%,SOV提高6.15%,QE提高6.49%;应用15.状态HMM,QE比7-状态HMM又提高5.74%.在15-状态:HMM预测中加入序列的同源信息后,Q3准确率比单序列15-状态HMM增加8.76%.结果表明,7-状态HMM预测能力优于3-状态HMM,15-状态HMM总体预测能力和7-状态HMM相当,但β折叠预测能力强于7-状态HMM.     

基于GVF模型的多目标边缘提取

针对活动轮廓模型的多目标边缘提取问题,本文提出了一种基于GVF模型的改进算法.该算法首先利用GVF模型对整个目标区域进行边缘提取,由于目标间存在滞止点区域,边缘提取的结果中将包含伪边缘.针对这些伪边缘进行分析并设置新的初始轮廓线,再次利用GVF模型进行目标边缘提取,并进行判断.将分别提取到的真实边缘首尾相连,去除伪边缘,可进行多次判断是否存在伪边缘,直至得到完整的多目标边缘.实验结果表明,该改进算法可以实现多目标边缘提取,且易于实现.     

一种基于HMM和CRF的双层分词模型

中文分词是自然语言处理的基础。根据各种分词模型的特性,提出一种双层分词模型。在底层用隐马尔可夫模型(HMM)进行粗分,在高层利用条件随机场(CRF)对文本进行重新标注。从实验结果证明,该模型是有效的。     

一种基于GPU的三维素描风格画的渲染方法

提出一种在GPU下将三维模型渲染成素描风格画的简单算法,通过分析素描面的特点,将整个渲染过程分为内部渲染和轮廓线的渲染。首先通过一种手绘风格生成算法实现素描风格纹理到模型的实时映射从而实现内部的非真实感渲染,再利用渲染到纹理技术通过边缘检测深度缓冲图像获取轮廓线,并对轮廓线图像采用一种多次采样叠加的算法生成素描风格画所特有的复线,最后将两种渲染结果进行图像的融合,产生最终的素描画风格。     

双重主动轮廓图像分割

本文首先利用小波变换得到原图像的粗分辨逼近,在粗分辨逼近中得到图像的一个粗尺度分割。由于逼近图像中噪声下降,尺寸减少,使得算法对参数的选取不太敏感,而且收敛速度加快。然后将第一次分割结果通过小波反变换返回到原始尺度上,将得到的近似轮廓曲线作为初始水平集函数再在原图像中演化得到更准确的分割,这样就得到了一种双重主动轮廓图像分割算法。由于初始轮廓曲线非常接近真实的轮廓曲线,所以很快就可以收敛到真实的轮廓。理论分析和数值结果表明,双重主动轮廓分割算法可以快速有效地分离出感兴趣目标。     

基于成像模型的遥感图像IHS融合

针对色度融合容易产生光谱扭曲的现象,本文利用光谱响应函数,传感器偏置和增益等参数从成像机理的角度解释了IHS融合.建立了描述从光谱密度到全色图像、多光谱图像再到融合图像的过程的新模型.通过对新模型的分析,本文提出了保持IHS融合光谱特性的条件方程,通过对模型有效的简化进而提出了新的融合方法.而且,本文首次提出利用最小二乘法确定光谱响应函数之间关系并计算融合过程所需要的参数.最后本文分别对QuikBird图像数据进行实验并与多种改进的色度融合方法做了深入的比较,试验证明本文提出的方法不但提高了多光谱图像的空间分辨力而且对保持光谱特性有更加明显的效果.     

Methods for automatic microarray image segmentation

This paper describes image processing methods for automatic spotted microarray image analysis. Automatic gridding is important to achieve constant data quality and is, therefore, especially interesting for large-scale experiments as well as for integration of microarray expression data from different sources. We propose a Markov random field (MRF) based approach to high-level grid segmentation, which is robust to common problems encountered with array images and does not require calibration. We also propose an active contour method for single-spot segmentation. Active contour models describe objects in images by properties of their boundaries. Both MRFs and active contour models have been used in various other computer vision applications. The traditional active contour model must be generalized for successful application to microarray spot segmentation. Our active contour model is employed for spot detection in the MRF score functions as well as for spot signal segmentation in quantitative array image analysis. An evaluation using several image series from different sources shows the robustness of our methods.     

结合数学形态学和Level Set超声图像的分割方法

针对噪声严重的超声图像,提出了一种结合数学形态学和Level Set的分割方法。首先采用全变差模型进行图像滤波,再通过交互式区域选择和数学形态学方法获得感兴趣目标的二值化图像,并把该二值化图像轮廓作为水平集方法的初始曲线。改进隐式测地活动轮廓模型(GAC)中的边缘检测函数,增强了处理弱边缘的能力。分割结果表明,该方法能够准确地提取出目标轮廓,同时减少了迭代次数和运算时间。     

Texture‐based parametric active contour for target detection and tracking

In recent years, active contour models (ACM) have been considered as powerful tools for image segmentation and object tracking in computer vision and image processing applications. This article presents a new tracking method based on parametric active contour models. In the proposed method, a new pressure energy called “texture pressure energy” is added to the energy function of the parametric active contour model to detect and track a texture target object in a texture background. In this scheme, the texture features of the contour are calculated by a moment‐based method. Then, by comparing these features with texture features of the target object, the contour curve is expanded or contracted to be adapted to the object boundaries. Experimental results show that the proposed method is more efficient and accurate in the tracking of objects compare to the traditional ones, when both object and background are textures in nature. © 2009 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 19, 187–198, 2009     

Object detetion and image segmentation using texture pressure energy in parametric active contour models

Abstract

In recent years, Active Contour Models (ACMs) have become powerful tools for object detection and image segmentation in computer vision and image processing applications. This paper presents a new energy function in parametric active contour models for object detection and image segmentation. In the proposed method, a new pressure energy called “texture pressure energy” is added to the energy function of the parametric active contour model to detect and segment a textured object against a textured background. In this scheme, the texture features of the contour are calculated by a moment based method. Then by comparing these features with texture features of the object, the contour curve is expanded or contracted in order to be adapted to the object boundaries. Experimental results show that the proposed method has more efficient and accurate segmenting functionality than the traditional method when both object and background have texture properties.     

基于Gabor脸和隐马尔可夫模型的人像识别算法

提出了基于Gabor小波变换和隐马尔可夫模型的人像识别算法。该算法先对人脸图像进行多分辨率的Gabor小波变换,采用主元分析法对每个结点进行降维,最后形成Gabor脸。把Gabor脸的每个特征结作为观测向量,对隐马尔可夫模型进行了训练,并把优化的模型参数用于人脸识别。实验结果表明,本文方法识别率高,复杂度较低,对部分遮挡的图像具有较大的容忍度。     

Studying Grout Propagation in Granular Soils by Image Processing Techniques

Abstract: This article presents two image‐processing techniques which allow studying grout propagation in granular soils. The first one deals with the evaluation of the cement grout concentration using UV spectrophotometry. The second is based on image segmentation processing: active contour model. We propose to develop the active contour model image analysis technique. This process, currently applied in the field of medicine, can be transposed to grout flow detection during soil injection. They are powerful image segmentation techniques that combine geometry, physics and approximation theory. Two distinct formulations exist to employ these techniques, parametric or geometric curves. These models have proven to be effective in segmenting and tracking non‐rigid structures. They exploit features derived from the image data together with a prior knowledge about the location, size and shape of these structures. We focus on a parametric approach for deformable models. They support highly intuitive interaction mechanisms. Those mechanisms allow, when necessary, researchers to bring their expertise to bear on the model‐based image interpretation task. In this paper, we show the development of the two different image analysis methods employed in our experiments for grout detection and grout concentration measurement.     

一种集成模糊特征的测地线主动轮廓模型的图像分割

构建有模糊特征控制的曲线法向力场集成到测地线主动轮廓模型,使得该模型能包含更多的图像分割信息.当用模糊特征表示图像区域信息时,改进模型保持了原来模型具有边缘指示特性,同时具有区域特征一致性的分割性能.改进模型提高了演化曲线抓取感兴趣目标的范围和提取凹凸区域的能力,可以适应对弱边缘、噪声干涉图像的分割.图像分割的实验说明新的模型对图像分割具有良好的性能.     

具有可调增益的模糊—PID电液主动控制悬架

针对车辆悬架受不平路面激励作用下的振动,进行电液主动隔振研究,在分析PID和模糊控制方法的基础上,对车辆悬架系统的单自由度电液模拟装置进行了理论和实验研究,提出增益自动跟踪和变系数的模糊－PID控制算法,根据输入信号偏差的大小分别采用模糊或PID控制,通过微型计算机与MCS－98单片联机调试控制,悬架振动的控制效果得到提高,从2.875Hz到7Hz的低频范围内取得了优于被动控制的减振效果,为主动隔振技术在车辆减振实时控制中的应用提供了理论根据和实验基础。