Infrared Image Series Automatic Registration and a History & Tendency Analysis

Infrared camera has been widely used in electrical power systems for predictive and preventive thermal fault diagnosis. Without historical IR images being considered, most thermograph analysis used at present are concerned with static image. The dynamic history and tendency analysis which compares the newly taken image with historical images is particularly attractive. Conclusions can be drawn from temperature variations rather than just from anomalous thermal pattern. However the spatial distortions prevent such dynamic analysis from being practicable. So the need to register infrared images arises. This paper proposes an automatic registration algorithm based on closed temperature contour matching for infrared image series of a target equipment taken at different times. Centroid Distance Function is used in image registration for the first time. The algorithm can remove spatial distortions, bring temperature variations into prominence and endow infrared camera with dynamic analysis ability. Experiment shows that the algorithm is simple, effective and extensible.     

基于二进制小波变换和改进SPIHT算法的图像编码方法

提出了一种基于二进制小波变换和改进SPIHT算法的图像编码方法.二进制小波变换将图像从实数域变换到实数域,消除像素之间的空间冗余性,得到了具有整数准确度的紧致描述.针对传统SPIHT算法解码图像视觉效果差的缺点,提出了改进方法.根据图像分析结果,将二进制小波变换变换系数按视觉重要性重新排序,通过对视觉重要系数优先编码,把量化误差集中在视觉不敏感区域,从而在不影响编码率失真性能的同时,有效地提高了解码图像的视觉效果.实验结果表明,和其它流行的编码算法相比,本文算法对不同性质的图像具有最优的编码性能和视觉效果.     

Entropy-Based Combined Metric for Automatic Objective Quality Assessment of Stitched Panoramic Images

Quality assessment of stitched images is an important element of many virtual reality and remote sensing applications where the panoramic images may be used as a background as well as for navigation purposes. The quality of stitched images may be decreased by several factors, including geometric distortions, ghosting, blurring, and color distortions. Nevertheless, the specificity of such distortions is different than those typical for general-purpose image quality assessment. Therefore, the necessity of the development of new objective image quality metrics for such type of emerging applications becomes obvious. The method proposed in the paper is based on the combination of features used in some recently proposed metrics with the results of the local and global image entropy analysis. The results obtained applying the proposed combined metric have been verified using the ISIQA database, containing 264 stitched images of 26 scenes together with the respective subjective Mean Opinion Scores, leading to a significant increase of its correlation with subjective evaluation results.     

An algorithm for encryption of secret images into meaningful images

Image encryption algorithms typically transform a plain image into a noise-like cipher image, whose appearance is an indication of encrypted content. Bao and Zhou [Image encryption: Generating visually meaningful encrypted images, Information Sciences 324, 2015] propose encrypting the plain image into a visually meaningful cover image. This improves security by masking existence of encrypted content. Following their approach, we propose a lossless visually meaningful image encryption scheme which improves Bao and Zhou's algorithm by making the encrypted content, i.e. distortions to the cover image, more difficult to detect. Empirical results are presented to show high quality of the resulting images and high security of the proposed algorithm. Competence of the proposed scheme is further demonstrated by means of comparison with Bao and Zhou's scheme.     

BMEFIQA: Blind Quality Assessment of Multi-Exposure Fused Images Based on Several Characteristics

A multi-exposure fused (MEF) image is generated by multiple images with different exposure levels, but the transformation process will inevitably introduce various distortions. Therefore, it is worth discussing how to evaluate the visual quality of MEF images. This paper proposes a new blind quality assessment method for MEF images by considering their characteristics, and it is dubbed as BMEFIQA. More specifically, multiple features that represent different image attributes are extracted to perceive the various distortions of MEF images. Among them, structural, naturalness, and colorfulness features are utilized to describe the phenomena of structure destruction, unnatural presentation, and color distortion, respectively. All the captured features constitute a final feature vector for quality regression via random forest. Experimental results on a publicly available database show the superiority of the proposed BMEFIQA method to several blind quality assessment methods.     

Distortion-invariant pattern recognition using synthetic discriminant function based multiple phase-shifted-reference fringe-adjusted joint transform correlation

This paper proposes a novel pattern recognition system for invariance to noise and distortions. The technique first generates a synthetic discriminant function of the target image from its different distorted versions. It then takes four different phase-shifted versions of the reference image, which are individually joint transform correlated with the given input scene. Thus the proposed algorithm produces a single cross-correlation signal corresponding to each potential target. Also a fringe-adjusted filter is designed to generate a delta-like correlation peak with high discrimination between the signal and the noise. The pattern recognition system is also designed for the identification of multiple targets belonging to multiple reference objects simultaneously in a given input scene. The proposed technique is investigated using computer simulation including real-life images in different complex environments.     

基于梯度一致性约束的多光谱/全色影像最大后验融合方法

多光谱/全色影像融合可以得到高空间分辨率的多光谱影像,在影像解译和分类等方面具有十分重要的意义。提出一种基于梯度一致性约束的遥感影像融合方法。该方法在最大后验概率框架下,通过梯度一致性约束建立理想高空间分辨率多光谱影像和全色影像之间的关系,并结合多光谱影像观测模型和Huber-Markov影像先验,构建融合目标函数,最后采用梯度下降法求解得到融合影像。本文方法在目标函数中引入了梯度一致性约束,克服了现有的同类方法受限于波段数量的缺陷,同时在求解中自适应确定每个波段的迭代步长,充分顾及了各波段的光谱特性,从而既确保了融合影像的光谱信息保真度,也提高了融合影像的空间信息融入度。通过IKONOS和WorldView-2影像对该方法进行了验证,并和GS,AIHS和AMBF等融合方法从定性和定量两方面进行了比较分析。实验结果表明,相比于其他方法,该方法可以在更好保持光谱信息的同时增强影像的空间分辨率,具有更广泛的适用范围和更佳的融合效果。     

No-Reference Quality Assessment for 3D Synthesized Images Based on Visual-Entropy-Guided Multi-Layer Features Analysis

Multiview video plus depth is one of the mainstream representations of 3D scenes in emerging free viewpoint video, which generates virtual 3D synthesized images through a depth-image-based-rendering (DIBR) technique. However, the inaccuracy of depth maps and imperfect DIBR techniques result in different geometric distortions that seriously deteriorate the users’ visual perception. An effective 3D synthesized image quality assessment (IQA) metric can simulate human visual perception and determine the application feasibility of the synthesized content. In this paper, a no-reference IQA metric based on visual-entropy-guided multi-layer features analysis for 3D synthesized images is proposed. According to the energy entropy, the geometric distortions are divided into two visual attention layers, namely, bottom-up layer and top-down layer. The feature of salient distortion is measured by regional proportion plus transition threshold on a bottom-up layer. In parallel, the key distribution regions of insignificant geometric distortion are extracted by a relative total variation model, and the features of these distortions are measured by the interaction of decentralized attention and concentrated attention on top-down layers. By integrating the features of both bottom-up and top-down layers, a more visually perceptive quality evaluation model is built. Experimental results show that the proposed method is superior to the state-of-the-art in assessing the quality of 3D synthesized images.     

Affine and polynomial mutual information coregistration for artifact elimination in diffusion tensor imaging of newborns

We have investigated the use of two different image coregistration algorithms for identifying local regions of erroneously high fractional anisotropy (FA) as derived from diffusion tensor imaging (DTI) data sets in newborns. The first algorithm uses conventional affine registration of each of the diffusion-weighted images to the unweighted (b = 0) image for each slice, while the second algorithm uses second-order polynomial warping. Similarity between images was determined using the mutual information (MI) criterion, which is the preferred 'cost' criterion for coregistration of images with significantly different image intensity distributions. We have found that subtle differences exist in the FA values resulting from affine and second-order polynomial coregistration and demonstrate that nonlinear distortions introduce artifacts of spatial extent similar to real white matter structures in the newborn subcortex. We show that polynomial coregistration systematically reduces the presence of erroneous regions of high FA and that such artifacts can be identified by visual inspection of FA maps resulting from affine and polynomial coregistrations. Furthermore, we show that nonlinear distortions may be particularly pronounced when acquiring image slices of axial orientation at the height of the nasal cavity. Finally, we show that third-order polynomial MI coregistration (using the images resulting from second-order coregistration as input) has no observable effect on the resulting FA maps.     

基于YCbCr颜色空间的Retinex低照度图像增强方法研究

针对Retinex理论的低照度图像增强算法中光照图像估计问题,提出一种基于YCbCr颜色空间的低照度图像增强方法.该方法将原始低照度图像从RGB(Red Green Blue)颜色空间转换到YCbCr颜色空间,提取该空间中Y分量构建为原始光照图像分量L1(x,y),并对L1(x,y)进行Gamma校正得到增强的光照图像分量L2(x,y),经Retinex模型得到增强图像R(x,y),采用多尺度细节增强方法对图像R(x,y)进行细节增强,得到最终增强图像Re(x,y).实验结果表明,所提方法不仅能有效提升亮度,避免亮度和色彩失真,增强了图像的细节信息并获得了更好的视觉效果,而且运行速度快.     

An unsupervised deep learning technique for susceptibility artifact correction in reversed phase-encoding EPI images

Echo planar imaging (EPI) is a fast and non-invasive magnetic resonance imaging technique that supports data acquisition at high spatial and temporal resolutions. However, susceptibility artifacts, which cause the misalignment to the underlying structural image, are unavoidable distortions in EPI. Traditional susceptibility artifact correction (SAC) methods estimate the displacement field by optimizing an objective function that involves one or more pairs of reversed phase-encoding (PE) images. The estimated displacement field is then used to unwarp the distorted images and produce the corrected images. Since this conventional approach is time-consuming, we propose an end-to-end deep learning technique, named S-Net, to correct the susceptibility artifacts the reversed-PE image pair. The proposed S-Net consists of two components: (i) a convolutional neural network to map a reversed-PE image pair to the displacement field; and (ii) a spatial transform unit to unwarp the input images and produce the corrected images. The S-Net is trained using a set of reversed-PE image pairs and an unsupervised loss function, without ground-truth data. For a new image pair of reversed-PE images, the displacement field and corrected images are obtained simultaneously by evaluating the trained S-Net directly. Evaluations on three different datasets demonstrate that S-Net can correct the susceptibility artifacts in the reversed-PE images. Compared with two state-of-the-art SAC methods (TOPUP and TISAC), the proposed S-Net runs significantly faster: 20 times faster than TISAC and 369 times faster than TOPUP, while achieving a similar correction accuracy. Consequently, S-Net accelerates the medical image processing pipelines and makes the real-time correction for MRI scanners feasible. Our proposed technique also opens up a new direction in learning-based SAC.     

基于Rician分布的散斑噪声的超声图像模拟算法*

为了衡量各种去噪算法的性能,在干净无噪声的图像上添加接近真实且可控的散斑噪声是非常重要的,提出一种基于Rician分布的不完全发育的斑点噪声的超声图像模拟算法。该算法考虑到了声波区域中包含孤立的强散射体的情况,同时结合了超声成像的扫描过程。以合成图像和肾脏图像为体模进行了模拟实验,并对最终生成的伪超声图像进行了噪声分布统计及拟合检验。实验结果表明该算法生成的伪超声图像在视觉上和理论上都接近真实的超声图像。     

A geometric correction based robust color image watermarking scheme using quaternion Exponent moments

Most of the existing color image watermarking schemes was designed to mark grayscale images or use the color components, which ignore the significant correlation between different color channels. Recently, several approaches were developed to process the color channels intrinsically, but they always cannot counterattack geometric distortions. It is a challenging work to design a robust color image watermarking scheme against geometrical distortions. In this paper, we propose a geometric correction based robust color image watermarking approach using quaternion Exponent moments (QEMs). The novelty of our approach is that (1) the QEMs are derived to deal with the color images in a holistic manner, and (2) the QEMs are exploited for estimating the geometric distortions parameters in order to permit watermark extraction. Experimental results confirm the validity of our approach and its higher robustness against geometrical distortions compared to alternative watermarking methods in the literature.     

基于成像系统物理特性的多光谱图像与全色波段图像融合

提出了一种基于成像系统物理特性的多光谱图像与全色波段图像融合算法。该算法采用àtrous小波变换提取全色波段图像的空间细节信息,并将提取的空间信息按照一定的注入模型调整后添加到各波段多光谱图像中去,得到具有高空间分辨力的多光谱图像。注入模型充分考虑了各波段成像传感器的相对光谱响应函数、地表物体对各波段的光谱反射率以及各波段的辐射调整系数等成像系统的物理特性,使融合后的多光谱图像在显著提高空间质量的同时,最大可能地保留了原始多光谱图像的光谱特性。对IKONOS卫星遥感影像的融合实验结果表明,该算法在光谱保留和空间质量提高方面较其它基于小波变换的融合算法都具有更高的性能。     

基于级联神经网络的实用型三维复合不变性多目标识别

以三 飞机模型作为待识别目标,模型真实场景,对用于多目标分类识别的级联神经网络重新进行了研究。实验发现畜产品上降的主要原因是实际采集的目标发生的复杂畸变与计算机模拟产生的效果并不一样。用采集得到的目标图像作为训练样本,对网络重新构造和训练,取得了好的实验结果。分析了其中涉及到目标定位、图像分割等图像预处理问题,提出了一种基于二值图像开矿学腐蚀运算的快速目标检测位法,可快速有效地对目标进行检测定 。     

Image fusion through local feature extraction by using multi-scale top-hat by reconstruction operators

The purpose of image fusion is to combine useful image features of different original images into the final fusion image, which will produce one useful result image for different applications. One of the main difficulties of image fusion is extracting useful image features of different original images. In some cases, useful image features are local image features of the whole image. To efficiently extract local image features and produce an efficient fusion result, an image fusion algorithm based on the extracted local image features by using multi-scale top-hat by reconstruction operators is proposed in this paper. Firstly, multi-scale local feature extraction using multi-scale top-hat by reconstruction operators is discussed. Then, based on the extracted multi-scale local features of different original images, the useful image features for image fusion are constructed. Finally, the constructed useful image features for image fusion are combined into the final fusion image. Experimental results on different types of images show that, the proposed algorithm performs well for image fusion.     

基于随机点扩散函数的多帧湍流退化图像自适应复原方法

针对湍流退化图像随机性的问题,提出了一种基于随机点扩散函数的多帧湍流退化图像自适应复原方法。首先介绍了随机点扩散函数的图像退化模型,并分析了点扩散函数随机性对图像复原造成的影响,建立了基于随机点扩散函数的多帧图像退化模型。在此基础上,建立了基于多帧退化图像的全变分复原模型,利用前向后向算子分裂法对模型进行求解,提高了算法的运算效率。然后,提出了一种新的自适应正则化参数选取方法,该方法利用全变分复原模型的目标函数计算正则化参数,当正则化参数收敛时,复原图像的峰值信噪比达到最大值,因此利用目标函数的相对差值作为自适应算法迭代终止的条件,可以获得最佳复原效果。最后通过实验分析,算法中退化图像的帧数应不大于10帧。实验结果表明:当取10帧退化图像时,AFBS算法运算时间与单帧的FBS算法相当,信噪比增益为1.4 dB。本文算法对图像噪声有明显的抑制作用,对湍流退化图像可以获得较好的复原效果。     

基于非采样Contourlet变换的遥感图像融合算法

为了使融合后的多光谱图像在尽可能保持原始多光谱图像光谱特性的同时,显著提高空间分辨力,提出了一种基于非采样Contourlet变换(NSCT)的遥感图像融合算法。算法首先对全色波段图像进行非采样Contourlet变换,得到全色波段图像的低频子带系数和各带通方向子带系数;然后针对多光谱图像的每一个波段,将其进行双线性插值后作为融合后多光谱图像的低频子带系数,对全色波段图像的各带通方向子带系数采用基于成像系统物理特性的注入模型(调整系数)进行局部调整后,作为融合后多光谱图像的各带通方向子带系数,从而得到融合后多光谱图像的非采样Contourlet变换系数;最后再经非采样Contourlet逆变换得到该波段具有高空间分辨力的多光谱图像。采用IKONOS卫星遥感图像进行了仿真实验,实验结果表明,该算法在光谱保留和空间质量提高方面优于其它传统的遥感图像融合算法。     

On the use of the FLAIR technique to improve the correction of eddy current induced artefacts in MR diffusion tensor imaging

Eddy current induced geometric distortions can only be accurately corrected in brain diffusion-weighted echo-planar (DW-EP) images for b-values less than approximately 300 s mm(-2) using the iterative cross-correlation (ICC) algorithm. This is due to the difference in signal intensity of the cerebrospinal fluid (CSF) compartment in the diffusion-weighted and baseline T(2)-weighted echo-planar (T2W-EP) images. At larger values of b, image misalignment artefacts can, however, be removed by directly correlating CSF-suppressed T2W-EP images with non-CSF-suppressed and CSF-suppressed DW-EP images. Separate phantom experiments can also be performed to provide eddy current calibration data. Here the ability of these methods to remove eddy current induced artefacts from DW-EP images collected in volunteer diffusion tensor imaging (DTI) experiments is investigated. Monte Carlo simulations show that in order for the ICC algorithm to produce accurate estimates of the eddy current induced distortions at b-values greater than 1000 s mm(-2), the degree of CSF suppression should be greater than approximately 80%. This condition is typically met for FLAIR inversion times between 0.5 and 0.8 of the spin-lattice relaxation time of CSF. In volunteer studies the most complete image realignment was provided by direct correlation of CSF-suppressed T2W-EP and DW-EP images acquired in the FLAIR DTI experiment. These results indicate that although calibration data obtained from brain or phantom images can significantly reduce eddy current induced distortions, the optimum image realignment achievable using post-processing methods is likely to be that obtained by direct image warping techniques.     

基于Markov随机场的自适应正则化三维显微图像复原

提出了基于马尔可夫随机场模型的正则化因子自适应调整三维显微图像复原算法,并用模拟序列样本和真实生物样本进行了实验.为了保持复原图像的边缘等细节信息,以Markov随机场模型作为图像的先验概率模型,对代价函数添加边缘约束惩罚项.其中,正则化因子在迭代过程中自适应地进行更新.实验结果表明此算法在对原始图像进行估计的同时,能够有效地保留图像的边缘等细节信息.而EM算法虽然能够有效地去除层间干扰,却丢失了大量的细节信息.