Integration of artificial intelligence with meta classifiers for the gully erosion susceptibility assessment in Hinglo river basin,Eastern India

The main aim of this study is to evaluate the gully erosion susceptibility coupling the artificial intelligence and machine learning ensemble approaches. In the present study, the multilayer perceptron neural network (MLP) was used as the base classifier and the hybrid ensemble machine learning methods i.e. Bagging and Dagging were used as the functional classifiers. The Hinglo river basin, an important tributary of the Ajay River was selected as the study area, consists with the parts of Chhotonagpur plateau and Rarh lateritic region. The study area is facing the gully erosion problems which are interrupted the growth of the agriculture. The gully erosion susceptibility maps (GESMs), prepared by MLP, MLP-Bagging and MLP-Dagging were classified into four classes such as low, moderate, high and very high susceptibility classes with the help of natural break method (NBM) in GIS environment. The very high susceptibility class covered 19.41% (MLP), 13.52% (MLP-Bagging) and 15.30% (MLP-Dagging) areas of the basin. For the evaluation and comparison of the models, receiver operating characteristics (ROC), accuracy, mean absolute error (MAE) and root mean square error (RMSE) were applied. Overall, all the gully erosion susceptibility models were performed as excellent. Integration of hybrid ensemble models with MLP has increase the accuracy of the MLP models. Among these models MLP-Dagging has achieved the highest accuracy in compare to the other models. The importance of the selected factors in the present study was assessed by the Relief-F method. The results show that the soil type factor has the highest predictive performance. Sensitivity analysis also showed soil type as most important factor. The gully erosion susceptibility maps (GESMs) are considered as the efficient tool which could be used to take the necessary steps for mitigating and controlling the soil erosion problem and sustainable environmental management and development.     

Spatial modeling of flood probability using geo-environmental variables and machine learning models,case study: Tajan watershed,Iran

The main objective of this study was to produce flood susceptibility maps for Tajan watershed, Sari, Iran using three machine learning (ML) models including Self-Organization Map (SOM), Radial Basis Function Neural Network (RBFNN), and Multi-layers Perceptron (MLP). To reach such a goal, different physical-geographical factors (criteria) were integrated and mapped. 212 flood inventory map was randomly divided into training and testing datasets, where 148 flood locations (70%) were used for training and the remaining 64 locations (30%) were employed for testing. Model validation was performed using several statistical indices and the area under the curve (AUC). The results of the correlation matrix showed, three factors slope (0.277), distance from river (0.263), and altitude (0.223) were the most important factors affecting flood. The accuracy evaluation of the flood susceptibility maps through the AUC method and K-index shows that in the validation phase RBFNN (AUC = 0.90) outperform the MLP (AUC = 0.839) and SOM (AUC = 0.882) models. The highest percentage flood susceptibility of the area in MLP, SOM and RBFNN models is related to moderate (28.7%), very low (40%) and low (37%), respectively. Also, the validation results of the models using the Relative Flood Density (RFD) approach showed that very high class had the highest RFD value.     

Exploration of groundwater potential zones mapping for hard rock region in the Jakham river basin using geospatial techniques and aquifer parameters

Geospatial techniques are useful to understand the groundwater resources assessment, development, and management. Groundwater mapping is essential to counter the excessive withdrawal of groundwater and fulfil the need of drinking, irrigation water in hard rock areas. Currently, the main problem is facing the groundwater level is decreasing due to less rainfall, climate changes and increasing demand, which is under in the hot zones. This results can be more useful for future challenges, sustainable development and demand to Jakham river basin. The main objective of this paper to demarcated the groundwater potential zones (GWPZ) using geospatial techniques for the hard rock region with the reference of aquifer parameters. In this paper, we have used to total eight thematic layers such as geomorphology, land use, soil, topographic elevation, slope, post-monsoon groundwater level (GWL), net recharge, and transmissivity, which are prepared from satellite data and field verification. All thematic layers were integrated for assigning the weights to demarcation of the groundwater potential zones in the RS and GIS environment. The selected thematic layers and features were assigned weightage and normalized by the analytic hierarchy process (AHP) technique. Finally, the thematic layers were systematically integrated using weighted overlay analysis within a GIS environment. We have classified into five GWPZ classes i.e. very high, high, moderate, low, and very low for basin area using GIS, and AHP methods. The study area result indicated that high and moderate zone, which is confined in the central part of the basin, covers 2.43 % and 43.88 % area, respectively. The low (49.21 %) and very low (4.25%) GW potential zone is under the confined aquifer in the high slope and rock outcrops formations near the basin boundary. The final GWPZ map was validated with groundwater level fluctuation data, which illustrates the accuracy of the adopted approach. This unique approach and conclusions of this work may also help to develop the framework and policies for swiftly analyzing groundwater recharge planning, development and locating the artificial recharge structures in the other semi-arid, arid and hard rock regions.     

An ensemble deep learning based shoreline segmentation approach (WaterNet) from Landsat 8 OLI images

Shorelines constantly vary due to natural, urbanization and anthropogenic effects such as global warming, population growth, and environmental pollution. Sustainable monitoring of coastal changes is vital in terms of coastal resource management, environmental preservation and planning. Publicly available Landsat 8 OLI (Operational Land Manager) images provide accurate, reliable, temporal and up-to-date information about coastal areas. Recently, the use of machine learning and deep learning algorithms have become widespread. In this study, we used our public Landsat 8 OLI satellite image dataset to create a majority voting method which is an ensemble automatic shoreline segmentation system (WaterNet) to obtain shorelines automatically. For this purpose, different deep learning architectures have been utilized namely as Standard U-Net, Dilated U-Net, Fractal U-Net, FC-DenseNet, and Pix2Pix. Also, we have suggested a novel framework to create labeling data from OpenStreetMap service to create a unique dataset called YTU-WaterNet. According to the results, IoU and F1 scores have been calculated as 99.59% and 99.79% for the WaterNet. The results indicate that the WaterNet method outperforms other methods in terms of shoreline extraction from Landsat 8 OLI satellite images.     

Estimation of stress and its use in evaluation of landslide prone regions using remote sensing data

Recently, use of remote sensing data for determining the orientation of stress has been demonstrated. The present study deals with the estimation of stress pattern in the part of the Himalayan region which shows the ongoing neo-tectonic activities. The study area falls into a tectonically active zone of the Central-Himalaya, with a complex geotectonic set-up confined by a number of faults. Efforts have been made to evaluate the technique as a fast algorithm for quick and time limited analysis of linear feature from which the orientation of the lineaments are estimated by using remote sensing data. Further, the estimation of stress and the lineament analysis have been used in mapping of landslide prone areas. Terrain information such as land cover, geology, lineament, faults, mega faults, geomorphology and drainage has been derived from the satellite imageries, and the existing thematic information has been updated to enable the quantification of landslide causative parameters. Spatial and temporal multi-layered information have been used for landslides hazard susceptibility analysis. The qualitative hazard analysis has been carried out using the map overlying techniques in GIS environment along the central part of Himalayan region. It has been observed that the high potential zones have been found to have very high lineament density, moderate to low drainage density and high slope areas of the terrain. On the basis of the geological and morphological analysis, it is further suggested that the combined impacts of the crushed nature of bed rock (due to the neo-tectonic activities), heavy rainfall and lack of vegetation cover cause persistent recurrence of landslides along this region. The role of earthquake on induction of landslides will be presented.     

Chlorophyll,calcite, and suspended sediment concentrations in the Bay of Bengal and the Arabian Sea at the river mouths

Chlorophyll and suspended sediment concentrations (SSC) and sea surface temperature (SST) are important parameters in assessing the productivity of coastal regions. Numerous rivers flow into the eastern (Ganga, Subernarekha, Mahanadi, Godavari, Krishna, Penner, and Kaveri) and western (Narmada, Tapti, and Indus) coasts of the Indian sub-continent. Using IRS P4 (Oceansat-1) Ocean Color Monitor (OCM) and Moderate Resolution Imaging Spectroradiometer (MODIS) data, we have retrieved chlorophyll, calcite, and SSC near the mouth of these rivers for the period during 2000–2004. The maxima of chlorophyll-a concentrations at the river mouth is much higher for the Himalayan and north India rivers (Ganga, Subernarekha, Mahanadi, and Indus) (10–14 mg/m³) compared to rivers in the southern parts of India (Kaveri and Penner) (∼4 mg/m³). The maxima of calcite concentration (∼45 moles/m³), chlorophyll (∼14 mg/m³), and sediment concentrations (∼9 g/m³) near river mouth are found to be influenced by river discharges (Ganga and Brahmaputra) during the monsoon season. The calcite concentration (∼45 moles/m³) at the mouth of Ganga river shows a major peak with the onset of monsoon season (June–July) followed by a maxima in chlorophyll-a with a time lag of 1–2 months. The Krishna, Kaveri, and Penner rivers show low chlorophyll concentrations (3–8 mg/m³), high calcite (0–40 moles/m³), and low SSC (<3 g/m³) compared to Narmada and Tapti rivers (chlorophyll-a 12–14 mg/m³, calcite 0–2 moles/m³, and SSC 13–19 g/m³). The Indus river shows similar behavior (maxima of chlorophyll ∼13 mg/m³ and SSC ∼8 g/m³) with respect to Ganga river except for high calcite concentration during winter months (∼25 moles/m³). The characteristics of the chlorophyll, calcite, and SSC at the mouth of these rivers show spatial and temporal variability along the eastern and westerns coasts of India which are found to differ widely. A comparison of the chlorophyll concentrations using OCM and MODIS data shows low chlorophyll concentrations in the Bay of Bengal as compared to the Arabian Sea.     

An approach to multi-temporal MODIS image analysis using image classification and segmentation

This paper discusses an approach for river mapping and flood evaluation based on multi-temporal time series analysis of satellite images utilizing pixel spectral information for image classification and region-based segmentation for extracting water-covered regions. Analysis of MODIS satellite images is applied in three stages: before flood, during flood and after flood. Water regions are extracted from the MODIS images using image classification (based on spectral information) and image segmentation (based on spatial information). Multi-temporal MODIS images from “normal” (non-flood) and flood time-periods are processed in two steps. In the first step, image classifiers such as Support Vector Machines (SVM) and Artificial Neural Networks (ANN) separate the image pixels into water and non-water groups based on their spectral features. The classified image is then segmented using spatial features of the water pixels to remove the misclassified water. From the results obtained, we evaluate the performance of the method and conclude that the use of image classification (SVM and ANN) and region-based image segmentation is an accurate and reliable approach for the extraction of water-covered regions.     

比特并行Reed—Solomon编码器的设计

研究高速ＲＳ码编码器设计问题。给出了最优对偶基的计算方法,研究了用对偶基下的ｂｉｔ－ｐａｒａｌｌｅｌ乘法器构成ＲＳ码系统码编码器。编码器可以达到较高的吞吐率。     

Mapping recent agricultural developments in China from satellite data

Landsat data have been employed to study and map agricultural developments in three regions of China: 1) Pearl River delta; 2) Nen River basin; and 3) Xinjiang Autonomous Region. Manual interpretation procedures used in conjunction with multi-date Landsat images and collateral information permitted rice yields to be estimated for the Pearl River delta in 1978. A combination of manual and computer-assisted analyses of Landsat data of Northeast China revealed that more than 15,000 km² of agricultural land in a 184,500 km² study area had been reclaimed from rangeland and marshland. These analyses also indicated a shift in cropping practices, with the foodcrops wheat and corn replacing cash crops such as soybeans. In the arid west, Landsat image data provided valuable input to a geographic information system (GIS). It appears the GIS approach will prove useful for evaluating agricultural land potential in the remote areas of China.     

一种平均矩独立重要性指标及其拒绝抽样方法

在矩独立重要性分析过程中,重要性指标往往用于衡量结构系统输出不确定性向输入变量不确定性的逆向分配问题。假设输入参数的方差可以减缩一定比例因子,那么矩独立重要性指标可以定义为该缩减因子的函数。同时,假设输入参数的方差缩减因子为一随机变量,那么可以取矩独立重要性指标函数的均值定义一个新的平均矩独立重要性指标。由于使用Sobol方法计算平均矩独立重要性指标的模型需要循环抽样,计算量很高,故引入拒绝抽样（RS）方法,通过重复利用矩独立重要性分析中的一组输入输出样本,就可以额外计算得到矩独立指标函数和平均矩独立重要性指标,这大大节约了计算成本。本文所提指标函数及平均指标的有效性和RS方法的准确性、高效性通过数值和工程算例得以验证。      

Hydro-geomorphic characteristics of the Indian (Peninsular) catchments: Based on morphometric correlation with hydro-sedimentary data

In terms of hydro-geomorphic characteristics, catchments in Peninsular India remained mostly unexplored except a few regional and local works that deal with tectonic, structural and paleo-climatic control on geomorphology. Catchment scale morphometric analyses deliver insights into dynamics, erosion capacity, probability of flood occurrence, lithological and structural control, and genetic response to the tectonics. The present study aimed to comprehend hydro-geomorphic characteristics of 12 major catchments in Peninsular India through GIS-based morphometric analysis. A total of 25 morphometric parameters were computed and several statistical analyses performed in establishing inter-correlation and classification of Indian rivers. Most of the rivers in Peninsular India were found 7th to 9th order catchments. Almost all basins showed a moderate relief ratio, hypsometric integral, ruggedness etc. Cauvery, Baitarni, and Brahmani showed exceptionally steeper gradient, high relief ratio, LS factor, and ruggedness index, which indicated higher erosion potential. Correlation among landscape variables revealed moderate scale dependency of few relief factors. Baitarni, Brahmani and Narmada showed higher hypsometric integral. A strong positive association between hypsometric integral and sediment yield suggested critically high erosion potential in catchments with high integral values. The present study provides some generic insights into the hydro-geomorphic characteristics with dissimilarity in lithology in Peninsular Indian catchments as a whole.     

Dynamics of total suspended solid concentrations in the lower Raidak river (Himalayan foreland Basin), India

Accurate assessment of total suspended solids (TSS) is one of the most important parameters for the management of health of aquatic ecosystem. Due to the limitation of traditional TSS measurement methods, recently developed remote sensing based algorithm (Wen algorithm) has been used to measure the TSS concentration in a Himalayan foothill river like Raidak. Additionally, to examine the spatiotemporal characteristics of total suspended solids (TSS), total three consecutive years (2019, 2020 and 2021) along with different seasons have been selected. The Nechad algorithm has been used to validate the recalibrated algorithm. The result also indicates that Wen algorithm is also highly consistent with the Nechad algorithm (R² value is greater than 0.90). It is also estimated that the concentration of TSS becomes very high particularly in monsoon months (242 mg/l, 270 mg/l and 246 mg/l for the year 2019, 2020 and 2021). The Raidak river catchment is located in the foothill zone of the lofty Himalayan range and the region is totally influenced by Indian summer monsoon. In the Raidak river course, the seasonal sediment flux is highly correlated with rainfall, stream flow, cross sectional area and high anthropogenic stress. The NDTI (Normalize Difference Turbidity Index) value (Turbidity) is greater in monsoon season than pre and post monsoonal periods (0.035, 0.0851 and 0.0201 for the years 2019, 2020 and 2021 respectively). ANOVA result shows a significant difference among the TSS values in different years (p=<0.05).     

Calculation of flood hydrographs using Landsat-derived land-use information in the Dreisam watershed,South-West Germany

Flood calculations in a watershed in South-West Germany have been carried out using a geographical data base including land-use, soil information and slope with a resolution of 64×104 m. The land-use information is result of a maximum-likelihood classification of MSS-data. As model for the calculation of the flood hydrographs the SCS-TR 20 was used. The single event calculations as well as the calculations using 24h rainfall of different return periods match the measured hydrographs well. The effect of a possible deforestation of the area due to forest damages has been simulated using 5 different scenarios. According to the simulations a total deforestation causes a 100 year peak discharge 5 times as high as with the current forest distribution.     

基于置信度的TOF与双目系统深度数据融合

为了解决铁路火车等结构单一环境的三维重建问题,提出了置信度的概念,将TOF系统与双目系统的互补性特点有效结合。通过联合标定,建立起TOF系统与双目系统的坐标关系,将TOF中的点映射到左相机视角下,得出双目系统左相机视角下的TOF测量视差图,再利用图像分割以及曲面拟合对其上采样处理至双目图像的分辨率大小。根据各系统特点定义置信度,确定数据融合的不同系统权重。利用Middlebury的数据集处理结果,融合后的匹配精度较双目系统精度提高一倍以上,且视差图的分辨率提升至与双目系统相同大小。      

基于边缘和结构的无参考屏幕内容图像质量评估

屏幕内容图像（SCI）是一种与传统自然图像不同的图像,具有更多的文本、图形以及特殊的布局。考虑文本、图形、图像和布局对屏幕内容图像质量的影响,提出了针对屏幕内容图像的基于边缘和结构的无参考质量评估（BES）算法。文本、图形和图像具有大量边缘,并且人类视觉系统对边缘高度敏感,因此BES算法首先使用Gabor滤波器的虚部提取边缘并计算每张屏幕内容图像的边缘特征。其次,提取一个结构特征来表示屏幕内容图像的布局。具体而言,利用Scharr滤波器计算得到一个局部二值模式（LBP）图,接着利用LBP图计算得到结构特征。最后,应用随机森林回归算法将边缘和结构特征映射为主观分数。实验结果表明,在数据库SIQAD和SCID上,所提出BES算法性能的皮尔森线性相关系数（PLCC）相对于对比算法中最先进的无参考算法,分别提高了2.63%和11.22%,甚至高于一些全参考算法。      

长波定时ASF修正的GIS软件

论述的长波定时ASF修正的GIS软件能快速数字化可视化显示和调用服务区域的地形地理信息、大地电参数和大气折射率等多种数据源。使用此软件可以非常方便容易地用不同模型计算在广阔区域接收到的多个长波信号在各种典型路径上的ASF和地波传播时间延迟,并提高定时精度。     

Remote sensing and GIS-based landslide hazard analysis and cross-validation using multivariate logistic regression model on three test areas in Malaysia

This paper presents the results of the cross-validation of a multivariate logistic regression model using remote sensing data and GIS for landslide hazard analysis on the Penang, Cameron, and Selangor areas in Malaysia. Landslide locations in the study areas were identified by interpreting aerial photographs and satellite images, supported by field surveys. SPOT 5 and Landsat TM satellite imagery were used to map landcover and vegetation index, respectively. Maps of topography, soil type, lineaments and land cover were constructed from the spatial datasets. Ten factors which influence landslide occurrence, i.e., slope, aspect, curvature, distance from drainage, lithology, distance from lineaments, soil type, landcover, rainfall precipitation, and normalized difference vegetation index (ndvi), were extracted from the spatial database and the logistic regression coefficient of each factor was computed. Then the landslide hazard was analysed using the multivariate logistic regression coefficients derived not only from the data for the respective area but also using the logistic regression coefficients calculated from each of the other two areas (nine hazard maps in all) as a cross-validation of the model. For verification of the model, the results of the analyses were then compared with the field-verified landslide locations. Among the three cases of the application of logistic regression coefficient in the same study area, the case of Selangor based on the Selangor logistic regression coefficients showed the highest accuracy (94%), where as Penang based on the Penang coefficients showed the lowest accuracy (86%). Similarly, among the six cases from the cross application of logistic regression coefficient in other two areas, the case of Selangor based on logistic coefficient of Cameron showed highest (90%) prediction accuracy where as the case of Penang based on the Selangor logistic regression coefficients showed the lowest accuracy (79%). Qualitatively, the cross application model yields reasonable results which can be used for preliminary landslide hazard mapping.     

基于R-D SSD模型航空发动机安装工位检测算法

为解决航空发动机在安装过程中大多实行人工安装、定位不精确等问题，在研究其自动化安装方法中，针对航空发动机安装工位的检测需求，提出了一种残差网络与膨胀卷积相融合的SSD改进算法(R-D SSD)。将经典SSD模型的主干网络VGG16替换为残差网络ResNet-101，并增加其输出特征图上的预选框数量，解决了原始算法对底层特征抓取能力不足的问题，进而弥补了对小目标检测效果较差的缺陷；利用膨胀卷积扩大网络的感受野，获取足够的安装工位边缘特征细节信息，在不改变网络结构的同时，保证了模型良好的实时性和对目标的检测精度。实验表明:对于小目标数据集和整个数据集，R-D SSD算法的平均检测精度较原始算法分别提高了8.6%和4.0%，可以满足航空发动机安装时平均检测精度不低于85%的要求。      

Study on extraction of crop information using time-series MODIS data in the Chao Phraya Basin of Thailand

In order to acquire the crop-related information in Chao Phraya Basin, time-series MODIS data were used in this paper. Although the spatial resolution of MODIS data is not very high, it is still useful for detecting very large-scale phenomenon, such as changes in seasonal vegetation patterns. After the data processing a general crop-related LULC (land use and land cover) map, cropping intensity map and cropping patterns map were produced. Analysis of these maps showed that the main land use type in the study area was farmland, most of which was dominated by rice. Rice fields mostly concentrated in the flood plains and double or triple rice-cropping system was commonly employed in this area. Maize, cassava, sugarcane and other upland crops were mainly distributed in the high alluvial terraces. Because these area often have water shortage problem particularly in the dry season which can support only one crop in a year, the cropping intensity was very low. However, some upland areas can be cultivated twice a year with crops which have short growing seasons. The crop information extracted from MODIS data sets were assessed by CBERS data, statistic data and so on. It was shown that MODIS derived crop information coincided well with the statistic data at the provincial level. At the same time, crop information extracted by MODIS data sets and CBERS were compared with each other which also showed similar spatial patterns.     

Response of the EIA ionosphere to the 7-8 May 2005 geomagnetic storm

In this paper, response of low latitude ionosphere to a moderate geomagnetic storm of 7–8 May 2005 (SSC: 1920 UT on 7 May with Sym-H minimum, ∼−112 nT around 1600 UT on 8 May) has been investigated using the GPS measurements from a near EIA crest region, Rajkot (Geog. 22.29°N, 70.74°E, Geomag.14°), India. We found a decrease in total electron content (TEC) in 12 h after the onset of the storm, an increase during and after 6 h of Sym-H deep minimum with a decrease below its usual-day level on the second day during the recovery phase of the storm. On 8 May, an increase of TEC is observed after sunset and during post-midnight hours (maximum up to 170%) with the formation of ionospheric plasma bubbles followed by a nearly simultaneous onset of scintillations at L-band frequencies following the time of rapid decrease in Sym-H index (−30 nT/h around 1300 UT).