复杂地物条件下基于线特征的异源景象匹配

为了寻求应用于复杂地物条件下异源景象匹配的算法,使其满足尺度和旋转不变性,受视觉成像系统的启发,利用初始简图可以表征图像大部分信息的特点,提出了一种新的基于线段对的异源图像匹配算法。首先,提取并筛选图像中能够表征图像信息的线段;然后,利用线段自身信息及线段的相对位置关系构建线段对特征;接着,通过线段对之间的相似性对图像进行粗匹配;最后,利用线段之间的拓扑关系进行精匹配。实验表明,本文方法对具有旋转、缩放和平移变换的异源图像的匹配正确率达到了75%以上,运算时间是传统匹配算法的1/5左右,基本满足了异源景象匹配应用对算法实时性和准确性的要求。     

Improved keypoint descriptors based on Delaunay triangulation for image matching

In the current keypoint-based image matching methods, not all the keypoints can be reliably matched because of the influence of noise, illumination, and image distortion. In the paper, we propose a novel method based on the Delaunay triangulation to detect and remove possible mismatches. Given previously matched keypoints by a detector in two images, the proposed method utilizes four steps to remove the mismatched point pairs: first, triangulating keypoints in the reference image, and producing a graph consisting of edges that connects the keypoints; second, drawing a graph in the test image by connecting the corresponding points as same as in the reference image; third, detecting abnormal edges in the test image using special constraints; fourth, detecting and removing mismatched point pairs based on the abnormal edges. The experimental results show that the method can detect the mismatches accurately and improve the robustness of current matching algorithms.     

改进SIFT变换与客观评价结合的图像配准算法

针对SIFT(Scale Invariant Feature Transform)算子在大幅复杂图像中提取的过多不稳定特征点及在只有少量重合区域下图像配准过程中出现的过多误匹配,导致图像配准精度下降。提出一种改进的SIFT算法,在对目标图像提取SIFT特征后,利用双向BBF(Best-Bin-First)匹配算法对提取的特征点进行匹配,采用SIFT描述子的尺度以及梯度方向信息建立最小邻域匹配剔除误匹配点,通过随机抽取一致性算法(RANSAC)进一步筛选匹配点,并利用最小二乘法结合多项式近似拟合出变换模型,利用局部均方根有效值(RMS)评价映射矩阵与实际图像的误差,找出并删除引起误差的误匹配点,迭代至配准图像符合评价标准后,计算出精确变换模型。实验结果表明,该算法提高了大幅复杂图像在少量重合区域时的配准精度。     

基于Shape Context和尺度不变特征变换的多模图像自动配准方法

提出了基于修正的尺度不变特征变换（SIFT）特征提取和Shape Context特征描述算子相结合的多模图像自动配准算法,该算法利用修正的SIFT算法提取多模图像中的特征点,然后采用Shape Context算子描述特征点,利用特征点周围区域边缘点的梯度方向形成特征向量。采用欧氏距离作为匹配标准对多模图像中特征点进行初始匹配,然后通过RANSAC算法消除误匹配的特征点对,并采用最小二乘法计算仿射变换参数,最后通过仿射变换和双线性插值实现图像配准。对红外图像和可见光图像的配准实验结果表明了本算法的有效性和稳定性。     

Rician noise reduction in magnetic resonance images using adaptive non-local mean and guided image filtering

This paper proposes a Rician noise reduction method for magnetic resonance (MR) images. The proposed method is based on adaptive non-local mean and guided image filtering techniques. In the first phase, a guidance image is obtained from the noisy image through an adaptive non-local mean filter. Sobel operators are applied to compute the strength of edges which is further used to control the spread of the kernel in non-local mean filtering. In the second phase, the noisy and the guidance images are provided to the guided image filter as input to restore the noise-free image. The improved performance of the proposed method is investigated using the simulated and real data sets of MR images. Its performance is also compared with the previously proposed state-of-the art methods. Comparative analysis demonstrates the superiority of the proposed scheme over the existing approaches.     

基于线性不变矩的特征图象配准算法研究

针对图象差异较大的光学图象与合成孔径雷达图象(SAR)的特点,设计了一种基于特征的图象配准的算法,包括特征提取、匹配、控制点选取、变换系数计算和误差计算等步骤.在特征提取中,对于雷达图象需要进行预处理并采用LOG提取图象轮廓.在匹配算法中,以线性不变矩作为特征量,对现有基于图象轮廓特征的匹配算法进行了改进.采用上海浦东地区的Radarsat-1S2模式图象与LandsatTM-5波段图象作为待配准图象.实验结果表明,该方法可以较好地完成光学图象与SAR图象的配准,有着较高的应用价值.     

基于对数解调的递变能量CT成像方法

复杂结构件由于有效厚度差异大和成像系统动态范围受限, 单一能量下的投影数据信息不完整, 常用CT重建算法及不完全数据重建算法无法在数据缺失严重的情况, 有效实现复杂结构件的CT重建. 为此论文提出基于对数解调的递变能量CT成像方法. 该方法在分析直接高动态CT成像所存在问题的基础上, 提出利用对数变换压缩递变能量投影序列动态范围, 并利用现有的基于图像灰度一致性的融合方法, 计算融合加权系数, 再经常规重建算法实现复杂构件的CT成像. 论文并以某复杂仪表为对象, 进行实验, 相比传统的固定能量成像方式, CT信息完整, 质量高. 从而说明论文所提出的方法, 能够实现CT系统动态范围的扩展, 实现复杂结构件的高动态CT成像.     

基于相位一致性和Hough变换的多源图像配准方法

由于红外图像与可见光图像对比度不同,常用基于梯度幅值的特征匹配方法难以正确配准。在分析红外图像与可见光图像成像机制的基础上,提出了一种结合相位一致性边缘检测与Hough变换的多源图像配准新方法。该算法首先采用高通滤波和平台直方图均衡方法对红外图像进行预处理以提高红外图像的对比度,再利用具有图像对比度不变性的相位一致性边缘检测法提取两幅图像的边缘,结合Hough变换选取图像空间中最长的线作为特征,采用改进相位相关法作为相似性度量,在对数极坐标域下计算出两幅图像的几何变形参数。仿真实验结果表明,该方法能够以较高查准率实现红外与可见光图像自动配准,并具有较强的鲁棒性。     

基于相位一致性和Hough变换的多源图像配准方法

由于红外图像与可见光图像对比度不同,常用基于梯度幅值的特征匹配方法难以正确配准。在分析红外图像与可见光图像成像机制的基础上,提出了一种结合相位一致性边缘检测与Hough变换的多源图像配准新方法。该算法首先采用高通滤波和平台直方图均衡方法对红外图像进行预处理以提高红外图像的对比度,再利用具有图像对比度不变性的相位一致性边缘检测法提取两幅图像的边缘,结合Hough变换选取图像空间中最长的线作为特征,采用改进相位相关法作为相似性度量,在对数极坐标域下计算出两幅图像的几何变形参数。仿真实验结果表明,该方法能够以较高查准率实现红外与可见光图像自动配准,并具有较强的鲁棒性。     

Electro-optic holography fringe control using diode laser current modulation

Electro-optic holography (EOH) is a whole field, laser-based, vibration measurement technique. Highly dense, unresolvable fringe patterns often limit the EOH measurement range. This paper presents a technique for increasing the measurement range of an EOH system using diode laser current modulation. This technique is known as frequency translated electro-optic holography (FTEOH). Using diode laser current modulation, the fringe patterns are based on higher-order Bessel functions and the sensitivity of the EOH system can either be increased or decreased. The amount of sensitivity reduction is limited only by the frequency limit of the signal generator that modulates the current supplied to the diode laser. EOH and FTEOH experimental results are presented and shown to closely compare.     

Multi-sensor image fusion using discrete wavelet frame transform

An algorithm is presented for multi-sensor image fusion using discrete wavelet frame transform (DWFT).The source images to be fused are firstly decomposed by DWFT. The fusion process is the combining of the source coefficients. Before the image fusion process, image segmentation is performed on each source image in order to obtain the region representation of each source image. For each source image, the salience of each region in its region representation is calculated. By overlapping all these region representations of all the source images, we produce a shared region representation to label all the input images. The fusion process is guided by these region representations. Region match measure of the source images is calculated for each region in the shared region representation. When fusing the similar regions, weighted averaging mode is performed; otherwise selection mode is performed. Experimental results using real data show that the proposed algorithm outperforms the traditional pyramid transform based or discrete wavelet transform (DWT) based algorithms in multi-sensor image fusion.     

Imaging freely moving subjects using continuous interleaved orthogonal magnetic resonance imaging

Magnetic resonance imaging has shown increasing clinical utility for the diagnosis of abnormalities in fetal development. MRI is not yet as effective for fetal imaging as ultrasound because of the difficulty of imaging freely moving subjects. We describe a design approach to overcome this difficulty. By interleaving orthogonal images of a subject, it is possible to rapidly and interactively localize the scan plane in a moving subject and confirm image plane orientation relative to the subject. We derive the equations necessary to optimize the tip angles for the acquisition of the orthogonal images so as to minimize artifact in the main image despite the long T1 of a fluid environment (e.g., amniotic fluid). To fully utilize the orthogonal images for rapid localization, it is critical to minimize the delay between acquisition and display, and to avoid segmented reconstruction techniques that are commonly used in high frame rate imaging. We demonstrate that this approach can be used to perform interactive scan plane localization on a moving subject and can obtain high temporal resolution images while confirming the image plane orientation relative to the subject.