基于层次集成学习的高光谱图像分类

为提高高光谱图像（HSI）分类精度,基于集成学习方法提出高光谱图像分类的层次集成学习新框架。采用两种集成学习策略：外部集成及内部集成。在外部集成阶段,构造多种高光谱图像的光谱和空间特征,使外部集成呈高度多样性,有利于提高分类精度;内部集成阶段,针对关联多特征集中的个体,Adaboost算法实现个体分类性能的提高。两组高光谱数据的实验结果表明,与原始的Adaboost和单分类器相比较,该方法在整体精度方面有更好的性能。     

优化端元提取方法的高光谱混合像元分解

高光谱图像的混合像元分解将原始图像分解为多种纯净地物及相应的丰度,端元提取是混合像元分解的关键技术. 针对传统算法计算速度慢、搜索范围较大的特点,基于改进的ICA （independent component analysis）算法以及优化的候选端元判断方法,提出了一种优化的混合像元分解方法. 首先使用改进的算法优化端元提取方法;然后利用相邻像素的光谱特征和空间特征信息,结合并行算法对候选端元进行优化;最后利用真实的高光谱数据对该方法的性能进行了验证. 验证结果表明:该方法能有效提高端元提取精度,降低复杂度,与经典的端元提取算法N-FINDER相比,准确度提高了3.55%,解混后得到的地物分类精度有了明显改善（总体分类精度提高了2.88%）.      

一种基于稀疏表示的交通图像去噪算法

将图像稀疏表示方法引入到交通图像处理中,实现了一种基于K-SVD的正交匹配追踪的交通图像去噪算法.该算法通过奇异值分解,DCT字典进行自适应更新,形成更能表示图像结构的超完备字典.实验结果表明,相对于传统图像增强方法(中值滤波、均值滤波、基于小波滤波)和基于DCT冗余字典的稀疏表示图像增强方法,该算法能更有效地去除交通图像噪声,得到更高的峰值信噪比.     

分块组合搜索的结构稀疏人脸识别模型

介绍一种新的稀疏表示人脸识别模型,在经典的稀疏表示分类模型基础上,利用数据字典的结构信息,考虑算法实现的可行性,提出了一种分块组合搜索的稀疏表示人脸识别模型,主要思想是将数据字典按类别自然分块,然后在数据块内进行组合搜索,再联合不同类别的组块,以寻找表示能力最强的组块。为验证模型的性能,使用结构贪婪算法实现分块组合搜索方法和其他的结构稀疏方法,并进行比较,实验显示分块组合搜索的人脸识别率高于其他结构稀疏方法,且此方法性能稳定,不受数据字典排列的影响。     

基于局部自相似的字典学习图像去噪方法

利用图像具有自相似的特点,结合稀疏表示理论,提出了一种新的图像去噪方法.该方法汇集相似的图像块,构造局部字典,能够更好的匹配图像的纹理和复杂结构.实验结果表明,该方法在抑制图像噪声的同时,能很好的保持图像的纹理和细节信息.     

基于压缩感知和字典学习的背景差分法

针对当使用背景差分法时,背景存在突变和渐变、图像数据的冗余和伪前景对目标检测的干扰等问题,提出一种基于稀疏表示和字典学习的背景差分法。该方法首先训练视频流得到其数据字典,并根据数据字典学习与稀疏表示理论建立背景模型,可以有效减少数据的冗余。然后根据目标及其邻域的密集度进行目标分割,以排除前景的干扰。最后再根据数据字典的更新算法,有效解决了背景的突变和渐变问题。实验结果表明,该方法具有可行性。     

自适应冗余字典学习的高光谱混合像元解混

针对线性稀疏解混模型无法准确识别真实端元造成丰度估计误差较大的问题,本文提出一种基于自适应冗余字典的高光谱混合像元解混算法.该算法根据地物在空间上的连续性,以及高光谱数据中信号成分与光谱库中物质光谱的强相关性,首先保留每个像元在光谱库上投影系数大于设定阈值所对应的光谱,将其作为与每个像元信号成分最匹配的光谱集合;然后合并该集合以构建高光谱数据的自适应冗余字典;最后利用ADMM算法求解高光谱数据在该字典上的丰度矩阵.仿真和实际高光谱数据实验结果表明,本文所提出的算法可减小丰度估计误差,在信噪比为15~35 dB时,其丰度估计准确性高于性能较优的SUnSAL算法约1~2 dB.      

基于多标记像素匹配的高分辨率遥感图像道路提取

针对高分辨率遥感图像道路提取方法存在提取不完整和误提取问题，提出一种基于多标记像素匹配的高分辨率遥感图像道路提取方法. 将待提取图像由RGB颜色空间转换到 Lab颜色空间，选取与照明强度弱相关的色相特征作为初始匹配项. 以矩形框的方式标记不同类型的道路，利用t 检验法剔除其匹配项的异常值，从而确定阈值来匹配道路像素点，利用局部纹理算子对匹配结果进行筛选. 利用道路区域的形态特征对匹配结果进行优化处理. 为验证所提方法的可行性和优越性，对不同传感器获取的高分辨遥感图像进行测试，与现有道路提取方法进行对比. 定性和定量精度评价结果表明，所提方法对不同类型道路的提取具有较高的精度.     

基于混沌变异粒子群优化算法的图像稀疏分解

提出了基于改进的粒子群优化（PSO）算法的匹配追踪算法,用于快速图像稀疏分解．改进的PSO算法利用尺度收缩混沌变异的精细局部搜索性能,使稀疏分解的匹配追踪算法具有良好的全局寻优能力,提高了稀疏分解在冗余字典中原子匹配的速度和准确度．用二维墨西哥草帽函数作为冗余字典的生成函数,以增强对图像边缘和轮廓的表达能力．仿真结果表明,用提出的算法实现图像稀疏分解比用遗传算法和PSO更快更有效,重建图像的视觉效果好．     

基于Hopfield神经网络的遥感图像超分辨率识别算法

提出基于Hopfield神经网络的遥感图像超分辨率目标识别算法，它是利用模糊分类技术进行模糊分类，然后用分类结果约束Hopfield神经网络的方法．通过实验，可知Hopfield神经网络在学习样本少时．也能够输出分辨率相对较高的地物目标信息．因此．基于Hopfield神经网络的遥感图像处理方法，能够提高遥感图像的目标分辨率．使其目标特征信息更清晰．     

Weak correlation dictionary construction method for sparse coding

For sparse coding, the weaker the correlation of dictionary atoms is, the better the representation capacity of dictionary will be. A weak correlation dictionary construction method for sparse coding has been proposed in this paper. Firstly, a new dictionary atom initialization is proposed in which data samples with weak correlation are selected as the initial dictionary atoms in order to effectively reduce the correlation among them. Then, in the process of dictionary learning, the correlation between atoms has been measured by correlation coefficient, and strong correlation atoms have been eliminated and replaced by weak correlation atoms in order to improve the representation capacity of the dictionary. An image classification scheme has been achieved by applying the weak correlation dictionary construction method proposed in this paper. Experimental results show that, the proposed method averagely improves image classification accuracy by more than 2%, compared to sparse coding spatial pyramid matching (ScSPM) and other existing methods for image classification on the datasets of Caltech-101, Scene-15, etc.     

Range profile target recognition using sparse representation based on feature space

A novel method is presented to improve the recognition rate of warhead in this paper. Firstly, a tool for electromagnetic calculation, like CST Microwave Studio, is used to simulate the frequency response of the electromagnetic scattering. Secondly, the echo and further the range profile are acquired from the frequency response by further processing. Thirdly, a set of discriminative features is extracted from the range profiles of the target. Fourthly, these features are used to construct a dictionary for the sparse representation classifier. Finally, the sample of the target can be classified by solving the sparsest coefficients. Since the reconstruction result is determined by a linear combination of the training samples, this method has a good robustness for the variable features. By formulating the problem within a feature-based sparse representation framework, the presented method combines the discriminative features of each sample during the sparse recovery process rather than in a postprocessing manner. Moreover, based on the feature representation space rather than a single feature or image pixel, the constructed dictionary exhibits both strong expressive and discriminative powers that can enhance the classification performance of the test sample. A series of test results based on the simulated data demonstrates the effectiveness of our method.     

黑白数字图像的有穷状态自动机表示方法

自动机理论是理论计算机科学的基础理论之一，在很多领域自动机有着广泛的应用，在将黑白图像进行像素地址编码的基础上使用语言来描述数字图像，从而得到用自动机来描述数字图像的方法，任意有穷分辨率黑白图像均可以用有穷状态自动机来表示，多分辨率图像能够用有穷状态自动机来描述当且仅当该图像中不同形状的子图像的个数为有限个．     

An information-based color feature representation and its application in detecting adult images

For many image classification tasks, color histogram is usually employed as an important “signature” to describe the color distribution of the image and infer the image content. However, most traditional color histograms cannot achieve satisfactory results in many image classification systems. In order to improve the accuracy and reduce the computational complexity of the classification task, an information-based color feature representation is proposed in this paper. The mutual information between the feature and the class label is adopted to evaluate the discriminative power of the feature. A novel quantization scheme is presented, which removes the redundant color components and combines the adjacent components together to generate a new feature to maximize the discriminative ability. An iterative algorithm is performed to derive the color space quantization and color feature generation. In order to illustrate the effectiveness of the proposed color representation, a specific image classification task, i.e., differentiating the adult images from benign ones, is employed. Experimental results show that our color feature achieves better classification performance and better efficiency compared with the traditional color histogram.     

基于高光谱成像技术的绝缘子污秽度预测

高光谱成像技术能对绝缘子进行非接触式成像,且具有多波段、图谱合一等特点. 为此,本文提出一种基于高光谱成像技术的绝缘子污秽度预测方法. 首先,利用高光谱成像仪对绝缘子进行成像,得到400～1 000 nm波段范围内的高光谱图像数据,并进行黑白校正;然后,获取感兴趣区域（region of interest,ROI）的反射率光谱曲线,进行Savitzky-Golay平滑、对数或一阶导数变换的预处理. 最后,联合部分的真实样本标签数据分别建立基于支持向量机的绝缘子污秽度预测（support vector machines-insulator contamination degree prediction,SVM-ICDP）和基于偏最小二乘回归的绝缘子污秽度预测（partial least squares regression-insulator contamination degree prediction,PLSR-ICDP）模型. 从实验结果中可知,当预处理方法采用一阶导数变换时,所建立的绝缘子污秽度预测模型效果最佳,即SVM-ICDP模型准确率达到91.84%;PLSR-ICDP模型的均方根误差（root mean square error,RMSE）为0.024 1.      

A generalized two-level Bregman method with dictionary updating for non-convex magnetic resonance imaging reconstruction

In recent years, it has shown that a generalized thresholding algorithm is useful for inverse problems with sparsity constraints. The generalized thresholding minimizes the non-convex p-norm based function with p 1, and it penalizes small coefficients over a wider range meanwhile applies less bias to the larger coefficients.In this work, on the basis of two-level Bregman method with dictionary updating(TBMDU), we use the modified thresholding to minimize the non-convex function and propose the generalized TBMDU(GTBMDU) algorithm.The experimental results on magnetic resonance(MR) image simulations and real MR data, under a variety of sampling trajectories and acceleration factors, consistently demonstrate that the proposed algorithm can efficiently reconstruct the MR images and present advantages over the previous soft thresholding approaches.