Augment low-field intra-operative MRI with preoperative MRI using a hybrid non-rigid registration method

Background

Preoperatively acquired diffusion tensor image (DTI) and blood oxygen level dependent (BOLD) have been proved to be effective in providing more anatomical and functional information; however, the brain deformation induced by brain shift and tumor resection severely impairs the correspondence between the image space and the patient space in image-guided neurosurgery.

Method

To address the brain deformation, we developed a hybrid non-rigid registration method to register high-field preoperative MRI with low-field intra-operative MRI in order to recover the deformation induced by brain shift and tumor resection. The registered DTI and BOLD are fused with low-field intra-operative MRI for image-guided neurosurgery.

Results

The proposed hybrid registration method was evaluated by comparing the landmarks predicted by the hybrid registration method with the landmarks identified in the low-field intra-operative MRI for 10 patients. The prediction error of the hybrid method is 1.92 ± 0.54 mm, and the compensation accuracy is 74.3 ± 5.0%. Compared to the landmarks far from the resection region, those near the resection region demonstrated a higher compensation accuracy (P-value = .003) although these landmarks had larger initial displacements.

Conclusions

The proposed hybrid registration method is able to bring preoperatively acquired BOLD and DTI into the operating room and compensate for the deformation to augment low-field intra-operative MRI with rich anatomical and functional information.     

Fall detection for multiple pedestrians using depth image processing technique

A fall detection method based on depth image analysis is proposed in this paper. As different from the conventional methods, if the pedestrians are partially overlapped or partially occluded, the proposed method is still able to detect fall events and has the following advantages: (1) single or multiple pedestrian detection; (2) recognition of human and non-human objects; (3) compensation for illumination, which is applicable in scenarios using indoor light sources of different colors; (4) using the central line of a human silhouette to obtain the pedestrian tilt angle; and (5) avoiding misrecognition of a squat or stoop as a fall. According to the experimental results, the precision of the proposed fall detection method is 94.31% and the recall is 85.57%. The proposed method is verified to be robust and specifically suitable for applying in family homes, corridors and other public places.     

基于多混沌系统的彩色图像加密方法

为了提高彩色图像加密的安全性和加密性能,设计了一种基于多混沌系统的彩色图像加密方法.将一个彩色图像分解为R、G、B三个灰度图像,使用MD5算法动态生成加密算法的初始值,然后使用三种不同的基于混沌的加密结构对三个图像进行加密.对R图像使用Feistel结构加密,其中Feistel结构的S一盒由Logistic混沌序列和Hyperhenon混沌序列组合产生;对G图像使用由Lorenz系统产生加密序列对图像进行代替和置换操作的加密结构;对B图像使用由分段线性混沌映射产生加密序列,然后加密图像的加密结构,再把加密后的图像结合起来生成加密后的图像.理论分析和实验结果表明,该加密方法能够较为有效地保证彩色图像加密的安全性.     

基于Gabor多尺度空间的不变兴趣点检测

针对以往仿射不变兴趣点的特征尺度不能直接断定的问题, 提出一种基于Gabor多尺度空间的不变兴趣点检测算法。该算法主要包括三个步骤：应用Gabor滤波器组与图像卷积建立图像Gabor多尺度空间; 通过极大值准则检测兴趣点并直接断定特征尺度; 采用二阶矩矩阵描述兴趣点局部结构。实验结果表明, 相比较其他Hessian-Affine、MSER等算法, 该算法在图像模糊和JPEG压缩情况下可重复率和可匹配率均取得最好结果, 是一种能有效直接提取特征尺度的兴趣点检测算法。     

基于反调和平均的张量形态学算子模型及其在彩色图像边缘检测中的应用

旨在建立基于张量模型的彩色图像形态学算子。通过将HSV彩色空间下的三维向量利用二阶对称正定张量进行建模表示, 提出一种基于反调和平均的彩色形态学算子。在此基础上, 借助于张量相似性度量构造基于参考张量的形态学算子, 该算子能够实现滤波的输入保持。通过基于形态学梯度的边缘检测实验验证模型的可行性和有效性。对比实验表明基于张量的彩色形态学模型相对于原有的向量模型在图像处理中具有更好的结果。     

遥感器像移与图像参数的分析建模研究

卫星平台运动会产生恶化像质的像移,像移是影响遥感图像质量的重要因素.针对这种情况,建立了像移和图像参数的关系模型,可通过卫星生成图像的内在特征定量研究卫星成像过程中产生的像移,而不必通过产生像移的卫星平台等因素建立物理模型来研究像移.首先模拟了遥感器在轨成像过程,通过人工制造出影响因素的干扰获得了仿真图像数据,然后定量分析了像移与图像参数之间的相关性,选出了相关性较强且适合建模的图像参数,最后通过对这些图像参数构成的向量的降维得到了反映图像质量的综合参数,并建立了综合参数与像移的关系模型,以通过图像综合参数来计算像移.实验结果表明,这种模型能够很好地反映图像综合指标与像移之间的关系.该研究通过图像参数定量研究像移,为在轨运行卫星的遥感器像移与外景条件下的遥感图像要素的相关建模研究提供了部分技术支持.     

一种新的基于压缩传感理论解决Pan-Sharpening问题的算法

Pan-Sharpening问题是将高分辨率全色图像和低分辨率多光谱图像融合,重构出高分辨率多光谱图像的过程。压缩传感理论解决了相应目标函数的稀疏解问题(求解含有l0范数的极小化问题)。利用压缩传感理论,提出一种新的解决Pan-Sharpening问题的算法:以离散余弦变换矩阵作为字典进行重构。然后针对重构过程中产生的块效应、色彩偏差和洗涤效应等问题进行相应的处理。QuickBird卫星图像的实验结果表明:与经典算法进行对比,该算法在效果上有一定的优势。     

基于Linux+ARM的多格式视频图像采集系统的设计

设计了基于S3C2440微处理器的视频图像采集系统,详细分析了该系统的启动过程。在一块裸板上搭建起整个视频图像采集系统。结合V4L2技术完成了支持YUV、MJPEG图像格式的USB摄像头和支持RGB图像格式的CMOS摄像头的多格式视频图像采集系统。测试结果表明,该系统可以在LCD显示屏上动态显示多格式摄像头所采集到的视频图像,且视频图像具有清晰度高、流畅性好、稳定性强等特点,具有进一步推广使用的广阔前景。     

Pattern recognition for characterization of pharmaceutical powders

A large part of pharmaceutical manufacturing involves the use of particulate materials. It is well known that both particle size and shape affect the physical characteristics of tablets. An image processing and analysis algorithm based on the invariant image moment technique was developed in this work to determine the particle shape by comparing features (moments) extracted from templates to those extracted from each of the objects in the image. First it determines the particle shape (rectangle, circle, etc.) and then calculates its specific dimensions (diameter, aspect ratio).The statistical validation of the vision system obtained a repeatability of 0.0012 in and 0.5% relative standard deviation and accuracy within 0.1 to 0.9% of the average value considered as true value. Also the pattern recognition technique indicated high precision and accuracy for images containing particles with some level of contact between them. The shape recognition of microcrystalline cellulose (MCC) indicated that particles of equant and acicular shape as defined by USP are predominant. The results suggest that image processing and analysis would be a suitable tool for pharmaceutical process analytical technologies (PAT) and process optimization.     

基于粒子群三次样条优化的图像边缘描述算法

针对手工选点生成的三次样条曲线无法准确地描述图像边缘的问题,以粒子群三次参数样条曲线优化算法为基础对现有的图像边缘算法进行了改进。在图像边缘上取若干点作为样条曲线的型值点,利用粒子群算法对这些型值点进行优化,使得到的样条曲线对图像边缘有很好的贴合。实验表明,该算法不仅能够快速拟合样条曲线,而且优化后的样条曲线形态能较好地贴合图像边缘轮廓。