融合对象性和视觉显著度的单目图像2D转3D

受对象性测度和视觉显著度的启发,提出一种适用于单目图像2D转3D的对象窗深度中心环绕分布假设,给出融合对象性测度和视觉显著度的单目图像深度估计算法。首先计算图像的视觉显著度并将其映射成深度;其次在图像上随机采样若干个窗,并计算这些窗的对象性测度;再次,定义一个能量函数用于度量深度和对象性测度对彼此的影响程度,并通过迭代优化的方法改进深度和对象性测度的估计结果;最后,根据深度信息进行3D视频合成。实验结果表明,融入对象性测度信息后,显著改进了基于视觉显著度2D转3D的深度估计质量,保证了估计深度在对象边界处的不连续过渡和其他区域的平滑过渡。     

基于颜色和梯度方向共生直方图的图像检索

为有效融合图像的形状、颜色等特征,提出一种基元梯度方向共生直方图和颜色直方图的彩色图像检索算法。首先将图像划分为4×4互不重叠的基元,并根据基元的梯度平均幅值把基元分为视觉均衡基元和视觉非均衡基元;接着计算视觉均衡基元的平均颜色值,并将其在HSV空间中量化到72色颜色空间,构建视觉均衡基元颜色直方图作为彩色图像的局部颜色描述子;然后将图像中各像素点颜色值在HSV空间量化到72色颜色空间,获得图像的全局颜色直方图;最后在视觉非均衡基元中构建梯度方向共生直方图描述图像的形状特征。融合局部颜色描述子、全局颜色直方图和形状特征构成彩色图像检索特征矢量。实验结果表明,所提算法能够准确描述彩色图像的颜色和形状特征,具有很好的旋转不变性和尺度不变性。相似性度量非常有效,查全率和查准率均有较大提高。     

基于结构统计的纹理分类

针对纹理图像在视觉上的结构表现，选用能说明任意二值纹理的基元序列，将基元所对应的区域进行平移或伸张以覆盖整个纹理图像，完成所有基元对应区域的象素统计，获得能说明纹理中各结构比例及粒度的统计曲线。用此方法重新分类了Brodatz的111个纹理图片。实验证明视觉上相似的纹理有相似的统计曲线(相同或接近的峰、谷或转折位置)。比较其他分类结果，该方法更具有视觉接近性。     

基于多特征融合的图像检索算法

为有效提取和描述图像特征,提高图像检索性能,提出一种基于纹理、颜色和形状多特征融合的图像检索算法。检测彩色图像的边缘,对其进行变换得到基元图像。遍历基元图像得到基元共生矩阵,对每个基元求梯度值得到基元梯度直方图。将彩色图像量化到64色颜色空间,得到对应的颜色直方图。利用上述3个特征量描述图像特征,并用于图像检索。实验结果表明,与BCTF和MCM算法相比,该算法的查全率和查准率较高,计算复杂度较低。     

基于内唇轮廓标定的唇印提取算法

针对现有特征唇印提取中信息处理量较大、识别率不高等问题,本文提出了基于内唇轮廓的特征唇印提取算法。算法首先在基元图像上对内唇轮廓特征点进行标定,建立基元唇印模型,然后通过均值计算构造普通模型,并利用Gabor变换对基元唇印和普通模型进行联合特征信息提取,最后通过相似度对比选择出特征唇印,以实现身份识别。仿真实验验证,本文的特征唇印提取算法在较低时空消耗下,具有较高的识别率,具有有效性和可用性。     

基于颜色和基元特征的图像检索

为了有效地提高图像检索性能,提出一种基于基元特征和颜色特征相结合的图像检索方法。检测彩色图像的边缘梯度,对其进行变换获得基元图像。以基元图像的重心为基准点计算各点到重心的距离,求得基元重心距离直方图。对基元图像中的每一个基元在4个不同方向投影,求得基元投影直方图。将彩色图像在RGB空间量化到64色颜色空间,得到彩色图像的颜色直方图。利用以上3个直方图描述图像特征,并用于图像检索。实验结果表明,该算法的查全率和查准率均有较大提高。     

裸眼3D液晶屏FPGA时序控制技术研究

裸眼3D液晶屏为图像显示载体,可把左右眼的图像准确送至双眼形成立体视觉。左右眼图像信号使用FPGA来准确控制帧同步、行同步与像素时钟,使图像像素准确地传输到屏幕相应位置。研究了裸眼3D屏的数据接口要求,通过裸眼3D显示算法和利用硬件描述语言建立了图像数据流传输模型,得出了裸眼3D字幕显示的可行逻辑设计。系统经过实验验证,运行稳定。     

基于基元结构的图像检索以及相关反馈

在本文中,我们提出了一种新颖的特征描述符用于图像检索,即基元结构直方图。它使用了颜色和边缘方向信息来描述图像的基元结构信息。在特征提取时,我们使用更符合人眼的视觉认知机制的HSV颜色空间。在相关反馈阶段,我们对权重进行不断的更新,以达到更好的检索效果。     

高效RGB格雷码与相移法结合的三维重建技术

针对传统相移法在三维(3D)轮廓重建时相位展开耗时长和精度较低的问题,提出了一种新的高效率的三原色(RGB)格雷码编码方式对相位解包裹。用彩色摄像机分别拍摄经由物体高度信息调制后变形的正弦光栅图像,每幅光栅图片相移π/2;在正弦光栅灰度值最大处投射RGB格雷码对条纹定级;根据左右图像的相位展开值和双目视觉原理对左右视图进行匹配,恢复模型的3D轮廓。提出的方法在64级条纹之内将解码需要的投影条纹数量减少到3幅,测量速度和精度都有明显提高。将提出的方法应用在模型表面测量和标准球测量实验中,结果表明,提出的方法提高了重建速率且精度小于0. 03 mm。     

2D到3D图像颜色空间层转换技术研究及实现

从3D相关技术入手,初步讨论了人类视觉及相关转换技术。接着重点描述基于3d红蓝立体原理的图像转换原理,从图像颜色空间角度,着重讨论2D转3D的相关技术及研究现状,并基于电视类节目图像给出一套完整系统设计及实现,并给出相应的实验结果。     

基于立体视觉的玉米叶片形态测量与三维重建

提出一种基于立体视觉的玉米叶片形态测量和重建的方法。利用双目立体视觉系统获取玉米叶片的两幅图像,通过图像分割技术和边缘检测算法对每幅图像中的玉米叶片进行边缘提取;利用极线约束和彩色图像RGB值对图像进行匹配,计算出叶片边缘的三维坐标,从而恢复叶片的三维边缘;利用对叶片边缘的恢复技术,对叶片曲面进行三维重建;根据恢复的区域点云,可以测出任意两点的空间距离,实现了对叶片的三维测量。试验结果表明,此方法能够很好地恢复玉米叶片的三维信息,为玉米叶片三维形态的无损、快速检测监测提供了新的方法。     

Reconstruction of 3D human body pose from stereo image sequences based on top-down learning

This paper presents a novel method for reconstructing a 3D human body pose from stereo image sequences based on a top-down learning method. However, it is inefficient to build a statistical model using all training data. Therefore, the training data is hierarchically divided into several clusters to reduce the complexity of the learning problem. In the learning stage, the human body model database is hierarchically constructed by classifying the training data into several sub-clusters with silhouette images. The data of each cluster in the bottom level is represented by a linear combination of examples. In the reconstruction stage, the proposed method hierarchically searches a cluster for the best matching silhouette image using a silhouette history image (SHI). Then, the 3D human body pose is reconstructed from a depth image using a linear combination of examples method. By using depth information to reconstruct 3D human body pose, the similar poses in silhouette images are estimated as different 3D human body poses. The experimental results demonstrate that the proposed method is efficient and effective for reconstructing 3D human body poses.     

IKONOS卫星图像的快速三维场景重建

IKONOS卫星立体像对一般经过图像匹配,采用RPC模型,通过繁杂的多项式迭代获取DEM,实现三维重建,其计算花费很大。在对城市三维重建时,认为城市局部的地面是平坦的,有近似相同的海拔,城市的三维信息集中就是建筑物的三维信息,针对IKONOS影像的特点,利用几何光学成像模型快速计算同名点的相对大地坐标,实现城市场景的快速三维重建。实验结果表明,场景重建结果有较好的精度,满足观测和处理的需求,实现了快速处理和分析环境数据。     

A multimodal approach for 3D face modeling and recognition using 3D deformable facial mask

We present a multimodal approach for face modeling and recognition. The algorithm uses three cameras to capture stereo images, two frontal and one profile, of the face. 2D facial features are extracted from one of the frontal images and a dense disparity map is computed from the two frontal images. Using the extracted 2D features and their corresponding disparities, we compute their 3D coordinates. We next align a low resolution 3D mesh model to the 3D features, re-project its vertices onto the frontal 2D image and adjust its profile silhouette vertices using the profile view image. We increase the resolution of the resulting 2D model at its center region to obtain a facial mask model covering distinctive features of the face. The 2D coordinates of the vertices, along with their disparities, result in a deformed 3D mask model specific to a given subject’s face. Our method integrates information from the extracted facial features from the 2D image modality with information from the 3D modality obtained from the stereo images. Application of the models in 3D face recognition, for 112 subjects, validates the algorithm with a 95% identification rate and 92% verification rate at 0.1% false acceptance rate.

基于高空间分辨率与立体像对遥感数据的建筑物三维信息提取

准确获取建筑的三维分布信息对于城市规划与管理、灾害风险评估与防范以及灾后救助等都具有非常重要的意义。针对目前建筑物信息提取研究集中于二维平面信息提取,三维信息提取研究较少,且方法自动化程度较低,实用性和和推广性不足,提出了综合立体像对和高空间分辨率两种遥感数据进行建筑物三维信息提取的方法。首先,基于小波变换融合方法对GeoEye\|1高空间分辨率全色和多光谱影像进行融合,然后运用面向对象方法对融合后的高空间分辨率遥感影像进行建筑物基底轮廓提取,再利用IRS\|P5立体像对反演地物高度,最后通过数据整合获得研究区建筑物的三维空间分布。研究结果表明:该方法可以充分利用不同遥感数据的优势,获得较高的提取精度;研究所需数据容易获取,方法具有较好的可操作性和推广性。     

Blind stereo image quality assessment inspired by brain sensory-motor fusion

The use of 3D and stereo imaging is rapidly increasing. Compression, transmission, and processing could degrade the quality of stereo images. Quality assessment of such images is different than their 2D counterparts. Metrics, which represent 3D perception by human visual system (HVS), are expected to assess stereoscopic quality more accurately. In this paper, inspired by brain sensory/motor fusion process, by combining the right and left images, we form two synthesized images. Effects of different structural distortions on statistical distributions of the synthesized images are analyzed. Based on the observed statistical changes, features are extracted from these synthesized images that can reveal type and severity of distortions. Then, we propose a stacked neural network model to learn the extracted features and accurately predict the quality of stereo images. This model is tested on 3D images of popular databases. Experimental results show the superiority of this method over state-of-the-art stereo image quality assessment approaches.     

Noise reduction in three-dimensional digital images

Two dimensional and three dimensional noise reduction techniques are used on real 3D images and compared. The comparison is based on the busyness of the resulting images and on their fidelity to the original images. The following methods, each with 2D and 3D versions, are reviewed: mean filtering, median filtering, nearest neighbor smoothing, selective averaging and maximum likelihood smoothing. The results suggest that the 3D techniques are more effective at removing noise and retaining image information content than the 2D techniques. The methods that produced the highest quality images were the nearest neighbor and maximum likelihood smoothing techniques. The mean and median filtering methods removed the most noise, but blurred the images. The selective averaging method provided intermediate results.     

基于射影不变量的三维重构

文中提出了利用射影不变量来求解基于图像对三维深度恢复问题。方法的基本思想是对于立体图像,利用密度段元素,引入了两个射影不变量来恢复密度段的深度信息。从这两个不变量,能推导立体图像中匹配的密度段对所满足的关系。利用这个关系,实现了密度段之间的匹配运算。这个方法能直接地从输入图像中得到密集和准确的深度,对变形的图像具有鲁棒性。     

Distortion conspicuity on stereoscopically viewed 3D images may correlate to scene content and distortion type

We describe two studies that were aimed towards increasing our understanding of how the visibility of distortions on stereoscopically viewed 3D images is affected by scene content and distortion types. By assuming that subjects' performance would be highly correlated with the visibility of local distorted patches, we analyzed subjects' performance in locating distortion patches when viewing stereoscopic 3D images. Subjects' performances are measured by whether they successfully locate a local distorted patch, the times they spent to finish the task, and subjective quality ratings given by subjects. The visual data used in this work are co‐registered stereo images with co‐registered “ground truth” range (depth) data. Varied statistical analysis methods were used to discuss the significance of our observations. Three observations are drawn from our analyses. First, blur, JPEG, and JP2K distortions in stereo 3D images may be suppressed if one of the left or right views is undistorted. Second, contrast masking ^a does not occur, or is reduced, while viewing white noise distorted stereo 3D images. Third, there is no depth/disparity masking effect when viewing stereo 3D images, but there may be (conversely) depth‐related facilitation ^b effects for blur, JPEG, and JP2K distorted stereo 3D images.