Estimating the parameters of forest inventory using machine learning and the reduction of remote sensing features

Locally computed statistics of image texture and a case-based reasoning (CBR) system were evaluated for mapping of forest attributes. Cluster analysis was preferred to regression models, as a pre-selection method of features. The best stand-based accuracy using satellite sensor images was 74.64 m⁻³ ha⁻¹ (36%) RMSE for stand volume, 1.98 m⁻³ ha⁻¹ a⁻¹ (49%) for annual increase in stand volume, where κ = 0.23 for stand growth classes and κ = 0.41 for dominant tree species in stands. The top pixel-based accuracy using orthophotos was 76.54 m⁻³ ha⁻¹ (41%) RMSE for stand volume, 1.87 m⁻³ ha⁻¹ a⁻¹ (44%) for annual increase in stand volume, where κ = 0.24 for stand growth classes and κ = 0.38 for dominant tree species in stands. Mean saturation in 30 m radius was the most useful feature when orthophotos were used, and standard deviation of Landsat ETM 6.2 values in 80 m radius was the best when satellite sensor images were used. The most valuable feature components (radii, channels and local statistics) for orthophotos were: 30 m kernel radius, lightness and the mean of pixel values; for satellite sensor images: 80 m kernel radius, near-infrared channel (ETM 4) and the mean of pixel values. Locally computed statistics.     

气溶胶定量遥感的机器学习方法综述

机器学习方法近年来取得突破进展,其遥感应用从目标识别和地物分类领域,发展到定量化反演的多个领域。气溶胶定量遥感因其机理复杂,反演参数的种类和精度受到限制,机器学习为气溶胶遥感带来了新的研究和应用技术手段。本文汇总现有研究进展将气溶胶机器学习方法归纳为卫星遥感反演气溶胶光学厚度AOD（Aerosol Optical Depth）、卫星遥感反演其他气溶胶参数、卫星遥感反演颗粒物浓度（PMx）、地基气溶胶遥感4类。结合作者研究工作,通过分析讨论,归纳机器学习用于气溶胶定量遥感的条件为：（1）物理模型无法使用;（2）已有模型卫星产品精度低;（3）已有模型精度高但计算速度低。从应用的角度来说,可以借助于更多的具有相关性的输入信息,发挥机器学习在反演产品种类、反演精度、计算效率等方面的优势;而对定量遥感来说,应该同时重视挖掘遥感数据本身的信息来提高反演能力,并通过误差分析等手段反馈对遥感机理的理解,使机器学习与遥感机理研究相互促进。     

An assessment of biodiversity hotspots using remote sensing and GIS

This study focuses on the assessment of the status of world’s remaining closed forests (WRCF), population distribution, and protected areas in global biodiversity hotspots using remote sensing and Geographic Information System (GIS). Conservation International (CI) has identified 25 eco-regions, called biodiversity hotspots that are especially rich in endemic species and are particularly threatened by human activities. This study uses globally consistent and comprehensive geo-spatial data sets generated using rerriote sensing and other sources, and the application of GIS layering methods. The consistent data set has made it possible to identify and quantify relationships between the WRCF, human population, and protected areas in biodiversity hotspots. It is expected that such information will provide a scientific basis for biodiversity hotspots management and assist in policy formulations at the national and international levels.     

Soil resource assessment and mapping using remote sensing and GIS

In this study digital image processing for physiographic analysis and soil resource mapping of Solani watershed was carried out using satellite remote sensing data and GIS. Digital image processing of satellite data facilitated in accurately delineating and identifying various soil mapping units. The physiography of the study area is mainly influenced by denudational and colluvial processes in the upper part and by sedimentation processes in the lower part. Topography of the land and nature of parent material along with the time factor seemed to have played a vital role in the genesis of soils. Majority of the mapping units are Typic Haplustepts with Entisols and Inceptisols being the major soil orders. The soils of the Siwalik hills experiences severe erosion, which prevents the maturation of soil profile. The present study demonstrated that satellite remote sensing and GIS is a valuable tool for physiographic analysis and soil resource mapping.     

Estimating fractional cover of crop,crop residue,and soil in cropland using broadband remote sensing data and machine learning

The fractional vegetation cover (FVC), crop residue cover (CRC), and bare soil (BS) are three important parameters in vegetation–soil ecosystems, and their correct and timely estimation can improve crop monitoring and environmental monitoring. The triangular space method uses one CRC index and one vegetation index to create a triangular space in which the three vertices represent pure vegetation, crop residue, and bare soil. Subsequently, the CRC, FVC, and BS of mixed remote sensing pixels can be distinguished by their spatial locations in the triangular space. However, soil moisture and crop-residue moisture (SM-CRM) significantly reduce the performance of broadband remote sensing CRC indices and can thus decrease the accuracy of the remote estimation and mapping of CRC, FVC, and BS. This study evaluated the use of broadband remote sensing, the triangular space method, and the random forest (RF) technique to estimate and map the FVC, CRC, and BS of cropland in which SM-CRM changes dramatically. A spectral dataset was obtained using: (1) from a field-based experiment with a field spectrometer; and (2) from a laboratory-based simulation that included four distinct soil types, three types of crop residue (winter-wheat, maize, and rice), one crop (winter wheat), and varying SM-CRM. We trained an RF model [designated the broadband crop-residue index from random forest (CRRF)] that can magnify spectral features of crop residue and soil by using the broadband remote sensing angle indices as input, and uses a moisture-resistant hyperspectral index as the target. The effects of moisture on crop residue and soil were minimized by using the broadband CRRF. Then, the CRRF-NDVI triangular space method was used to estimate and map CRC, FVC, and BS. Our method was validated by using both laboratory- and field-based experiments and Sentinel-2 broadband remote-sensing images. Our results indicate that the CRRF-NDVI triangular space method can reduce the effect of moisture on the broadband remote-sensing of CRC, and may also help to obtain laboratory and field CRC, FVC, and BS. Thus, the proposed method has great potential for application to croplands in which the SM-CRM content changes dramatically.     

基于多核学习的高分辨率遥感图像目标检测方法

为更有效地实现复杂场景中的多类目标同时检测,本文提出了一种基于多核学习算法进行目标检测的框架。该方法由特征提取和模型训练2个阶段组成。特征提取阶段,引入了多尺度下的点特征、表观特征同时对多类目标进行综合描述;模型训练阶段,分别采用加权相加和相乘2种方法将提取的各个基础特征组合起来,在支持向量机的框架下对各特征所代表的基础核权重进行学习。将训练所得的分类器结合滑动窗搜索技术对遥感图像进行目标检测实验,结果表明,与传统单核支持向量机相比,准确率更高。     

Sustainability assessment of sodic land reclamation using remote sensing and GIS

Sodicland reclamation in the Indo-Gangetic plains is being done on a large scale in the states of Uttar Pradesh, Punjab and Haryana in India. However, in certain areas, the reclamation has been reported to be unsustainable and the soils are reverting back to sodicity condition. A study was conducted in one of the reclamation sites of Etawah district for sustainability assessment of sodic land reclamation using remote sensing, Geographic Information system (GIS) and ancillary ground information. Multitemporal satellite data were used for delineation of reclaimed sodiclands and reverted sodic land. Field survey was conducted to find out the various causative factors. Groundwater level information and detailed field survey data were analysed in GIS environment. Results showed that in the reclamation site covering 3,905 ha. in 57 villages of the district, about 27 per cent of reclaimed lands were reverted to sodicity. High water table condition, improper drainage, nearness to canal (within 500 m), and hard pan in the sub-soil were found to be the reasons for unsustainability of reclamation.     

Land cover classification and economic assessment of citrus groves using remote sensing

The citrus industry has the second largest impact on Florida's economy, following tourism. Estimation of citrus area coverage and annual forecasts of Florida's citrus production are currently dependent on labor-intensive interpretation of aerial photographs. Remotely sensed data from satellites has been widely applied in agricultural yield estimation and cropland management. Satellite data can potentially be obtained throughout the year, making it especially suitable for the detection of land cover change in agriculture and horticulture, plant health status, soil and moisture conditions, and effects of crop management practices. In this study, we analyzed land cover of citrus crops in Florida using Landsat Enhanced Thematic Mapper Plus (ETM+) imagery from the University of Maryland Global Land Cover Facility (GLCF). We hypothesized that an interdisciplinary approach combining citrus production (economic) data with citrus land cover area per county would yield a correlation between observable spectral reflectance throughout the year, and the fiscal impact of citrus on local economies. While the data from official sources based on aerial photography were positively correlated, there were serious discrepancies between agriculture census data and satellite-derived cropland area using medium-resolution satellite imagery. If these discrepancies can be resolved by using imagery of higher spatial resolution, a stronger correlation would be observed for citrus production based on satellite data. This would allow us to predict the economic impact of citrus from satellite-derived spectral data analysis to determine final crop harvests.     

Accuracy assessment in cotton acreage estimation using Indian remote sensing satellite data

The accuracy of cotton crop classification using satellite data has been assessed with respect to a detailed land cover map prepared by field survey. The effect of spatial resolution on classification accuracy was studied using LISS-I (spatial resolution 72.6 m) and LISS-II data (spatial resolution 36.25 m) of the Indian remote sensing satellite IRS-1B. The performances of the maximum likelihood and the minimum distance to mean as classifiers have also been assessed. LISS-II data have been found to give a higher classification accuracy. The estimate of cotton acreage using LISS-II data was closer to that obtained from the base map. The maximum likelihood classifier (MXL) and the minimum distance to mean (MDM) classifier performed equally well.     

Prioritization of watersheds using morphometric parameters and assessment of surface water potential using remote sensing

In the present study, prioritization of sub-watersheds was carried out on the basis of sediment production rate. Further, basic hydrologic information such as peak rate of runoff and annual surface water potential were also assessed for the study watersheds and these are essential requisites for effective watershed management. The 10 sub watersheds of Tarai development project area are selected for the present study. Morphometric parameters pertaining to study area are used in the estimation of sediment production rate. The sediment production rate in the study area varies between 2.45 to 11.0 ha-m/100 km²/year. The remote sensing data has been utilized for generating land use/land cover data which is an essential prerequisite for land and water resource planning and development. The remote sensing data can especially play significant role in collection of real time information from remote areas of river basins for generation of parameters required for hydrologic modeling.     

多分类SVM主动学习及其在遥感图像分类中的应用

针对应用传统分类器和被动学习的方法,难以满足遥感图像处理实际应用的要求这一困境,提出了一种新的基于多分类SVM的主动学习方法,与被动学习的随机选择不同,主动学习是在少量标记类别的初始训练样本集基础上,通过反复迭代主动学习的方式,得到最有利SVM分类器性能的样本为支持向量。研究表明,这种方法直接避免了大量的计算,可有效地减少样本训练时需要标记样本的数目,并取得较为理想的分类效果。     

Integration of remote sensing,county-level census,and machine learning for century-long regional cropland distribution data reconstruction

The Lower Mississippi Alluvial Valley (LMAV) was home to about ten million hectare bottomland hardwood (BLH) forests in the Southern U.S. It experienced over 80 % area loss of the BLH forests in the past centuries and large-scale afforestation in recent decades. Due to the lack of a high-resolution cropland dataset, impacts of land use change (LUC) on the LMAV ecosystem services have not been fully understood. In this study, we developed a novel framework by integrating the machine learning algorithm, county-level agricultural census, and satellite-based cropland products to reconstruct the LMAV cropland distribution during 1850–2018 at a 30-m resolution. Results showed that the LMAV cropland area increased from 0.78 × 10⁴ km² in 1850 to 6.64 × 10⁴ km² in 1980 and then decreased to 6.16 × 10⁴ km² in 2018. Cropland expansion rate was the largest in the 1960s (749 km² yr⁻¹) but decreased rapidly thereafter, whereas cropland abandonment rate increased substantially in recent decades with the largest rate of 514 km² yr⁻¹ in the 2010s. Our dataset has three notable features: (1) the depiction of fine spatial details, (2) the integration of the county-level census, and (3) the inclusion of a machine-learning algorithm trained by satellite-based land cover product. Most importantly, our dataset well captured the continuous increasing trend in cropland area from 1930–1960, which was misrepresented by other cropland datasets reconstructed from the state-level census. Our dataset would be important to accurately evaluate the impacts of historical deforestation and recent afforestation efforts on regional ecosystem services, attribute the observed hydrological changes to anthropogenic and natural driving factors, and investigate how the socioeconomic factors control regional LUC pattern. Our framework and dataset are crucial to developing managerial and policy strategies for conserving natural resources and enhancing ecosystem services in the LMAV.     

高分辨率遥感影像解译中的机器学习范式

高分辨率遥感影像解译是遥感信息处理领域的研究热点之一,在遥感大数据知识挖掘与智能化分析中起着至关重要的作用,具有重要的民用和军事应用价值.传统的高分辨率遥感影像解译通常采用人工目视解译方式,费时费力且精度低.所以,如何自动、高效地实现高分辨率遥感影像解译是亟待解决的问题.近年来,随着人工智能技术的飞速发展,采用机器学习...     

Understanding spatial variations of malaria in Vietnam using remotely sensed data integrated into GIS and machine learning classifiers

Abstract

This study examines the potentials of remotely sensed data, GIS and some machine learning classifiers and ensemble techniques in the investigation of the non-linear relationship between malaria occurrences and socio-physical conditions in the Dak Nong province of Viet Nam. Accuracy assessment was determined with Receiver Operating Characteristic (ROC) curve and pair t-test. The results showed that the area under ROC of Random Subspace ensemble model performed better than the other models based on statistical indicators. Comparing pair t-test with Area Under Curve values showed a slight difference of about 1%. Therefore ensemble techniques had significantly improved the performance of the base classifier. However, the performances might vary according to geographic locations. It is concluded that the machine learning classifiers combined with remotely sensed data and GIS is promising for malaria vulnerability mapping, and the derived maps can be used as a fundamental basis for programmes on spatial disease control.     

Segregation of soilscapes using statistics in remote sensing

The use of remotely sensed data in the form of vertical aerial photographs has been in practice since several decades. Launching of Landsat satellites and their capability together multispectral scanner data afforded soil scientists enhanced capability for mapping soils. Technological advances in computer processing of Landsat MSS data coupled with ancillary information provided added advantages. Still difficulties exist in identifying the interference to vegetation and in segregation of narrowly defined soilscapes. Utility of statistical data, obtainable from computer analysis, and an aid in understanding the reflectance characteristics of various types of vegetation and soilscapes has been discussed in this paper. The final output obtained by computer processing compared with existing soil maps of the area registered more future prospects for utilisation of statistics in segregation of soilscapes and vegetation.     

Deep learning for geometric and semantic tasks in photogrammetry and remote sensing

ABSTRACT

During the last few years, artificial intelligence based on deep learning, and particularly based on convolutional neural networks, has acted as a game changer in just about all tasks related to photogrammetry and remote sensing. Results have shown partly significant improvements in many projects all across the photogrammetric processing chain from image orientation to surface reconstruction, scene classification as well as change detection, object extraction and object tracking and recognition in image sequences. This paper summarizes the foundations of deep learning for photogrammetry and remote sensing before illustrating, by way of example, different projects being carried out at the Institute of Photogrammetry and GeoInformation, Leibniz University Hannover, in this exciting and fast moving field of research and development.     

Assessment of sedimentation in the Linganmakki Reservoir using remote sensing

The reservoir surveys carried out using the conventional equipments like theodolite, plane table, sextant, range finders, sounding rods, echo-sounders and slow moving boats etc. are time consuming and sometime it takes up to three years to complete the survey of a major reservoir like Hirakud. During such longtime of survey, the siltation pattern and the bed levels also get changed. So, the need to update the sediment measurement techniques and to introduce latest technology available in the field was felt to overcome the difficulties faced in the conventional method especially in major reservoirs. With this objective, the present study of the assessment of the sedimentation in Linganmakki reservoir is carried out using remote sensing and GIS techniques. The revised capacities in the zone of reservoir levels (533.05 – 548.78 m) for the year 1989–1990 and 2001–2002 in the zone of 532.20 – 545.91 m of reservoir levels were determined, which is found 2837.84 M cum and 2207.95 M cum respectively. Based on these results, the sedimentation rates in the zone (533.05 – 548.78 m) come out to be 8.96 ham/100 km²/year and for the zone (532.20 – 545.91 m) is 8.57 ha-m/100 km²/year. It is found that the results are comparable during and the uniform sediment rate of 8.57 ha-m/100 km²/ year can be considered.     

Structure and its impact in Uthangarai and Thirthamalai region, Tamil Nadu, India using remote sensing technique

The Uthangarai-Thirthamalai region in Dharmapuri district is a typical Precambrian tract in Tamil Nadu. IRS-1C satellite imagery was used for interpreting folds, faults, fractures and lineaments, in this region. As interpreted from the satellite imagery, there are five major structural domains, which control the Ponnaiyar, Pambar, Vaniyar and Kovilar river courses and other drainage pattern in this region. Presence of fault rock, epidotization and sympathetic fractures along such folds and faults reveals the intensity of tectonism in this sector. The interpreted lineaments have been classified into five major groups on the basis of their geometry and orientation. From the rose diagram, it is inferred that the NNW-SSE trending lineaments are fall in major frequency domain. The springs observed near Hanumanthirtham and Thirthamalai region could be structurally originated. The geomorphology interpreted from the satellite data has shown significant morphological expression of folds, faults and lineaments. The geophysical data have been analysed and correlated with intensity of weathering by fractures and lineaments.