改进的基于深度学习的遥感图像分类算法

针对传统的基于深度学习的遥感图像分类算法未能有效融合多种深度学习特征，且分类器性能欠佳的问题，提出一种改进的基于深度学习的高分辨率遥感图像分类算法。首先，设计并搭建一个七层卷积神经网络；其次，将高分辨率遥感图像样本输入到该网络中进行网络训练，得到最后两个全连接层输出作为遥感图像两种不同的高层特征；再次，针对该网络第五层池化层输出，采用主成分分析（PCA）进行降维，作为遥感图像的第三种高层特征；然后，将上述三种高层特征通过串联的形式进行融合，得到一种有效的基于深度学习的遥感图像特征；最后，设计了一种基于逻辑回归的遥感图像分类器，可以对遥感图像进行有效分类。与传统基于深度学习的遥感图像分类算法相比，所提算法分类准确率有较高提升。实验结果表明，该算法在分类准确率、误分类率和Kappa系数上表现优异，能实现良好的分类效果。     

基于蚁群优化分类规则挖掘的遥感图像分类研究

蚁群优化算法作为群智能理论的主要算法之一,已经成功应用在众多研究领域的优化问题上,但是在遥感数据处理领域还是一个新的研究课题。蚁群优化具有自组织、合作、通信等智能化优点,对数据无需统计分布参数的先验知识,因此在遥感数据处理领域具有很大的潜在优势。介绍了将蚁群优化分类规则挖掘算法应用到遥感图像分类研究领域的理论与算法流程。并采用北京地区的CBERS遥感数据作为实验数据,通过蚁群优化算法构造分类规则,对选择的遥感数据进行了分类实验,并和最大似然分类方法进行对比,实验结果表明,蚁群优化分类规则挖掘算法为遥感图像的分类提供了一种新方法。     

基于AdaBoost的组合分类器在遥感影像分类中的应用*

运用组合分类器的经典算法AdaBoost将多个弱分类器-神经网络分类器组合输出,并引入混合判别多分类器综合规则,有效提高疑难类别的分类精度,进而提高分类的总精度.最后以天津地区ASTER影像为例,介绍了基于AdaBoost的组合分类算法,并在此基础上实现了天津地区的土地利用分类.分类结果表明,组合分类器能有效提高单个分类器的分类精度,分类总精度由81.13%提高到93.32%.实验表明基于AdaBoost的组合分类是遥感图像分类的一种新的有效方法.     

Cost-sensitive boosting for classification of imbalanced data

Classification of data with imbalanced class distribution has posed a significant drawback of the performance attainable by most standard classifier learning algorithms, which assume a relatively balanced class distribution and equal misclassification costs. The significant difficulty and frequent occurrence of the class imbalance problem indicate the need for extra research efforts. The objective of this paper is to investigate meta-techniques applicable to most classifier learning algorithms, with the aim to advance the classification of imbalanced data. The AdaBoost algorithm is reported as a successful meta-technique for improving classification accuracy. The insight gained from a comprehensive analysis of the AdaBoost algorithm in terms of its advantages and shortcomings in tacking the class imbalance problem leads to the exploration of three cost-sensitive boosting algorithms, which are developed by introducing cost items into the learning framework of AdaBoost. Further analysis shows that one of the proposed algorithms tallies with the stagewise additive modelling in statistics to minimize the cost exponential loss. These boosting algorithms are also studied with respect to their weighting strategies towards different types of samples, and their effectiveness in identifying rare cases through experiments on several real world medical data sets, where the class imbalance problem prevails.     

基于剪枝网络的知识蒸馏对遥感卫星图像分类方法

针对目前遥感图像在应用卷积神经网络分类时需要大量计算,并占用大量内存的问题,提出了一种基于剪枝网络的知识蒸馏对遥感图像分类方法.以模型剪枝理论为基础,在网络结构中引入注意力机制,加强对重要特征的提取之后,并对网络进行模型剪枝,然后引入知识蒸馏技术对模型进行迁移学习,补偿模型剪枝之后分类精度的损失.为了证明方法的先进性与可靠性,利用在NWPU-RESISC45遥感卫星数据集上,与同类算法进行对比实验.实验结果表明,所提方法不仅在分类精度有更好的表现,并且在模型大小上更具有优势.     

A comparison of multiple classifier combinations using different voting-weights for remote sensing image classification

Remote sensing image classification is a common application of remote sensing images. In order to improve the performance of Remote sensing image classification, multiple classifier combinations are used to classify the Landsat-8 Operational Land Imager (Landsat-8 OLI) images. Some techniques and classifier combination algorithms are investigated. The classifier ensemble consisting of five member classifiers is constructed. The results of every member classifier are evaluated. The voting strategy is experimented to combine the classification results of the member classifier. The results show that all the classifiers have different performances and the multiple classifier combination provides better performance than a single classifier, and achieves higher overall accuracy of classification. The experiment shows that the multiple classifier combination using producer’s accuracy as voting-weight (MCC_mod2 and MCC_mod3) present higher classification accuracy than the algorithm using overall accuracy as voting-weight (MCC_mod1).And the multiple classifier combinations using different voting-weights affected the classification result in different land-cover types. The multiple classifier combination algorithm presented in this article using voting-weight based on the accuracy of multiple classifier may have stability problems, which need to be addressed in future studies.     

一种基于模糊关联分类的遥感图像分类方法

遥感图像分类是遥感领域的研究热点之一.提出了一种基于自适应区间划分的模糊关联遥感图像分类方法(fuzzy associative remote sensing classification,FARSC).算法根据遥感图像分类的特点,利用模糊C均值聚类算法自适应地建立连续型属性模糊区间,使用新的剪枝策略对项集进行筛选从而避免生成无用规则,采用一种新的规则重要性度量方法对多模糊分类规则进行融合,从而有效地提高分类效率和精确度.在UCI数据和遥感图像上所作实验结果表明,算法具有较高的分类精度以及对样本数量变化的不敏感性,对于解决遥感图像分类问题,FARSC算法具有较高的实用性,是一种有效的遥感图像分类方法.     

遥感影像投影寻踪回归分类模型

将投影寻踪回归分析技术引入遥感影像分类中,详尽叙述遥感影像投影寻踪回归分类模型的建立和实现过程。将广州地区的TM影像用于分类实验,并用混合蛙跳算法来优化投影寻踪回归分类模型中的参数矩阵,取得了较为理想的分类效果。此外,还进一步分析了投影中心的设定、调整以及优化算法和岭函数个数对投影寻踪回归模型分类精度的影响。实验结果表明,该模型易于优化实现,稳定性强,模型中岭函数的个数对投影寻踪回归模型的分类精度没有显著影响。     

Boosting Ensemble Learning for Hyperspectral Image Classification Using Tangent Collaborative Representation

Recently, Collaborative Representation Classification (CRC) has attracted much attention in hyperspectral image analysis. Due to uses the tangent plane to estimate the local manifold of the test sample. Tangent Collaborative Representation Classification (TCRC) achieve better performance. Furthermore, in order to improve the classification accuracy and reliability of hyperspectral remote sensing images, a novel Boosting-based Tangent Collaborative Representation ensemble method (Boost TCRC) for hyperspectral image classification is proposed. In this algorithm, Boost TCRC algorithm choose TCRC as base classifier and adjust the weight of the training samples adaptively by using the principle of Boosting. Increasing the weight of the misclassified samples so that the classifier concentrates on the training samples that are difficult to classify. Then assigns the weights according to the classification performance of the base classifier based on the residual domain fusion. Finally, the principle of minimum reconstruction error is adopted to classify the test sample. The performance of the proposed algorithm was comprehensively evaluated by hyperspectral remote sensing image data such as HyMap (Hyperspectral Mapper) and AVIRIS (Airbone Visible Infrared Imaging Spectrometer). The Boosting method can effectively improve the classification effect of the TCRC algorithm. For HyMap data, the overall classification accuracy and kappa coefficient of Boost TCRC algorithm are 93.73% and 0.920 8 respectively. Two precision values are higher than TCRC algorithm by 2.82% and 0.032 3, and are higher than the AdaBoost ELM algorithm by 1.81% and 0.022 5. For AVIRIS data, the overall classification accuracy and kappa coefficient of Boost TCRC algorithm are 84.11% and 0.8120 respectively. Two precision values are higher than TCRC algorithm by 3.97% and 0.049 3, and are higher than AdaBoost ELM algorithm by 12.02% and 0.143 6.     

利用概率主题模型的遥感影像半监督分类

土地覆盖是自然环境与人类活动相互作用的中心,而土地覆盖信息主要是通过遥感影像分类来获取,因此影像分类是遥感影像分析的最基本问题之一。在参考基于概率主题模型的高分辨率遥感影像聚类分析的基础上,通过半监督学习最典型的生成模型方法引出了基于概率主题模型的半监督分类（SS-LDA）算法。借鉴SS-LDA模型在文本识别应用的流程,构建了基于SS-LDA算法的高分辨率遥感影像分类的基本流程。通过实验证明,相对于传统的非监督分类与监督分类算法,SS-LDA算法能够获取较高精度的影像分类结果。     

一种用于不平衡数据分类的改进AdaBoost算法

真实世界中存在大量的类别不平衡分类问题,传统的机器学习算法如AdaBoost算法,关注的是分类器的整体性能,而没有给予小类更多的关注。因此针对类别不平衡学习算法的研究是机器学习的一个重要方向。AsymBoost作为AdaBoost的一种改进算法,用于类别不平衡学习时,牺牲大类样本的识别精度来提高小类样本的分类性能。AsymBoost算法依然可能遭遇样本权重过大造成的过适应问题。据此提出了一种新型的AdaBoost改进算法。该方法通过对大类中分类困难样本的权重和标签进行处理,使分类器能够同时获得较好的查准率和查全率。实验结果表明,该方法可以有效提高在不平衡数据集上的分类性能。     

基于Boosting的高光谱遥感切空间协同表示集成学习方法

近年来,协同表示分类(Collaborative Representation Classification,CRC)算法成为高光谱遥感影像分类的研究热点,尤其是切空间协同表示分类(Tangent Space Collaborative Representation,TCRC)利用切平面估计测试样本的局部流形,其分类精度得到了显著提高。为进一步提升高光谱遥感影像分类的准确性和可靠性,提出了基于Boosting的高光谱遥感影像切空间协同表示分类算法(Boosting-based Tangent Space Collaborative Representation Classification,Boost TCRC)。Boost TCRC算法采用TCRC算法作为基分类器,通过Boosting原理自适应地调整训练样本的权重,增大错分样本的权重从而使得分类器专注于较难分类的训练样本,然后在基于残差域融合时根据基分类器的分类表现赋予其权重,最终采用最小重构误差的原则对测试样本进行分类。实验采用HyMap(Hyperspectral Mapper)和AVIRIS(Airbone Visible Infrared Imaging Spectrometer)等高光谱遥感影像数据对所提出算法的性能进行了综合评价,结果表明：基于Boosting的集成方式可有效提升TCRC算法的分类效果。针对HyMap数据,Boost TCRC算法总体分类精度和Kappa系数分别为93.73%和0.920 8,两种精度指标分别高于TCRC算法2.82%和0.032 3,同时分别高于AdaBoost ELM算法1.81%和0.022 5。对于AVIRIS数据,Boost TCRC算法总体分类精度和kappa系数为84.11%和0.812 0,两种精度指标分别高于TCRC算法3.97%和0.049 3,同时分别高于AdaBoost ELM算法12.02%和0.143 6。     

基于边缘特征和AdaBoost分类的遥感图像云检测

传统遥感图像云检测方法在处理山地、雪地、暗云等场景时极易发生错判,准确度较低。通过对遥感图像中云与地物的不同特点进行分析,提出一种新的遥感图像边缘特征描述方法,结合图像的边缘特征和灰度特征使用AdaBoost分类器进行云图分类,并利用图像的空间相关性对分类结果进行修正。经10万余幅图像测试结果表明:该算法与传统算法相比准确度极大提高,正确率达到96%以上,且运算速度快,满足实时性要求。     

Boosting a fast neural network for supervised land cover classification

It is demonstrated that the use of an ensemble of neural networks for routine land cover classification of multispectral satellite data can lead to a significant improvement in classification accuracy. Specifically, the AdaBoost.M1 algorithm is applied to a sequence of three-layer, feed-forward neural networks. In order to overcome the drawback of long training time for each network in the ensemble, the networks are trained with an efficient Kalman filter algorithm. On the basis of statistical hypothesis tests, classification performance on multispectral imagery is compared with that of maximum likelihood and support vector machine classifiers. Good generalization accuracies are obtained with computation times of the order of 1 h or less. The algorithms involved are described in detail and a software implementation in the ENVI/IDL image analysis environment is provided.     

基于模糊双支持向量机的遥感图像分类研究

遥感图像的分类是研究土地利用变化的基础。传统的遥感图像分类方法存在运算速度慢、精度比较低和难以收敛等问题。提出了一种基于模糊双支持向量机的多类分类方法,将模糊技术引入到双支持向量机中,赋予不同样本以不同的模糊隶属度,然后将模糊双支持向量机推广到多类分类中,最后将新方法应用到遥感图像分类中。实验表明,新方法比传统的支持向量机多类分类方法有较高的分类精度,并且有较强的抗噪声能力,在运行时间上也是可行的。模糊双支持向量机是一种有效的遥感图像分类方法。     

Analysis of classification accuracy for pre-filtered multichannel remote sensing data

Noise is one of the main factors degrading the quality of original multichannel remote sensing data and its presence influences classification efficiency, object detection, etc. Thus, pre-filtering is often used to remove noise and improve the solving of final tasks of multichannel remote sensing. Recent studies indicate that a classical model of additive noise is not adequate enough for images formed by modern multichannel sensors operating in visible and infrared bands. However, this fact is often ignored by researchers designing noise removal methods and algorithms. Because of this, we focus on the classification of multichannel remote sensing images in the case of signal-dependent noise present in component images. Three approaches to filtering of multichannel images for the considered noise model are analysed, all based on discrete cosine transform in blocks. The study is carried out not only in terms of conventional efficiency metrics used in filtering (MSE) but also in terms of multichannel data classification accuracy (probability of correct classification, confusion matrix). The proposed classification system combines the pre-processing stage where a DCT-based filter processes the blocks of the multichannel remote sensing image and the classification stage. Two modern classifiers are employed, radial basis function neural network and support vector machines. Simulations are carried out for three-channel image of Landsat TM sensor. Different cases of learning are considered: using noise-free samples of the test multichannel image, the noisy multichannel image and the pre-filtered one. It is shown that the use of the pre-filtered image for training produces better classification in comparison to the case of learning for the noisy image. It is demonstrated that the best results for both groups of quantitative criteria are provided if a proposed 3D discrete cosine transform filter equipped by variance stabilizing transform is applied. The classification results obtained for data pre-filtered in different ways are in agreement for both considered classifiers. Comparison of classifier performance is carried out as well. The radial basis neural network classifier is less sensitive to noise in original images, but after pre-filtering the performance of both classifiers is approximately the same.     

基于自编码神经网络的航空物探遥感数据分类方法研究

航空物探遥感数据的采集过程中受到电磁波辐射等外界因素的影响,导致航空物探遥感数据分类准确率较低,为此提出基于自编码神经网络的航空物探遥感数据分类方法;根据航空物探对象的基本特征,设置遥感数据的分类标准;通过辐射校正、几何纠正、噪声消除等步骤,完成航空物探遥感数据的预处理;构建自编码神经网络,利用自编码神经网络算法,从光谱、形状、纹理等方面提取遥感数据特征,通过特征匹配确定航空物探遥感数据的所属类型;通过分类性能测试实验得出结论：所提方法的全局遥感数据分类成功率和错误率的平均值分别为99.8%和0.6%,局部遥感数据分类的成功率和错误率的平均值分别为99.8%和0.3%,即所提方法在分类性能方面具有明显优势。     

基于多分类AdaBoost改进算法的TEE标准切面分类

针对超声图像样本冗余、不同标准切面因疾病导致的高度相似性、感兴趣区域定位不准确问题,提出一种结合特征袋（BOF）特征、主动学习方法和多分类AdaBoost改进算法的经食管超声心动图（TEE）标准切面分类方法。首先采用BOF方法对超声图像进行描述;然后采用主动学习方法选择对分类器最有价值的样本作为训练集;最后,在AdaBoost算法对弱分类器的迭代训练中,根据临时强分类器的分类情况调整样本更新规则,实现对多分类AdaBoost算法的改进和TEE标准切面的分类。在TEE数据集和三个UCI数据集上的实验表明,相比AdaBoost.SAMME算法、多分类支持向量机（SVM）算法、BP神经网络和AdaBoost.M2算法,所提算法在各个数据集上的G-mean指标、整体分类准确率和大多数类别分类准确率都有不同程度的提升,且比较难分的类别分类准确率提升最为显著。实验结果表明,在包含类间相似样本的数据集上,分类器的性能有显著提升。     

基于特征整合的卷积神经网络草地分类算法

为提高遥感影像草地分类的精度,分析了卷积神经网络中提取图像特征的特点,提出了一种基于特征整合深度神经网络的遥感影像特征提取算法。首先,将遥感影像数据进行PCA白化处理,降低数据之间的相关性,加快神经网络学习的速率;其次,将从卷积神经网络中提取到的浅层特征和深层特征进行双线性整合,使得整合后的新特征更加完善和优化;最后,对遥感数据进行训练,由于新特征中有效信息的增加,使得特征表达能力得到提高,达到提高草地分类准确率的目的。实验结果表明:该算法能够有效地提高草地分类的准确率,分类精度达到94.65%,相较于卷积神经网络、BP神经网络和基于SVM的分类算法分别提高了4.3%、10.39%和15.33%。     

基于半监督模糊C均值算法的遥感影像分类

遥感影像数据因其固有的不确定性与复杂性,导致传统的无监督分类算法难以对其准确建模。基于模糊集理论的模式识别方法可以有效地表达数据的模糊性,其中二型模糊集能更好地刻画类间多重不确定性,而半监督法可以利用少量先验知识来解决算法对数据的泛化性问题,因此提出一种基于半监督的自适应区间二型模糊C均值遥感影像分类方法（SS-AIT2FCM）。首先,结合半监督和进化论思想,提出一种新的模糊权重指数选取方法,以提升自适应区间二型模糊C均值聚类算法的鲁棒性与泛化性,使算法更适用于光谱混叠严重、覆盖面积大、地物丰富的遥感数据分类;然后,通过对少量标记样本的软约束监督,对区间二型模糊算法迭代过程进行优化指导,来挖掘数据的最优表达。实验选用了北京颐和园区域的SPOT5多光谱遥感影像数据和广东横琴岛区域的Landsat TM多光谱遥感影像数据,对现有流行的模糊分类算法和SS-AIT2FCM的分类结果进行了比较。结果表明,SS-AIT2FCM获得了更高的分类精度与更清晰的类别边界,且有较好数据泛化能力。