3 Dimensional Irradiance Imaging with a Single Camera System *

本文提出了用一个摄像机重建三维发光体辐射场分布的方法．连续的三维发光体可以看成若干个互相平行的二维发光断层的组合,对发光体某断层进行聚焦摄像,得到的图像是该断层的高斯像和其他断层的离焦像的叠加像．对不同断层分别聚焦摄像,得到一组辐射图像,对图像进行变换、处理,采用图像反演算法,可以重建各断层的辐射场．本文用实验装置模拟了具有两断层的三维发光体,获得了初步结果,证明了该理论的正确性．     

瞬态火焰图像处理系统开发及其在实验教学中的应用

热能动力工程装置中涉及瞬态火焰图像的检测与处理,如内燃机燃烧室中火焰温度的检测以及燃烧火焰脉动特性的检测。基于双色测温理论及傅里叶变换原理开发了一套瞬态火焰图像处理系统,结合高速摄像仪获得的瞬态燃烧火焰图像,能够计算瞬态火焰图像的温度分布、燃烧脉动的幅频特性。该系统已用于瞬态火焰图像检测处理的实验教学中,检测了定容弹燃用柴油的温度分布以及层流扩散乙烯火焰的脉动特性。     

物体表面不同角度采集的三维图像配准的一种新算法

提出了一种能有效实现物体表面三维图像数据配准的算法.该算法利用了三维数据采集时的三维点的阵列信息,这些三维点的阵列信息在采集原始三维数据时可以很容易获得,同时,还利用了三维数据矢量/顶点相似特征作为数据匹配的基础.采用迭代最小均方误差算法来自适应优化变换矩阵参数.这些方法可以有效地提高三维图像配准的性能,加快匹配过程的运算速度.实验结果表明该算法可以获得较好的配准后三维图像的主观效果.尽管该方法主要针对人头模型,但经过少许修改后即可适用于其他物体.     

基于姿态和光照不变的三维人脸识别研究

为了解决由姿态和光照变化引起的误识率,提出了一种利用稀疏统计形变模型,进行三维人脸形状重建,通过调整角度,解决人脸图像姿态发生改变的方法.运用球面谐波商图像的方法,估计原始输入图像的光照属性,对生成的模型进行标准化的姿态和光照重构,解决了原始输入图像中由于姿态和光照不同所引起的识别误差.实验表明,该方法能够有效地提高对三维人脸的识别率.     

多孔介质三维重建的训练图像分析(英文)

为了使砂岩三维重建结果更接近于原始样本,提出了一种基于平稳性的二维训练图像分析方法.基于纹理特征分析了训练图像的二阶统计特征以获取尺度平稳性;通过多点密度函数分析训练图像的多点统计特征,获取多点统计平稳性.实验结果表明:对于尺度平稳性和多点统计平稳性都相对较高的训练图像,重建得到的三维结构的局部渗透性和两点统计概率特性更接近于真实三维结构的相关特性.而多点统计平稳性较高但尺度平稳性较低的训练图像,与多点统计平稳性较低而尺度平稳性较高的训练图像相比,更接近于真实三维图像.因此,训练图像的平稳性评估对于三维重建的二维薄片图像选取具有重要作用,同时,高阶统计分析方法更为准确.     

多聚焦像素级数字图像信息融合

根据图像的清晰恢复结合数字图像处理的基本理论算法,采用多聚焦图像融合技术,提出基于像素级的多聚焦数字图像的融合算法.用统计学的方法计算出不同聚焦图像上的聚焦点,并将其取出合成为一张清晰图像.根据以上算法设计了图像融合软件,并进行了实验验证     

基于正弦图融合的金属伪影校正方法（英文）

针对金属伪影严重降低了CT图像中组织结构清晰度的问题,提出了基于正弦图融合的金属伪影校正方法.首先,通过预先设置的阈值对原始CT图像进行分割,得到金属图像.对原始CT图像和金属图像进行投影生成原始投影正弦图和金属投影正弦图.使用插值校正方法校正含有金属伪影的CT图像和均值滤波维持校正后CT图像的边界.滤波之后图像被投影成滤波图像正弦图.根据金属图像正弦图在原始正弦图和滤波后图像对应正弦图中的位置,分别得到正弦图P_M~D(在原始正弦图中)和P_M~C (在滤波后图像对应正弦图中).然后,按照一定的比例,将P_M~D 和P_M~C 融合成正弦图P_M~F ,并通过融合正弦图P_M~F ,P~O和P_M~D 得到最终正弦图.最后,用滤波反投影重建算法将最终正弦图重建成校正之后的图像,并将金属信息补偿到校正图像上.临床图像上的实验表明:与经典金属伪影校正方法相比,所提出的基于正弦图融合方法在金属伪影去除和组织结构特征保存方面能够得到更好的效果.     

基于CT、MRI断层图像的人体三维建模

人体三维几何建模及其可视化显示是生物医学学科的一项基础性研究。CT和MRI等医学成像技术的断层图像可为三维建模提供基础。章论述和介绍了基于CT、MRI断层图像的三维建模系统的关键技术和相关实现原理。     

利用扭曲向列型液晶光阀实现对数字全息再现像质的改善

提出一种提高全息图像的清晰度的新方法.即原始图像经过微分运算和计算全息的变换后,得到它的边缘全息图,然后对该边缘全息图的对比度进行调节,再通过液晶光阀的调制作用,得到这幅调整后的全息图的再现像.这样得到的全息图的再现像比原来边缘全息图更为清晰.文中推导了黑色和白色的灰度调节函数,同时借助实验数据利用Matlab推导了光强与灰度的近似公式,并利用这些公式对图形进行了改善,实现了更为清晰的全息再现演示.     

基于局部目标特征提取和NSCT变换的多聚焦图像融合算法

使用多聚焦图像融合(Multi focus image fusion,MFIF)方法的最终目的是增加LR多聚焦图像中的高频成分,该方法采用低分率(low resolution,LR)多聚焦图像恢复技术让高分辨率(high resolution,HR)多聚焦图像恢复.在IEDCNN参考模型的基础上采用多聚焦图像融合模型,并利用局部目标特征提取和NSCT变换相结合的动态模型结构,还在图像融合时引入了蚁群算法.经过实验表明:本文采用的方法比IEDCNN更具优势,尤其是图像融合效果,清晰度更佳,同时也让振铃效应得到了有效的抑制.     

基于三维柏林噪声与位移贴图的真实感水面生成算法

提出了基于三维柏林噪声与位移贴图的真实感水面生成算法.三维柏林噪声图像可以看作是一系列连续的二维噪声图像,使用位移映射将这一系列的噪声图像作用到水面网格上,作为水面网格点的高度值的激励源.最终模拟了一个连续动态变化的水面环境.     

Environment map building and localization for robot navigation based on image sequences

SLAM is one of the most important components in robot navigation. A SLAM algorithm based on image sequences captured by a single digital camera is proposed in this paper. By this algorithm, SIFT feature points are selected and matched between image pairs sequentially. After three images have been captured, the environment＇s 3D map and the camera＇s positions are initialized based on matched feature points and intrinsic parameters of the camera. A robust method is applied to estimate the position and orientation of the camera in the forthcoming images. Finally, a robust adaptive bundle adjustment algorithm is adopted to optimize the environment＇s 3D map and the camera＇s positions simultaneously. Results of quantitative and qualitative experiments show that our algorithm can reconstruct the environment and localize the camera accurately and efficiently.     

Web-based interactive visualization of 3D video mosaics using X3D standard

INTRODUCTION The visualization of roadside buildings in virtual space using synthetic photorealistic view is one of the common methods for representing the background scene of Car Navigation Systems (CNS) and Internet map services. Since most of these background scenes are implemented by two-dimensional (2D) images, they may give somewhat monotonous impression due to fixed viewpoint position and orientation. For more interactive visualizations that enable arbitrary ad- justment of the …     

基于数学形态学的锅炉火焰图像骨架提取

本基于二值数学形态学，建立了数字图像骨架提取方法，并给出了相应的算法．同时，对结构元素序列进行了研究，结构元素序列的选取是骨架提取的关键．最后，应用数字图像骨架提取算法，对锅炉火焰图像进行了处理．处理结果表明，所提取的骨架能够准确地描述火焰图像的几何形状．     

3D Reconstruction technique for tomographic PIV

Tomographic particle image velocimetry(Tomo-PIV) is a state-of-the-art experimental technique based on a method of optical tomography to achieve the three-dimensional(3D) reconstruction for threedimensional three-component(3D-3C) flow velocity measurements. 3D reconstruction for Tomo-PIV is carried out herein. Meanwhile, a 3D simplified tomographic reconstruction model reduced from a 3D volume light intensity field with 2D projection images into a 2D Tomo-slice plane with 1D projecting lines, i.e., simplifying this 3D reconstruction into a problem of 2D Tomo-slice plane reconstruction, is applied thereafter. Two kinds of the most well-known algebraic reconstruction techniques, algebraic reconstruction technique(ART) and multiple algebraic reconstruction technique(MART), are compared as well. The principles of the two reconstruction algorithms are discussed in detail, which has been performed by a series of simulation images, yielding the corresponding reconstruction images that show different features between the ART and MART algorithm, and then their advantages and disadvantages are discussed. Further discussions are made for the standard particle image reconstruction when the background noise of the pre-initial particle image has been removed. Results show that the particle image reconstruction has been greatly improved. The MART algorithm is much better than the ART. Furthermore, the computational analyses of two parameters(the particle density and the number of cameras), are performed to study their effects on the reconstruction. Lastly, the 3D volume particle field is reconstructed by using the improved algorithm based on the simplified 3D tomographic reconstruction model, which proves that the algorithm simplification is feasible and it can be applied to the reconstruction of 3D volume particle field in a Tomo-PIV system.     

Simulation of turbulent combustion flame feature based on fractal theory for SI engines

The flame structure of gasoline engine is complicated and has the characteristic of fractal geometry. A fractal combustion model was used to simulate the engine working cycle. Based on this model, the fractal dimension and laminar flame surface area of turbulent premixed flames were studied under different working conditions. The experimental system mainly includes an optical engine and a set of photography equipment used to shoot the images of turbulent flame of spark-ignition engine. The difference box-counting method was used to process 2D combustion images. In contrast to the experimental results, the computational results show that the fractal combustion model is an effective method of simulating the engine combustion process. The study provides a better understanding for flame structure and its propagation.     

Simulation study on radiative imaging of combustion flame in furnace

Radiative imaging of combustion flame in furnace of power plant plays an increasingly important role in combustion diagnosis. This paper presents a new method for calculating the radiative imaging of three-dimensional (3D) combustion flame based on Monte Carlo method and optical lens imaging. Numerical simulation case was used in this study. Radiative images were calculated and images obtained can not only present the energy distribution on the charge-coupled device (CCD) camera target plane but also reflect the energy distribution condition in the simulation furnace. Finally the relationships between volume ele- ments and energy shares were also discussed.     

序列二维医学图象的体绘制法

Volume rendering of 3D data sets composed of sequential 2D medical images has become an important branch in image processing and computer graphics.To help physicians fully understand deep-seated human organs and focuses(e.g.a tumour)as 3D structures.in this paper,we present a modified volume rendering algorithm to render volumetric data,Using this method.the projection images of structures of interest from different viewing directions can be obtained satisfactorily.By rotating the light source and the observer eyepoint,this method avoids rotates the whole volumetric data in main memory and thus reduces computational complexity and rendering time.Experiments on CT images suggest that the proposed method is useful and efficient for rendering 3D data sets.     

Relighting multiple color textures

INTRODUCTION When constructing complete 3D models, multi- ple color texture maps are often necessary. Suppose that the object of interest is fixed and images of it are acquired by one or more cameras from different po- sitions and directions. The viewpoints and the illu- mination may be different for each acquisition. The geometric model of the object may come from merging of several 3D models. The 2D-3D registra-tion between the textures and the geometric model is done by a previous st…     

基于粗糙集和遗传算法在锅炉火焰图像分割中的研究

粗糙集善于从不明确数据集中发现有价值信息,锅炉火焰图像不确定信息量较为复杂,利用粗糙集并结合遗传算法对火焰图像进行分割优化,通过Matlab进行仿真,仿真结果表明,粗糙集工具结合遗传算法提高了图像分割的准确性,具有很好应用价值。