基于可逆神经网络的神经辐射场水印

针对面向隐式表达的神经辐射场的3D模型的版权问题,将神经辐射场水印的嵌入与提取视为一对图像变换的逆问题,提出了一种利用可逆神经网络水印保护神经辐射场版权方案。利用二维图像的水印技术以实现对三维场景的保护,通过可逆网络中的正向过程在神经辐射场的训练图像中嵌入水印,利用逆向过程从神经辐射场渲染出的图像提取水印,实现对神经辐射场以及三维场景的版权保护。但神经辐射场在渲染过程中会造成水印信息丢失,为此设计了图像质量增强模块,将渲染图像通过神经网络进行恢复然后再进行水印提取。同时在每个训练图像中均嵌入水印来训练神经辐射场,实现多个视角下均可提取水印信息。实验结果表明了提出的水印方案,达到版权保护的目的,证明方案的可行性。     

A region-based multi-sensor image fusion scheme using pulse-coupled neural network

For most image fusion algorithms split relationship among pixels and treat them more or less independently, this paper proposes a region-based image fusion scheme using pulse-coupled neural network (PCNN), which combines aspects of feature and pixel-level fusion. The basic idea is to segment all different input images by PCNN and to use this segmentation to guide the fusion process. In order to determine PCNN parameters adaptively, this paper brings forward an adaptive segmentation algorithm based on a modified PCNN with the multi-thresholds determined by a novel water region area method. Experimental results demonstrate that the proposed fusion scheme has extensive application scope and it outperforms the multi-scale decomposition based fusion approaches, both in visual effect and objective evaluation criteria, particularly when there is movement in the objects or mis-registration of the source images.     

Training pattern replication and weighted class allocation in artificial neural network classification

In some image classifications the importance of classes varies, and it is desirable to weight allocation to selected classes. Often the desire is to weight allocation in favour of classes that are abundant in the area represented by an image at the expense of the less abundant classes. If there is prior knowledge on the distribution of class occurrence, this weighting can be achieved with widely used statistical classifiers by setting appropriate a prioriprobabilities of class membership. With an artificial neural network, the incorporation of prior knowledge is more problematic. An approach to weight class allocation in an artificial neural network classification by replicating selected training patterns is presented. This investigation focuses on a series of classifications in which some classes were more abundant than others, but the same number of training cases were available for each class. By replicating the training patterns of abundant classes the representation of the abundant classes in the training set is increased, reflecting more closely the relative abundance of the classes in an image. Significant increases in classification accuracy were obtained by replicating the training patterns of abundant classes. Furthermore, in comparison against a discriminant analysis for the classification of synthetic aperture radar imagery, the results showed that training pattern replication could be used to weight class allocation with an effect similar to that of incorporating a prioriprobabilities of class membership into the discriminant analysis, and resulted in a significant 20.88%, increase in classification accuracy. This increase in classification accuracy was obtained without any new information, but was the result of making fuller use of what was available.     

基于可逆网络的轻量化图像隐写方案

目前,基于深度学习的隐写模型的隐写容量有所提高,但由于网络结构复杂,需要大量的时间来训练。针对此问题进行研究,提出轻量化的可逆神经网络结构,并以此设计了高效图像隐写方案,采用基于密集连接的可逆神经网络实现图像的隐藏与恢复,在减少可逆块的数量的同时,增加每个可逆块中可逆函数f（·）、r（·）和y（·）的卷积块数量来保证图像质量。能够显著降低计算和存储开销,使得模型在计算资源有限的设备上更加高效运行,使模型开发和迭代的过程更加高效,有效地节省了宝贵的计算资源。载体图像与秘密图像通过正向隐藏可逆变换生成含密图像,含密图像与随机变量通过反向恢复可逆变换得到恢复图像。实验结果表明,与HiNet算法相比,轻量级网络结构能够取得良好的图像质量和安全性,同时将训练时间缩短了46%,隐写时间缩短了28%。     

The effects of data filtering on neural network learning

This paper describes various experiments involving the pre-processing of training data in order to improve the performance of a neural network classifier. To minimise the effect of noisy training samples we applied k-nearest neighbour filtering to the training data. This effectively detects and eliminates outliers, which would otherwise degrade the learning process of the neural network. For a range of neural network complexities, and for two classification tasks, the filtering operation increased classification accuracy.     

小波系数指导的全色锐化网络

针对现有全色锐化网络无法同时兼顾空间信息与光谱信息保留的问题,提出一种基于小波系数指导的由融合网络和指导网络组成的全色锐化网络。融合网络分别提取PAN和MS图像的多级特征,并在同一级别进行特征的选择和融合,融合后的特征分别用于指导后一级别特征的提取;指导网络用于学习HRMS与已知的输入图像的小波系数之间的映射关系,并利用学习到的映射对融合网络的输出提供额外的监督。实验结果表明,该方法能够在保留MS图像光谱信息的同时恢复尽可能多的空间信息。在模拟数据集和真实数据集上的对比实验也表明,该方法融合效果优于其他传统方法和深度学习方法,具有一定的实用价值。     

基于深度神经网络的遥感图像分类算法综述

准确、高效的遥感图像分类是遥感图像解析的重要研究内容之一。近年来,随着机器学习技术的发展,深度神经网络日渐成为一种有效的遥感图像分类处理方法。分析了遥感图像分类目前存在的一些问题,并简要阐述了几种典型的深度神经网络的原理结构;然后根据遥感图像分类的研究现状和深度神经网络对遥感图像分类的研究现状,总结了深度神经网络在遥感图像分类技术应用中的发展趋势。     

基于BP神经网络的多传感器数据融合技术优化

传统的数据融合算法要求获得比较精确的对象数学模型,对于复杂的难于建立模型的场合无法适用。为解决上述问题,提出了一种基于BP神经网络算法的多传感器数据融合方法,对对象的先验要求不高,具有较强的自适应能力。仿真结果表明,采用BP神经网络对传感器数据进行融合处理大大提高了传感器的稳定性及其精度,效果良好。     

A comparison between functional networks and artificial neural networks for the prediction of fishing catches

In recent years, functional networks have emerged as an extension of artificial neural networks (ANNs). In this article, we apply both network techniques to predict the catches of the Prionace Glauca (a class of shark) and the Katsowonus Pelamis (a variety of tuna, more commonly known as the Skipjack). We have developed an application that will help reduce the search time for good fishing zones and thereby increase the fleets competitivity. Our results show that, thanks to their superior learning and generalisation capacities, functional networks are more efficient than ANNs. Our data proceeds from remote sensors. Their spectral signatures allow us to calculate products that are useful for ecological modelling. After an initial phase of digital image processing, we created a database that provides all the necessary patterns to train both network types.     

Hybrid deep neural network model for human action recognition

In this paper, we propose a hybrid deep neural network model for recognizing human actions in videos. A hybrid deep neural network model is designed by the fusion of homogeneous convolutional neural network (CNN) classifiers. The ensemble of classifiers is built by diversifying the input features and varying the initialization of the weights of the neural network. The convolutional neural network classifiers are trained to output a value of one, for the predicted class and a zero, for all the other classes. The outputs of the trained classifiers are considered as confidence value for prediction so that the predicted class will have a confidence value of approximately 1 and the rest of the classes will have a confidence value of approximately 0. The fusion function is computed as the maximum value of the outputs across all classifiers, to pick the correct class label during fusion. The effectiveness of the proposed approach is demonstrated on UCF50 dataset resulting in a high recognition accuracy of 99.68%.     

粗神经网络及其在图像融合中的应用研究

粗集理论能够优选数据,但容错性与推广能力比较弱;而神经网络具有较强的自组织、容错以及推理能力,却不能优选数据。把这两种理论结合起来,使之发挥各自优势,然后把它们用于图像融合之中,并提出了一种基于粗神经网络的图像融合方法,该方法使用遗传算法作为神经网络的训练算法。通过仿真实验表明,在对融合来自同一景物的多幅带噪声图像的应用中,该方法取得了很好的效果。     

基于特征融合卷积神经网络的车型精细识别

为解决现有车型精细识别方法中存在识别精度低、模型参数规模大等问题,提出一种基于特征融合卷积神经网络的车型精细识别方法。设计两个独立网络(UpNet、DownNet)分别用于提取车辆正面图像的上部和下部特征,在融合网络(FusionNet)中进行特征融合,实现车型的精细识别。相较于现有的车型精细识别方法,该方法在提高识别精度的同时,有效压缩了模型参数规模。在基准数据集CompCars下进行大量实验的结果表明,该方法的识别精度可达98.94%,模型参数大小仅为4.9MB。     

递归式多阶特征融合图像超分辨率算法

目的 近年来,卷积神经网络在解决图像超分辨率的问题上取得了巨大成功,不同结构的网络模型相继被提出。通过学习,这些网络模型对输入图像的特征进行抽象、组合,进而建立了从低分辨率的输入图像到高分辨率的目标图像的有效非线性映射。在该过程中,无论是图像的低阶像素级特征,还是高阶各层抽象特征,都对像素间相关性的挖掘起了重要作用,影响着目标高分辨图像的性能。而目前典型的超分辨率网络模型,如SRCNN（super-resolution convolutional neural network）、VDSR（very deep convolutional networks for super-resolution）、LapSRN（Laplacian pyramid super-resolution networks）等,都未充分利用这些多层次的特征。方法 提出一种充分融合网络多阶特征的图像超分辨率算法：该模型基于递归神经网络,由相同的单元串联构成,单元间参数共享;在每个单元内部,从低阶到高阶的逐级特征被级联、融合,以获得更丰富的信息来强化网络的学习能力;在训练中,采用基于残差的策略,单元内使用局部残差学习,整体网络使用全局残差学习,以加快训练速度。结果 所提出的网络模型在通用4个测试集上,针对分辨率放大2倍、3倍、4倍的情况,与深层超分辨率网络VDSR相比,平均分别能够获得0.24 dB、0.23 dB、0.19 dB的增益。结论 实验结果表明,所提出的递归式多阶特征融合图像超分辨率算法,有效提升了性能,特别是在细节非常丰富的Urban100数据集上,该算法对细节的处理效果尤为明显,图像的客观质量与主观质量都得到显著改善。     

基于神经网络与对比度的多聚焦图像融合技术

提出了一种基于小波对比度和神经网络的多聚焦图像融合算法。首先对各源图像进行小波变换,根据变换后系数计算出图像的小波对比度,选取源图像部分区域小波对比度作为前馈神经网络的训练样本,调整神经网络权重;然后用训练好的神经网络组合融合图像的小波系数,对组合后的系数进行一致性校验;最后对该系数进行小波逆变换,得到融合图像。实验结果表明,该算法能够较好地解决多聚焦图像融合问题,生成的融合图像效果优于传统图像融合方法。     

Artificial neural network visual model for image quality enhancement

An artificial neural network visual model is developed, which extracts multi-scale edge features from the decompressed image and uses these visual features as input to estimate and compensate for the coding distortions. This provides a generic postprocessing technique that can be applied to all the main coding methods. Experimental results involving postprocessing of the JPEG and quadtree coding systems show that the proposed artificial neural network visual model significantly enhances the quality of reconstructed images, both in terms of the objective peak signal-to-noise ratio and subjective visual assessment.     

基于BP神经网络的传感器交叉敏感性抑制

多传感器系统通常都存在交叉敏感性问题,传感器静态性不仅受某一个环境参量的影响,有时甚至受多个非目标参量的影响,致使测量精度低。提出一种基于BP神经网络的传感器的交叉敏感性抑制的方法,并应用到实际的检测系统中。通过对应用神经网络前后的数据进行比较分析,表明该方法大大降低了交叉干扰,提高了测量精度,达到了预期效果。     

遥感图象分类的统计融合方法及性能分析

基于数据层的统计数据融合方法,以提高遥感图象的分类性能为目的,实现了一种新的可调参数的图象分类方法。用这种方法对ＴＭ图象和ＳＡＲ图象进行了一系列的实验,并对实验的结果进行了分析,从而得出关于数据层统计信息融合方法的有益的结论。     

Blind quality assessment for screen content images via convolutional neural network

With the wide propagation of cloud and mobile computing, screen content images (SCIs) have become more indispensable in our daily lives. Compared to natural scene images (NSIs), SCIs possess many particular characteristics, like mixed contents, extremely sharp edges, and text graphics. Consequently, more challenges occur in the feature extraction, which is used to reflect the distortion, during the quality assessment of SCIs. Recently, some convolutional neural network (CNN) models have been designed by automatically learning feature to evaluate the quality. In this paper, we develop a novel blind quality assessment method for SCIs via the CNN. First, compared with existing CNN-based methods, the proposed method avoids the disadvantage of training with image patches, and it is the pioneering attempt that takes the entire image as inputs. Second, instead of the image gray value, the original image is decomposed into two portions, i.e., the predicted and unpredicted portions, according to the internal generative mechanism (IGM) theory as the input of CNN. Through the CNN, all features of the image are learned automatically from beginning to end, and the network finally outputs the predicted score. Since existing SCI database is too small, to fully train the network, we collected 30000 SCIs and employed a high-accuracy full-reference quality assessment metric of SCI to compute scores as the training labels. Experimental results on SIQAD database demonstrate that the proposed method is comparable to reference-based SCI quality assessment metrics and is superior to the state-of-the-art NSI quality assessment metrics.     

条件生成对抗遥感图像时空融合

目的 卫星遥感技术在硬件方面的局限导致获取的遥感图像在时间与空间分辨率之间存在矛盾,而时空融合提供了一种高效、低成本的方式来融合具有时空互补性的两类遥感图像数据（典型代表是Landsat和MODIS （moderate-resolution imaging spectroradiometer）图像）,生成同时具有高时空分辨率的融合数据,解决该问题。方法 提出了一种基于条件生成对抗网络的时空融合方法,可高效处理实际应用中的大量遥感数据。与现有的学习模型相比,该模型具有以下优点：1）通过学习一个非线性映射关系来显式地关联MODIS图像和Landsat图像;2）自动学习有效的图像特征;3）将特征提取、非线性映射和图像重建统一到一个框架下进行优化。在训练阶段,使用条件生成对抗网络建立降采样Landsat和MODIS图像之间的非线性映射,然后在原始Landsat和降采样Landsat之间训练多尺度超分条件生成对抗网络。预测过程包含两层：每层均包括基于条件生成对抗网络的预测和融合模型。分别实现从MODIS到降采样Landsat数据之间的非线性映射以及降采样Landsat与原始Landsat之间的超分辨率首建。结果 在基准数据集CIA （coleam bally irrigation area）和LGC （lower Gwydir catchment）上的结果表明,条件生成对抗网络的方法在4种评测指标上均达到领先结果,例如在CIA数据集上,RMSE （root mean squared error）、SAM （spectral angle mapper）、SSIM （structural similarity）和ERGAS （erreur relative global adimensionnelle desynthese）分别平均提高了0.001、0.15、0.008和0.065;在LGC数据集上分别平均提高了0.001 2、0.7、0.018和0.008 9。明显优于现有基于稀疏表示的方法与基于卷积神经网络的方法。结论 本文提出的条件生成对抗融合模型,能够充分学习Landsat和MODIS图像之间复杂的非线性映射,产生更加准确的融合结果。     

Pan-sharpening:a fast variational fusion approach

A fast variational fusion model based on partial differential equations (PDEs) is presented for pansharpening. The functional framework consists of several energy terms. The gradient energy term is created by calculating the gradient vector field of the panchromatic image and the geometry of the pan is injected into the multi-spectral bands. The radiometric reduction energy term and the channel correlation energy term are defined to decrease the radiometric distortion and preserve the correlation of multi-spectral channels while enforcing the inter-bands coherence. Inspired by the shock-filtering model, an inverse diffusion term for image enhancement is put to PDEs which are deduced by minimizing the energy functional. In comparison with the state-of-the-art fusion approaches based on `a trous wavelet and non-sampled contourlet, our model can obtain the fused image with a high spatial or spectral quality by adjusting the weight coefficients of the energy terms. It can also achieve a rather good trade-off between the spatial resolution improvement and the spectral quality preserving. Our model’s computational complexity for one time step is only O(N).