Gully erosion spatial modelling: Role of machine learning algorithms in selection of the best controlling factors and modelling process

This investigation assessed the efficacy of 10 widely used machine learning algorithms (MLA) comprising the least absolute shrinkage and selection operator (LASSO), generalized linear model (GLM), stepwise generalized linear model (SGLM), elastic net (ENET), partial least square (PLS), ridge regression, support vector machine (SVM), classification and regression trees (CART), bagged CART, and random forest (RF) for gully erosion susceptibility mapping (GESM) in Iran. The location of 462 previously existing gully erosion sites were mapped through widespread field investigations, of which 70% (323) and 30% (139) of observations were arbitrarily divided for algorithm calibration and validation. Twelve controlling factors for gully erosion, namely, soil texture, annual mean rainfall, digital elevation model (DEM), drainage density, slope, lithology, topographic wetness index (TWI), distance from rivers, aspect, distance from roads, plan curvature, and profile curvature were ranked in terms of their importance using each MLA. The MLA were compared using a training dataset for gully erosion and statistical measures such as RMSE (root mean square error), MAE (mean absolute error), and R-squared. Based on the comparisons among MLA, the RF algorithm exhibited the minimum RMSE and MAE and the maximum value of R-squared, and was therefore selected as the best model. The variable importance evaluation using the RF model revealed that distance from rivers had the highest significance in influencing the occurrence of gully erosion whereas plan curvature had the least importance. According to the GESM generated using RF, most of the study area is predicted to have a low (53.72%) or moderate (29.65%) susceptibility to gully erosion, whereas only a small area is identified to have a high (12.56%) or very high (4.07%) susceptibility. The outcome generated by RF model is validated using the ROC (Receiver Operating Characteristics) curve approach, which returned an area under the curve (AUC) of 0.985, proving the excellent forecasting ability of the model. The GESM prepared using the RF algorithm can aid decision-makers in targeting remedial actions for minimizing the damage caused by gully erosion.     

基于深度学习的赤潮发生预报方法研究

赤潮作为海洋灾害,对海洋渔业、生态、经济,以及人类生产、生活造成了严重影响。一直以来,赤潮受到研究者的广泛关注,但由于它的形成机制比较复杂,使得赤潮预报极具挑战性。针对赤潮预报的研究问题,本文收集了厦门海域赤潮发生前后的海洋监测数据,结合皮尔逊相关系数、散布矩阵、复相关系数方法,分析多环境因子与赤潮发生多要素的关联情况,重点采用基于深度学习的LSTM与CNN融合方法,挖掘环境因子的时序依赖,发现序列数据的局部特征,对赤潮发生进行预报。在厦门一号和厦门二号数据集中,本方法在预报未来12 h内的赤潮情况时,RMSE、MAE误差分别达到0.521 8、0.504 3。通过协同对比模型进一步确定赤潮发生的预报概率,在两个数据集上的最终预报准确率分别为67.58%和63.49%。本研究为赤潮的分析预报提供了探索经验,证明了将深度学习方法应用于赤潮预报的可行性。     

建筑物立面分类与DIM点云滤波

针对目前基于近景摄影测量方法构建建筑物立面模型过程中因密集影像匹配(DIM)点云噪声所引起的建筑物立面TIN网格模型畸变问题,本文借鉴机器学习中样本学习的思想,对建筑物立面进行了分类并对DIM点云提出了相应的滤波方法,以达到去除DIM点云噪声和改善其TIN网格模型畸变的目的.其中,针对平面结构立面,采取先对点云样本进行...     

深度学习GoogleNet模型支持下的中分辨率遥感影像自动分类

提出了一种基于深度学习技术的遥感分类方法,它能有效解决中分辨率影像在分类过程中出现的像元混分问题。研究选用2016年5月12日武汉市Landsat 7 ETM+遥感影像,基于GoogleNet模型中的Inception V3网络结构,借助迁移学习方法,构建出遥感分类模型,实现了对武汉市主城区4类典型地物（不透水层、植被、水体和其他用地）的自动分类提取,并将分类结果与传统最大似然分类（ML）结果进行了对比分析。研究表明：基于深度学习方法的遥感影像总体分类精度高达88.33%,Kappa系数为0.834 2,明显优于传统ML方法总体分类精度83%和Kappa系数0.755 0,而且有效抑制了地物在分类过程中出现的像元混分现象。     

基于CART集成学习的城市不透水层百分比遥感估算

利用Landsat ETM^＋遥感数据，提出了一种基于CART集成学习的ISP遥感亚像元估算方法，将Boosting重采样技术引入CART分析中，用于提高ISP估算的精度。实验结果表明，该方法的ISP估算性能优于传统的单一CART学习算法，从ETM^＋影像中估算的ISP值与真实值之间的相关系数达到0．91，平均偏差为11．16％。     

A novel hybrid approach of Bayesian Logistic Regression and its ensembles for landslide susceptibility assessment

Abstract

A novel artificial intelligence approach of Bayesian Logistic Regression (BLR) and its ensembles [Random Subspace (RS), Adaboost (AB), Multiboost (MB) and Bagging] was introduced for landslide susceptibility mapping in a part of Kamyaran city in Kurdistan Province, Iran. A spatial database was generated which includes a total of 60 landslide locations and a set of conditioning factors tested by the Information Gain Ratio technique. Performance of these models was evaluated using the area under the ROC curve (AUROC) and statistical index-based methods. Results showed that the hybrid ensemble models could significantly improve the performance of the base classifier of BLR (AUROC?=?0.930). However, RS model (AUROC?=?0.975) had the highest performance in comparison to other landslide ensemble models, followed by Bagging (AUROC?=?0.972), MB (AUROC?=?0.970) and AB (AUROC?=?0.957) models, respectively.     

基于机器学习的地震动特征提取与模拟以2021年云南漾濞6.4级地震为例

随着地震动数据数量的增长和质量的提高,将基于数据驱动的机器学习方法应用到地震动模拟中有重要意义。以2021年5月21日云南漾濞M_S6.4地震为例,利用主成分析方法从前震及余震地震动记录中提取特征母波时程,将地震动三要素作为模拟误差约束,在求解母波的线性组合系数时使用多目标优化算法寻优,最终找到帕累托最优解作为模拟目标台站记录时的组合系数,得到模拟地震动时程。结果表明:主成分析法在对实际地震动记录进行特征提取后,得到的特征母波时程可以在一定程度上保留原始数据的主要信息;考虑幅值、频谱和持时这三要素的角度去控制模拟误差,可以使得模拟的地震动时程更加接近真实记录。提出的基于特征提取的地震动模拟方法可以为基于小震数据合成大震地震动提供参考。     

基于深度学习的循环水养殖鳗鲡(Anguilla)计数研究

鳗鲡(Anguilla)作为我国优质水产养殖种类,精准掌握其数量对高效养殖有重要意义。为实现对循环水养殖鳗鲡的准确计数,提出了一种基于深度学习的改进Faster RCNN模型。针对检测目标即鳗鲡头部尺寸小的问题,选择在特征提取网络ResNet50中加入FPN结构来作为模型的骨干网络,以提取并融合多尺度的特征;针对原模型锚框都是基于人工经验设置的,并不适用于鳗鲡数据集的问题,使用k-means聚类算法对训练集中标注的鳗鲡头部检测框进行聚类分析,获得了适合鳗鲡数据集的15种不同尺度的锚框;针对图像中存在鳗鲡头部重叠的问题,选择使用Soft-NMS算法替代原NMS算法对RPN部分生成的候选框进行筛选,以减少模型对鳗鲡重叠部分的漏检情况。试验结果表明:改进后的Faster RCNN模型对鳗鲡头部的检测精度(mAP^0.5)高达96.5%,较原Faster RCNN模型(Backbone为ResNet50)显著提升了14%,与SSD300和YOLOV3模型相比分别显著提升了24.9%和15%;在鳗鲡计数上,利用改进后的Faster RCNN模型检测结果进行计数,计数准确率达到90%以上,提升了模型对鳗鲡的检测识别能力。     

高分辨率光学遥感场景分类的深度度量学习方法

针对高分辨率光学遥感影像场景具有同类型内部差异大、不同类型间相似度高导致部分场景识别困难的问题,本文提出了一种深度度量学习方法。首先在深度学习模型的特征输出层上为每类预设聚类中心,其次基于欧氏距离方法设计均值中心度量损失项,最后联合交叉熵损失项以及权重与偏置正则项构成模型的损失函数。该方法的目标是在特征空间上使同类型特征聚集并扩大类型间的距离以提高分类准确率。试验结果表明,本文方法有效地提升了分类准确率。在RSSCN7、UC Merced和NWPU-RESISC45数据集上,与现有方法相比,分类准确率分别提高了1.46%、1.09%和2.51%。