基于梯度塔形分解的多传感器图像融合

给出了基于梯度塔形分解的像素级图像融合新方法,基本思想是,先对原图像进行梯度塔形分解;其次,按照融合规则,采用基于区域特性的加权算子去构造融合图像对应的梯度金字塔;最后,通过逆梯度塔形变换重构融合图像,该方法被成功地用于图像的融合处理,实验结果表明,该融合方法十分有效,获得的融合图像更符合人的视觉特性,更有利于机器视觉。     

Fusion algorithm for multisensor images based on discrete multiwavelet transform

The authors review the notion of multiwavelets and describe the use of the discrete multiwavelet transform (DMWT) in image fusion processing. Multiwavelets are extensions from scalar wavelets, and have several advantages in comparison with scalar wavelets. Multiwavelet analysis can offer more precise image analysis than wavelet multiresolution analysis. A novel fusion algorithm is presented for multisensor images based on the discrete multiwavelet transform that can be performed at pixel level. After the registering of source images, a pyramid for each source image can be obtained by applying decomposition with multiwavelets in each level. The multiwavelet decomposition coefficients of the input images are appropriately merged and a new fused image is obtained by reconstructing the fused multiwavelet coefficients. This image fusion algorithm may be used to combine images from multisensors to obtain a single composite with extended information content. The results of experiments indicate that this image fusion algorithm can provide a more satisfactory fusion outcome.     

基于NSCT的多光谱和全色图像的融合

提出了一种基于无下采样Contourlet变换的多光谱和全色图像的融合方法.该方法在对多光谱影像进行IHS变换的基础上,对多光谱的I分量和高分辨率的全色影像分别进行无下采样Contourlet变换(NSCT),然后对分解得到的近似分量以及各层金字塔各方向的细节分量利用本文提出的一定的融合准则分别对近似分量和细节分量进行影像融合,最后通过无下采样Conlourlet逆变换得到新的I分量,与H,S分量一起还原到RGB空间,最终得到融合后的高分辨率多光谱彩色图像.本文采用一组多光谱图像和全色图像数据进行融合实验,其实验融合图像的目视效果和统计指标均优于传统的IHS融合方法、小波融合方法以及Contourlet变换方法.     

QuickBird Image Fusion by a Multirresolution-Multidirectional Joint Image Representation

A new fusion methodology for Multiespectral and Pancromatic images, has been proposed. This methodology is based on a joint multiresolution-multidirectional representation of the source images. For that an unique directional filters bank of low computacional complexity has been used.This kind of image representation allows an appropriated selection of the information extracted from the source images, avoiding some of the limitations inherent to other multiresolution fusión methods. The final aim is to obtain fused images with a high spectral and spatial quality simultaneously. The source images corresponds to the captured by Quickbird satellite (panchromatic and multispectral). The high quality of the obtained results shows the potential of the joint multiresolution-multidirectional representation for images fusion.     

基于滚动引导滤波的红外与可见光图像融合算法

针对红外与可见光图像融合时,易产生细节丢失、噪声抑制不佳等问题,本文提出了一种改进的滚动引导滤波融合算法。该算法充分利用了滚动引导滤波边缘和局部亮度保持特性,在通过均值滤波将输入图像分解为基础层与细节层的基础上,结合滚动引导滤波与高斯滤波获取输入图像的显著图,利用不同尺度参数的引导滤波对显著图优化得到权重图,将权重图作为权重分别指导基础层与细节层的融合,最后联合融合后的子图重构得到融合图像。针对3类测试数据进行的融合实验表明,与非下采样轮廓波变换、基于引导滤波、基于显著性检测的两个尺度的图像融合等经典方法相比,本文方法得到的融合图像不但从主观视觉效果上细节信息更丰富、目标对比度加强,并且在非线性相关信息熵、相位一致性等6项客观评价指标上均具有较好的效果。     

基于多尺度和注意力模型的红外与可见光图像融合

针对红外与可见光图像在融合后容易出现伪影,小目标轮廓不清晰等问题,提出一种基于多尺度特征与注意力模型相结合的红外与可见光图像融合算法。通过5次下采样提取源图像不同尺度的特征图,再将同一尺度的红外与可见光特征图输入到基于注意力模型的融合层,获得增强的融合特征图。最后把小尺度的融合特征图进行5次上采样,再与上采样后同一尺度的特征图相加,直到与源图像尺度一致,实现对特征图的多尺度融合。实验对比不同融合框架下融合图像的熵、标准差、互信息量、边缘保持度、小波特征互信息、视觉信息保真度以及融合效率,本文方法在多数指标上优于对比算法,且融合图像目标细节明显轮廓清晰。     

基于对比度的多分辨图像融合

本文给出一种基于对比度的多分辨图像融合算法.首先,利用小波变换得到待融合图像的多分辨分析,同时得到图像的多分辨对比度序列;然后以对比度为判据,在图像的多分辨分析的相应各级上进行融合,得到融合图像的多分辨分析;最后,利用小波逆变换重构融合图像.算法使用同一场景可见光图像和红外图像进行了验证,结果表明,融合图像完好地显示了源图像各自的信息.     

联合多流融合和多尺度学习的卷积神经网络遥感图像融合方法

为尽可能保持原始低分辨率多光谱(LRMS)图像光谱信息的同时,显著提高融合后的多光谱图像的空间分辨率,该文提出一种联合多流融合和多尺度学习的卷积神经网络遥感图融合方法.首先将原始MS图像输入频谱特征提取子网得到其光谱特征,然后分别将通过梯度算子处理全色图像得到的梯度信息和通过卷积后的全色图像与得到的光谱特征图在通道上拼...     

Perceptual-based image fusion for hyperspectral data

Three hierarchical multiresolution image fusion techniques are implemented and tested using image data from the Airborne Visual/Infrared Imaging Spectrometer (AVIRIS) hyperspectral sensor. The methods presented focus on combining multiple images from the AVIRIS sensor into a smaller subset of images white maintaining the visual information necessary for human analysis. Two of the techniques are published algorithms that were originally designed to combine images from multiple sensors, but are shown to work well on multiple images from the same sensor. The third method presented was developed specifically to fuse hyperspectral images for visual analysis. This new method uses the spatial frequency response (contrast sensitivity) of the human visual system to determine which features in the input images need to be preserved in the composite image(s) thus ensuring the composite image maintains the visually relevant features from each input image. The image fusion algorithms are analyzed using test images with known image characteristics and image data from the AVIRIS hyperspectral sensor. After analyzing the signal-to-noise ratios and visual aesthetics of the fused images, contrast sensitivity based fusion is shown to provide excellent fusion results and, in every case, outperformed the other two methods     

Context enhancement through infrared vision: a modified fusion scheme

In the night vision applications, visual and infrared images are often fused for an improved awareness of situation or environment. The fusion algorithms can generate a composite image that retains most important information from source images for human perception. The state of the art includes manipulating in the color spaces and implementing pixel-level fusion with multiresolution algorithms. In this paper, a modified scheme based on multiresolution fusion is proposed to process monochrome visual and infrared images. The visual image is first enhanced based on corresponding infrared image. The final result is obtained by fusing the enhanced image with the visual image. The process highlights the features from visual image, which is most suitable for human perception.     

基于VGG网络的双波段图像融合方法

针对红外与可见光图像中物体信息具有各自优点的情况,提出了基于VGG网络的红外与可见光图像融合方法来提高对夜间或复杂背景情况下的物体检测识别能力。首先将图像分别输入到一个经过训练得到的VGG网络中,经过不同的卷积层提取各自的特征图;然后将特征图经过ZCA白化处理,去除冗余信息;再通过归一化处理,将特征图的维度降到二维,并通过双三次插值法将其缩放到与源图像尺寸一致;最后通过加权取平均得到融合后的图像。实验结果表明,本文的方法在第四和第五层卷积得到的融合结果优于前三层的融合结果。同时,本文融合方法与其他3种融合方法相比视觉效果较好,在标准差、平均梯度、相关系数、熵值等评价指标上分别平均提升了12.79 %、11.04 %、9.94 %和2.54 %,并且在融合时间上保持在1秒以内。这说明该方法融合效果较好,速度较快,能够较多地保留红外与可见光图像信息和较好地提升目标的显著性。     

A new flexible bi-orthogonal filter design for multiresolutionfilterbanks with application to image compression

We propose a new fast method with great potential for multiresolution pyramid decomposition of signals and images. The method allows unusual flexibility in choosing a filter for any task involving the multiresolution analysis and synthesis. Using our method, one can choose any low-pass filter for the multiresolution filtering. This method enabled us to choose the best filters for set partitioning in hierarchical trees (SPIHT) image compression for the corresponding support sizes. The compression results for our seven tap filters are better than those of the 9/7 wavelet filters and approximately the same as those of the 10/18 filters, while at the same time our seven tap filters are faster than 10/18 filters     

针对人类视觉系统的双波段红外图像融合算法研究

提出了一种针对人类视觉系统的双波段红外图像融合算法。首先采用非下采样Contourlet变换把源图像分解成尺多度、多方向上的子带,然后采用绝对值最大算子的方法对高频部分进行融合,采用基于区域方差的方法对低频部分进行融合,最后进行重构得到融合图像。对试验中采集的双波段红外图像进行了融合实验,并将融合效果与梯度金字塔变换方法和小波变换方法进行了比较。实验结果表明,针对人类视觉系统的双波段红外图像融合算法有效可行,可以获得比常用的算法更好的融合效果。     

一种基于变分的红外和微光遥感图像融合模型

为了充分利用红外和微光遥感图像中的互补信息,使其便于目视解译,提出了 一种基于变分的图像融合方法。该变分模型定义了细节注入项和结构保真项,在保持红外和微光图像 光谱特性的同时,还改进了融合图像的空间细节和结构特性;引入了正则化能量项,保证了泛函最优解 的平滑性。基于梯度下降流,通过数值迭代获得了融合图像。实验结果表明,该模型能够获取兼具 丰富细节信息和光谱信息的融合图像。与Laplacian金字塔分解方法和多孔小波方法相比,本文 方法具有更佳的融合性能。     

一种基于稀疏表示的红外与微光图像的融合方法

根据人类视觉系统及信号的过完备稀疏表示理论,提出了一种基于稀疏表示的红外与微光图像融合算法。该方法首先把图像分割成部分重叠的图像块,由正交匹配追踪算法完成图像块的稀疏分解;然后采用最大值融合准则选择融合系数并完成图像块的重构,得到融合结果图像。实验结果表明,本文算法的融合效果优于小波变换法、Laplacian塔型方法以及PCA方法等传统融合方法。     

基于对比度金字塔与双边滤波的非对称红外与可见光图像融合

为了同时保留红外图像的特征信息和可见光图像的细节信息,提出了一种基于对比度金字塔的非对称红外与可见光图像融合方法。首先,使用对比度金字塔对红外与可见光图像进行高频与低频信息分解,然后对高频部分采用绝对值取大方法进行融合,对于低频部分采用基于双边滤波的方法对红外与可见光图像进行非对称的处理;其次,使用对比度金字塔的逆变换得到融合后图像。对融合图像进行主观视觉和客观指标评价,结果表明该算法在突出目标特征信息和保留细节特征方面表现优异。     

Multi-focus image fusion with joint guided image filtering

Multi-focus image fusion is the activity of synthesizing multiple images of different focusing settings to construct a fully focused image. Many of the latest methods for image fusion rarely consider the structural differences between the guidance image and the input image, and do not retain well the important source image features while producing a fully focused image. To address this issue, a method exploiting a combination of static and dynamic filters (SDF) is proposed herein. This combination has good edge smoothing characteristics and strong robustness against artifacts such as gradient inversion and global strength migration. First, SDF is utilized in order to decompose the source image into structure and texture layers. Secondly, a morphological gradient operator filter is used to calculate the significance map of different levels of the source. Thirdly, the maximum pixel value of the significance map is used to construct the binary decision graph of the two source images. Then, the structure and texture layers are fused with the aid of the binary decision graph, and subsequently the final fusion image is created by combining the fused structure layer and texture layer. This process ensures that spatial consistency is preserved. Tests on grayscale and color multi-focus image sets show that the proposed method has better performance than that of any of the existing methods according to both objective and subjective evaluation.     

多尺度特征融合的膨胀卷积残差网络高分一号影像道路提取

针对全卷积神经网络多次下采样操作导致的道路边缘细节信息损失和道路提取不准确的问题,本文提出了多尺度特征融合的膨胀卷积残差网络高分一号影像道路提取方法。首先,通过目视解译的方法制作大量的道路提取标签数据;其次,在残差网络ResNet-101的各个残差块中引入膨胀卷积和多尺度特征感知模块,扩大特征点的感受野,避免特征图分辨率减小和道路边缘细节特征的损失;然后,通过叠加融合和上采样操作将各个尺寸的道路特征图进行融合,得到原始分辨率大小的特征图;最后,将特征图输入Sigmoid分类器中进行分类。实验结果表明:本文方法的提取精度优于经典全卷积神经网络模型,准确率达到了98%以上,有效保留了道路的完整性及其边缘的细节信息。     

基于多尺度对比度塔和方向滤波器组的图像融合

本文从图像的特性出发,将方向信息与多分辨分析相结合,提出了基于方向滤波器组的融合思想.进而考虑到人眼的视觉特性,利用对比度塔分解(CP)捕获图像的低频分量,利用方向滤波器组(DFB)获得图像的高频分量,提出了一种基于多尺度对比度塔和方向滤波器组(CPDFB)的图像融合方法,同时分析了该方法的融合机理和计算复杂度,较好地实现了复杂图像的融合.仿真实验验证了其可行性和有效性.     

自注意力引导的红外与可见光图像融合算法

为确保源图像中的显著区域在融合图像保持显著,提出了一种自注意力引导的红外与可见光图像融合方法。在特征学习层引入自注意力学习机制获取源图像的特征图和自注意力图,利用自注意力图可以捕获到图像中长距离依赖的特性,设计平均加权融合策略对源图像的特征图进行融合,最后将融合后的特征图进行重构获得融合图像。通过生成对抗网络实现了图像特征编码、自注意力学习、融合规则和融合特征解码的学习。TNO真实数据上的实验表明,学习到注意力单元体现了图像中显著的区域,能够较好地引导融合规则的生成,提出的算法在客观和主观评价上优于当前主流红外与可见光图像融合算法,较好地保留了可见光图像的细节信息和红外图像的红外目标信息。