被动式太赫兹图像目标检测研究

研究提出一种被动式太赫兹扫描成像的目标检测方法。用0.22 THz被动式扫描成像系统采集原始图像,在滤波去噪后用针孔像分析法获得太赫兹扫描成像系统的点扩展函数,进而用Lucy-Richardson算法重构图像,通过灰度变换和边缘检测增强图像对比度和目标分辨力。实验结果表明,算法能有效改善被动式太赫兹图像质量,提高成像系统探测隐藏可疑物的能力。     

Comparative analysis of the resolving power of optoelectronic imaging systems upon incoherent and coherent illumination of objects

Results of calculations of the resolving power (according to Foucault) of optoelectronic imaging systems with incoherent and coherent illumination under the conditions of atmospheric phase distortions are presented. It is suggested that incoherent optical images are formed in the case of incoherent illumination and speckle images are formed in the case of coherent illumination. The shape of objects observed is estimated from these images. The resolving powers are shown to be practically the same in the formation of incoherent optical images and speckle images in the case of strong phase distortions and smoothing of images by a matrix photodetector. At the same time, the limiting resolving power in the formation of incoherent optical images is twice as large as in the formation of speckle images.     

不均匀光照和重影的仪表图像二值化方法

高压计量仪表识别过程中需要对图像进行二值化操作,然而拍摄的仪表图像多出现光照不均和数字重影现象,导致传统方法对仪表图像的二值化困难。为此,提出一种基于卷积神经网络的二值化方法,用于对复杂光照下含数字重影的仪表图像二值化。该网络使用的数据集为真实环境下的仪表图像,首先对输入的图像进行降维提取特征,然后反卷积重建图像前景,最后输出二值图。将设计的网络与传统的二值化方法进行对比,实验结果表明,经该网络训练得到的二值图数字清晰且无重影,且测得的交并比（IoU）平均值为95.12,与样本标签图像的相似度最高,能够有效解决复杂环境下有重影的仪表图像二值化问题。     

Analysis of speckle images to assess surface roughness

Digital speckle images are photographed for different aluminum rough surfaces using a CCD camera. The obtained speckle images are fed to a PC and analyzed making use of the MATLAB program. The computerized binary images are investigated. The signal-to-noise ratio is computed from these numerical images. It is shown that the surface roughness of the examined surfaces is dependent upon the degree of agglomeration of the speckle images.     

Deep unregistered multi-contrast MRI reconstruction

Multiple magnetic resonance images of different contrasts are normally acquired for clinical diagnosis. Recently, research has shown that the previously acquired multi-contrast (MC) images of the same patient can be used as anatomical prior to accelerating magnetic resonance imaging (MRI). However, current MC-MRI networks are based on the assumption that the images are perfectly registered, which is rarely the case in real-world applications. In this paper, we propose an end-to-end deep neural network to reconstruct highly accelerated images by exploiting the shareable information from potentially misaligned reference images of an arbitrary contrast. Specifically, a spatial transformation (ST) module is designed and integrated into the reconstruction network to align the pre-acquired reference images with the images to be reconstructed. The misalignment is further alleviated by maximizing the normalized cross-correlation (NCC) between the MC images. The visualization of feature maps demonstrates that the proposed method effectively reduces the misalignment between the images for shareable information extraction when applied to the publicly available brain datasets. Additionally, the experimental results on these datasets show the proposed network allows the robust exploitation of shareable information across the misaligned MC images, leading to improved reconstruction results.     

一种基于特征点跟踪的电子稳像算法

提出一种基于特征点跟踪的稳像算法。该算法从参考帧图像中提取出一组特征点,然后在后续帧中进行基于Kalman滤波的特征点的跟踪,匹配参考帧图像中特征点的坐标和当前图像中基于Kalman滤波修正后得到的特征点的坐标,并通过仿射模型求出位移量及旋转参数,最后进行反向变换,从而得到稳定的视频图像。实验结果表明:该算法稳像效果好,运算复杂度低,且具有较强的鲁棒性。     

一种从粗到精的红外和可见光卫星图像配准方法

气象卫星所携带的多种传感器可以获得可见光、红外、多光谱等多模态的卫星图像,目前处理这些多模态图像的一个重要手段是数据融合分析方法,而获取不同模态图像空间对应关系的图像配准是数据融合分析的前提和基础。针对多模态气象卫星图像的配准问题,重点研究红外图像和可见光图像的配准问题,并根据红外图像和可见光图像的特点,提出了一种由粗到精的两阶段配准方法。在粗配准阶段,将Fourier-Mellin变换应用于红外和可见光图像的边缘图像上,并通过变换图像在频域的关系实现了图像配准仿射变换参数的快速计算;在精配准阶段,基于图像的Harris算子检测红外图像和可见光图像的特征点,并通过特征点局部区域的互相关函数实现特征点的匹配,最终通过匹配特征点求得精确配准的变换参数。文章提出的由粗到精的图像配准方法,有效结合了Fourier-Mellin变换对边缘图像配准的高效性和Harris算子图像配准的准确性,是红外和可见光图像配准的一种新方法。利用FY-2D气象卫星获取的红外和可见光图像进行了配准实验,实验结果表明所提出的方法具有良好的鲁棒性和较高的配准精度。     

基于非采样Contourlet变换高分辨率遥感图像配准

为了提高高分辨率遥感图像配准的精确度,将非采样Contourlet变换应用于高分辨率遥感图像配准算法中.首先对高分辨率遥感图像进行非采样Contourlet变换.利用非采样Contourlet变换的平移不变性在变换域提取图像的边缘并选择合适的阈值准确地得到图像的边缘特征点.然后利用归一化互相关匹配法和概率支撑法对特征点进行匹配.最后通过三角形局部变换映射甬数实现图像配准.实验结果表明,该方法更能准确地提取高分辨率遥感图像的特征点,大大提高了正确匹配的概率,与基于小波方法的图像配准效果相比有更高的准确性和稳健性.     

A hybrid strategy for correcting geometric distortion in echo-planar images

A hybrid strategy for geometric distortion correction of echo-planar images is demonstrated. This procedure utilizes standard field mapping for signal displacement correction and the so-called reverse gradient acquisition for signal intensity correction. (The term reverse gradient refers to an acquisition of two sets of echo-planar images with phase encoding gradients of opposite polarity.) The hybrid strategy is applied to human brain echo-planar images acquired with and without diffusion-weighting. A comparison of the hybrid distortion corrected images to those corrected with standard field mapping only demonstrates much better performance of the hybrid method. A variant of the hybrid method is also demonstrated which requires the acquisition of only one pair of opposite polarity images within a set of images.     

A post-processing technique for displaying vessels from routine fast-spin-echo images: MRI-derived angiography.

Fast-spin-echo magnetic resonance (MR) images are routine components of a standard MR brain examination. On these images, blood vessels are visible as black flow void. We report that by applying an enhancement filter to a stack of routine fast-spin-echo MR images, projected angiographic images can be generated. The vascular detail in the projected image is similar to that observed in a phase-contrast image. In addition to its advantage in obtaining vessel information from routine images, the proposed post-processing technique is fast, easy to implement and completely automatic. These images provide additional vessel information that is useful when MR angiography is unavailable or as an aid in planning dedicated MR angiographic studies.     

基于二通道不可分加性小波的多光谱图像融合

提出了伸缩矩阵为[1,1;1,-1]的不可分小波的一种构造方法,并把它应用于多光谱图像与高分辨力全色图像的融合中。提出了一种二通道对称的不可分小波滤波器组的构造方法,设计出多组具有紧支撑、对称性和正交性的不可分小波6×6滤波器组,利用此类滤波器组中的低通滤波器对图像进行加性分解与重构。采用三种模式NAWS、NAWRGB和NAWL对ETM 及其它类多光谱图像与高分辨力全色图像进行融合研究。实验结果表明,该方法对多光谱图像与高分辨力全色图像的融合有较好的融合效果,与传统的融合方法相比,该方法既能保持多光谱图像的光谱信息,又能保持高分辨力全色图像的高分辨力特性,克服了传统张量积小波融合方法不能得到高分辨力图像的不足,并能节约运算量。     

Improvement of the acquisition of a large amount of MR images on a conventional whole body system

Modern whole body MR systems are equipped with echo-planar-imaging capability, which allows the measurement of a single slice in a fraction of a second or of thousands of images in few minutes. A considerable restriction to the acquisition of series containing large amounts of images in patient examinations is the time-consuming data handling time of the images at conventional systems, which includes the time to insert the images into the systems database. We propose the arrangement of several images on a new image with a large matrix size like a mosaic. The handling time depends mostly on the number of images without consideration of their matrix size. Therefore, image handling is strongly reduced by the use of such mosaic images.     

Effect of Scattering on Spectral Imaging of Exposed Cortical Tissue for Brain Function Measurement

Optical imaging of an exposed cortex for brain function measurement is an attractive method for both clinical and physiological investigations. Spectral imaging of the exposed cortical tissue enables measurement of the activity-dependent changes in oxy- and deoxy-haemoglobin independently. Since the diffusely reflected light from the exposed cortex is highly scattered in the cortical tissue, the wavelength dependence of the optical properties of the tissue is likely to affect the images of oxy- and deoxy-haemoglobin changes obtained from the spectral images. In this study, the spectral images of the model of exposed cortical tissue are calculated by Monte Carlo simulation to investigate the influence of blurring by tissue scattering on the brain function measurement by the imaging. The images of the change in oxy- and deoxy-haemoglobin obtained from the spectral images at near infrared wavelengths are broadened and blurred due to the scattering in the cortical tissue. The influence of cross talk is more significant in the image calculated from the spectral images at the visible wavelengths. Inappropriate choice of the spectral range of images might increase cross talk and error in the images of oxy- and deoxy-haemoglobin changes.     

Approach to Generate Radiometric Images

A new technique to generate radiometric images from the optical images is introduced in this paper. This technique can be applied in generating the source images for purpose of image-matching navigation. For comparison, two radiometric images are presents, one is imaged by a 8mm ground-based radiometer and the other is generated by this technique.     

Synthesis of three-dimensional phase images of nanoobjects: Numerical simulation

A way to synthesize three-dimensional phase-diffraction images of nanoobjects is proposed based on the numerical method of backward propagation of a field from the best focusing plane. The obtained phase images are used for solving the applied problem of object defect finding by methods of pattern recognition. The probability of defect detection by phase images is shown to be higher than for amplitude images.     

Depth-control method for integral imaging

We present a method for controlling the depth of three-dimensional (3D) images reconstructed by integral photography. Incoherent light is reflected from 3D objects, propagates through a lens array, and is captured as the first elemental images by a capturing device. The second elemental images of the 3D images are generated by numerical processing from the first elemental images in accordance with the desired depth. The optical reconstruction of 3D images at the desired depth by the second elemental images is confirmed experimentally.     

Linear restoration of partially coherent images

The method of superposition of images used previously to restore coherent or incoherent degraded images, is adapted to the restoration of partially coherent images.     

航天光学遥感器在轨MTF评价的一元方法

通过对航天光学遥感器MTF模型和遥感图片的分析,从图像中提取出与MTF有关的特征信息,采用人工神经网络(ANN)作为工具,将这些特征信息作为ANN的输入向量。在对大量MTF已知的遥感图片进行训练后,ANN可以对未知的遥感图片进行MTF测试。这种方法被称为MTF的一元评价方法,即通过对遥感器传输下来的任意一幅地面景物图像进行MTF的在轨评价,无需在地面铺设特定形状靶标或已知的参考图片。实验结果表明,平均评价误差约为5%,具有很强的抗噪声能力。     

一种改进的基于小波方向对比度的无缝拼接算法

为了提高图像无缝拼接效果,对现有平滑算法进行分析,在此基础上提出一种基于小波方向对比度的中缝平滑算法。该算法首先对源图像进行基于统计参数的预处理,使得图像整体亮度趋于一致,然后利用小波变换对图像进行多分辨率分解,对于分解后的图像提出采用一种基于改进的小波方向对比度的规则进行平滑处理。文中还通过几种图像客观评价标准对仿真结果进行了评价,结果表明：该方法在两图像灰度相差较大时也能够取得较好的平滑效果,对光照具有鲁棒性,并且平滑后的图像可以保留原始图像的细节信息,使得图像在保证清晰度的前提下达到无缝拼接。     

应用径向基函数去除数字图像中的椒盐噪声

去除图像中的椒盐噪声可转化为二维曲面的重建问题。选用Multi-Quadric函数对图像中损失的信息构造插值格式,自动选择待插值点和插值参考点并求解插值方程组得到处理后的图像。实验证明,本方法可以在很少破坏图像细节的情况下去除大部分甚至全部噪声,并且在噪声密度非常大的情况下仍然可以还原相当多的图像信息。对噪声密度为50％和90％的单色Lena图像进行处理,该方法得到的信噪比比自适应中值滤波高6dB以上。