Pub Date : 2022-01-02DOI: 10.1080/07038992.2021.1908118
R. H. Fraser, D. Pouliot, Jurjen van der Sluijs
Abstract Reindeer lichens (Cladonia spp.) are an important food source for woodland and barren ground caribou herds. In this study, we assessed Cladonia classification accuracy in a rocky, Canadian Shield landscape near Yellowknife, Northwest Territories using both Unmanned Aerial Vehicle (UAV) sensors and high-resolution satellite sensors. At the UAV scale, random forest classifications derived from a multispectral, visible-near infrared sensor (Micasense Altum) had an average 5% higher accuracy for mapping Cladonia (i.e., 95.5%) than when using a conventional color RGB camera (DJI Phantom 4 RTK). We aggregated Altum lichen classifications from three 5 ha study sites to train random forest regression models of fractional lichen cover using predictor features from WorldView-3 and Planet CubeSat satellite imagery. WorldView models at 6 m resolution had an average 6.8% RMSE (R 2 = 0.61) when tested at independent study sites and outperformed the 6 m Planet models, which had a 9.9% RMSE (R 2 = 0.34). These satellite results are comparable to previous lichen mapping studies focusing on woodlands, but the small cover of Cladonia in our study area (11.6% or 16.8% within the barren portions) results in a high relative RMSE (62.2%) expressed as a proportion of mean lichen cover.
{"title":"UAV and High Resolution Satellite Mapping of Forage Lichen (Cladonia spp.) in a Rocky Canadian Shield Landscape","authors":"R. H. Fraser, D. Pouliot, Jurjen van der Sluijs","doi":"10.1080/07038992.2021.1908118","DOIUrl":"https://doi.org/10.1080/07038992.2021.1908118","url":null,"abstract":"Abstract Reindeer lichens (Cladonia spp.) are an important food source for woodland and barren ground caribou herds. In this study, we assessed Cladonia classification accuracy in a rocky, Canadian Shield landscape near Yellowknife, Northwest Territories using both Unmanned Aerial Vehicle (UAV) sensors and high-resolution satellite sensors. At the UAV scale, random forest classifications derived from a multispectral, visible-near infrared sensor (Micasense Altum) had an average 5% higher accuracy for mapping Cladonia (i.e., 95.5%) than when using a conventional color RGB camera (DJI Phantom 4 RTK). We aggregated Altum lichen classifications from three 5 ha study sites to train random forest regression models of fractional lichen cover using predictor features from WorldView-3 and Planet CubeSat satellite imagery. WorldView models at 6 m resolution had an average 6.8% RMSE (R 2 = 0.61) when tested at independent study sites and outperformed the 6 m Planet models, which had a 9.9% RMSE (R 2 = 0.34). These satellite results are comparable to previous lichen mapping studies focusing on woodlands, but the small cover of Cladonia in our study area (11.6% or 16.8% within the barren portions) results in a high relative RMSE (62.2%) expressed as a proportion of mean lichen cover.","PeriodicalId":48843,"journal":{"name":"Canadian Journal of Remote Sensing","volume":"48 1","pages":"5 - 18"},"PeriodicalIF":2.6,"publicationDate":"2022-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1080/07038992.2021.1908118","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41697287","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-01-02DOI: 10.1080/07038992.2022.2024683
C. Hopkinson, C. Coburn, L. Chasmer
Dedication This Special Issue is dedicated to the memory of our friend and colleague, Dr. Martin Isenburg. Martin made valuable and colorful contributions to our symposium by hosting a workshop and giving a video presentation from his home and ’laser’ chicken farm in Costa Rica. The creator of the widely popular LAStools software, and an avid traveler and trainer in the international lidar community, he fell victim to the global pandemic in 2021. He will be sadly missed by all whose lives he touched.
{"title":"41st Canadian Symposium on Remote Sensing Special Issue: A Virtual Conference","authors":"C. Hopkinson, C. Coburn, L. Chasmer","doi":"10.1080/07038992.2022.2024683","DOIUrl":"https://doi.org/10.1080/07038992.2022.2024683","url":null,"abstract":"Dedication This Special Issue is dedicated to the memory of our friend and colleague, Dr. Martin Isenburg. Martin made valuable and colorful contributions to our symposium by hosting a workshop and giving a video presentation from his home and ’laser’ chicken farm in Costa Rica. The creator of the widely popular LAStools software, and an avid traveler and trainer in the international lidar community, he fell victim to the global pandemic in 2021. He will be sadly missed by all whose lives he touched.","PeriodicalId":48843,"journal":{"name":"Canadian Journal of Remote Sensing","volume":"48 1","pages":"1 - 4"},"PeriodicalIF":2.6,"publicationDate":"2022-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46107737","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-01-02DOI: 10.1080/07038992.2021.2014797
L. McLeod, Evan R. DeLancey, Erin M. Bayne
Abstract Yellow Rail (Coturnicops noveboracensis) are a highly specialized wetland obligate bird. They are a species at risk in Canada and very little is known about their abundance in the wetlands of the western boreal forest. Emerging technologies have enabled us to effectively survey for Yellow Rail and other wetland birds in remote areas by using ground-based remote sensors (autonomous recording units; ARUs) to conduct passive acoustic monitoring. We analyzed bird data from the first four years (2013–2016) of an ongoing monitoring program led by the Bioacoustic Unit at the Alberta Biodiversity Monitoring Institute. We developed species abundance models using satellite data from Sentinel-1 and Sentinel-2 processed in Google Earth Engine. We identified covariates from both synthetic aperture radar and optical remote sensing that had strong predictive capacity for this wetland bird (AUC = 0.96). Approximately 1.5% of available wetland habitat in our northeast Alberta study area was predicted to be highly suitable for Yellow Rail.
{"title":"Spatially Explicit Abundance Modeling of a Highly Specialized Wetland Bird Using Sentinel-1 and Sentinel-2 Modélisation spatialement explicite de l’abondance d’un oiseau très spécifique aux zones humides à l’aide de Sentinel-1 et de Sentinel-2","authors":"L. McLeod, Evan R. DeLancey, Erin M. Bayne","doi":"10.1080/07038992.2021.2014797","DOIUrl":"https://doi.org/10.1080/07038992.2021.2014797","url":null,"abstract":"Abstract Yellow Rail (Coturnicops noveboracensis) are a highly specialized wetland obligate bird. They are a species at risk in Canada and very little is known about their abundance in the wetlands of the western boreal forest. Emerging technologies have enabled us to effectively survey for Yellow Rail and other wetland birds in remote areas by using ground-based remote sensors (autonomous recording units; ARUs) to conduct passive acoustic monitoring. We analyzed bird data from the first four years (2013–2016) of an ongoing monitoring program led by the Bioacoustic Unit at the Alberta Biodiversity Monitoring Institute. We developed species abundance models using satellite data from Sentinel-1 and Sentinel-2 processed in Google Earth Engine. We identified covariates from both synthetic aperture radar and optical remote sensing that had strong predictive capacity for this wetland bird (AUC = 0.96). Approximately 1.5% of available wetland habitat in our northeast Alberta study area was predicted to be highly suitable for Yellow Rail.","PeriodicalId":48843,"journal":{"name":"Canadian Journal of Remote Sensing","volume":"48 1","pages":"37 - 54"},"PeriodicalIF":2.6,"publicationDate":"2022-01-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45286735","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-20DOI: 10.1080/07038992.2021.1997347
A. Moghtaderi, F. Moore, Hojatollah Ranjbar
Abstract In the last fifty years, satellite images have been used to map the Earth’s surface at a variety of scales. Two satellite multispectral sensors (Sentinel-2 MSI and ASTER) have great utility for lithological discrimination in areas of good rock exposures. This study was conducted in order to test the ability of these sensors to discriminate igneous and metamorphic lithologies in the Chadormalu paleocrater and evaluate the image types and processing methodologies. The MNF (Minimum Noise Fraction) transform, Mathematical Evaluation Method (MEM), Spectral Angle Mapper (SAM), Mixture Tuned Matched Filter (MTMF), and band ratios were performed on near and short wave infrared ASTER and Sentinel-2 bands. Comparison of the results from several methods demonstrates that the MEM method can detect lithological units with very low false detection and better matching with ground truth data. Moreover, this study indicates that the results produced by the MEM algorithm on Sentinel-2 MSI data are more accurate than the results produced with ASTER data in the same area. Therefore, the MEM algorithm seems to be well suited for image classification involving multispectral databases such as ASTER and Sentinel-2 images.
{"title":"Testing ASTER and Sentinel-2 MSI Images to Discriminate Igneous and Metamorphic Rock Units in the Chadormalu Paleocrater, Central Iran","authors":"A. Moghtaderi, F. Moore, Hojatollah Ranjbar","doi":"10.1080/07038992.2021.1997347","DOIUrl":"https://doi.org/10.1080/07038992.2021.1997347","url":null,"abstract":"Abstract In the last fifty years, satellite images have been used to map the Earth’s surface at a variety of scales. Two satellite multispectral sensors (Sentinel-2 MSI and ASTER) have great utility for lithological discrimination in areas of good rock exposures. This study was conducted in order to test the ability of these sensors to discriminate igneous and metamorphic lithologies in the Chadormalu paleocrater and evaluate the image types and processing methodologies. The MNF (Minimum Noise Fraction) transform, Mathematical Evaluation Method (MEM), Spectral Angle Mapper (SAM), Mixture Tuned Matched Filter (MTMF), and band ratios were performed on near and short wave infrared ASTER and Sentinel-2 bands. Comparison of the results from several methods demonstrates that the MEM method can detect lithological units with very low false detection and better matching with ground truth data. Moreover, this study indicates that the results produced by the MEM algorithm on Sentinel-2 MSI data are more accurate than the results produced with ASTER data in the same area. Therefore, the MEM algorithm seems to be well suited for image classification involving multispectral databases such as ASTER and Sentinel-2 images.","PeriodicalId":48843,"journal":{"name":"Canadian Journal of Remote Sensing","volume":"48 1","pages":"214 - 238"},"PeriodicalIF":2.6,"publicationDate":"2021-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43563285","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-13DOI: 10.1080/07038992.2021.2003701
M. Shokr, M. Dabboor, Mélanie Lacelle, Tom Zagon, B. Deschamps
Abstract Fully polarimetric (FP) SAR systems offer parameters that describe and quantify the scattering mechanisms from the surface cover. These are usually derived from decomposition of matrices derived from the original scattering matrix from observations at each pixel. Power from scattering mechanisms have potential for retrieval of sea ice information, which cannot be derived using traditional backscatter (magnitude or phase) measured by single- or dual-polarization SAR systems. This study investigates the potential of selected FP parameters that represent the power of three scattering mechanisms, in addition to the total power, in identifying ice types and surface features for operational use. Parameters were obtained from a set of 62 RADARSAT-2 Quad-pol data over Resolute Passage, central Arctic, during the period September-December 2017. A scattering-based color-composite scheme was developed. Analysis of the examined color images was supported by information from regional ice charts and SAR image interpretations from the Canadian Ice Service. Case studies are presented to demonstrate the potential of the proposed color-composite tool. Open water, new ice, multi-year ice and a few surface features including rafted, ridged and smooth/rough surfaces can be identified better in the color images. Physical interpretation of the relative power from the given scattering mechanisms is explained for the relevant ice types and surfaces.
{"title":"Observations from C-Band SAR Fully Polarimetric Parameters of Mobile Sea Ice Based on Radar Scattering Mechanisms to Support Operational Sea Ice Monitoring","authors":"M. Shokr, M. Dabboor, Mélanie Lacelle, Tom Zagon, B. Deschamps","doi":"10.1080/07038992.2021.2003701","DOIUrl":"https://doi.org/10.1080/07038992.2021.2003701","url":null,"abstract":"Abstract Fully polarimetric (FP) SAR systems offer parameters that describe and quantify the scattering mechanisms from the surface cover. These are usually derived from decomposition of matrices derived from the original scattering matrix from observations at each pixel. Power from scattering mechanisms have potential for retrieval of sea ice information, which cannot be derived using traditional backscatter (magnitude or phase) measured by single- or dual-polarization SAR systems. This study investigates the potential of selected FP parameters that represent the power of three scattering mechanisms, in addition to the total power, in identifying ice types and surface features for operational use. Parameters were obtained from a set of 62 RADARSAT-2 Quad-pol data over Resolute Passage, central Arctic, during the period September-December 2017. A scattering-based color-composite scheme was developed. Analysis of the examined color images was supported by information from regional ice charts and SAR image interpretations from the Canadian Ice Service. Case studies are presented to demonstrate the potential of the proposed color-composite tool. Open water, new ice, multi-year ice and a few surface features including rafted, ridged and smooth/rough surfaces can be identified better in the color images. Physical interpretation of the relative power from the given scattering mechanisms is explained for the relevant ice types and surfaces.","PeriodicalId":48843,"journal":{"name":"Canadian Journal of Remote Sensing","volume":"48 1","pages":"197 - 213"},"PeriodicalIF":2.6,"publicationDate":"2021-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42077336","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-12-07DOI: 10.1080/07038992.2021.2010523
Ahmed Ziane, A. Douaoui, I. Yahiaoui, M. Pulido, M. Larid, A. Gulakhmadov, Xi Chen
Abstract Since decades ago, the Lower Cheliff plain is under the continuous influence of soil salinization induced by mismanagement of the groundwater resources. The main purpose of this study was to estimate and map soil salinity using both Salinity Index (SI) and Artificial Neural Networks (ANN). In doing so, a total of 796 soil samples of Electrical Conductivity (EC, dS.m–1) measured in laboratory combined to spectral parameters data of Landsat-8 OLI, by applying a Salinity Index (SI) and used also to training the ANN model (80% of total data), the rest of the dataset (20%) was retained for validation with both methods. The results of applying an ANN estimator based on the reflectance values of three bands: green (B3), red (B4) and near-infrared (B5) as learning input neurons, proved their interest in the estimation of EC given a high determination coefficient (R2 = 0.80) between the values of simulated truth and ground, compared to the results obtained using only the SI method giving a moderate precision (R2 = 0.42). Regarding the soil salinity mapping, the two methods generated contrasting results, the SI estimates that 68.5% of the total area is affected by salinity (underestimation) meanwhile the ANN gave an estimation of 78.8%. In a conclusion, the estimation and mapping of soil salinity using the SI method has been upgraded significantly when ANN was involved.
{"title":"Upgrading the Salinity Index Estimation and Mapping Quality of Soil Salinity Using Artificial Neural Networks in the Lower-Cheliff Plain of Algeria in North Africa","authors":"Ahmed Ziane, A. Douaoui, I. Yahiaoui, M. Pulido, M. Larid, A. Gulakhmadov, Xi Chen","doi":"10.1080/07038992.2021.2010523","DOIUrl":"https://doi.org/10.1080/07038992.2021.2010523","url":null,"abstract":"Abstract Since decades ago, the Lower Cheliff plain is under the continuous influence of soil salinization induced by mismanagement of the groundwater resources. The main purpose of this study was to estimate and map soil salinity using both Salinity Index (SI) and Artificial Neural Networks (ANN). In doing so, a total of 796 soil samples of Electrical Conductivity (EC, dS.m–1) measured in laboratory combined to spectral parameters data of Landsat-8 OLI, by applying a Salinity Index (SI) and used also to training the ANN model (80% of total data), the rest of the dataset (20%) was retained for validation with both methods. The results of applying an ANN estimator based on the reflectance values of three bands: green (B3), red (B4) and near-infrared (B5) as learning input neurons, proved their interest in the estimation of EC given a high determination coefficient (R2 = 0.80) between the values of simulated truth and ground, compared to the results obtained using only the SI method giving a moderate precision (R2 = 0.42). Regarding the soil salinity mapping, the two methods generated contrasting results, the SI estimates that 68.5% of the total area is affected by salinity (underestimation) meanwhile the ANN gave an estimation of 78.8%. In a conclusion, the estimation and mapping of soil salinity using the SI method has been upgraded significantly when ANN was involved.","PeriodicalId":48843,"journal":{"name":"Canadian Journal of Remote Sensing","volume":"48 1","pages":"182 - 196"},"PeriodicalIF":2.6,"publicationDate":"2021-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47656583","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-17DOI: 10.1080/07038992.2021.1986810
Jinzhou Liu, Li Fang, Huifang Shen, Shudong Zhou
Abstract Rapid and accurate acquisition of glacier regional changes is of great significance to the study of glaciers. Among all satellite images, Synthetic Aperture Radar (SAR) data has a great advantage in monitoring the glaciers in harsh weather conditions. Conventionally, glacier boundaries are manually delineated on images. However, this is a time-consuming process, especially in the batch process of large-area data. In this paper, we propose a Multiscale Joint Deep Neural Network (MJ-DNN) for large-scale glaciers contour extraction using single-polarimetric SAR intensity images. Based on U-Net, the proposed method has been improved in three aspects. First, Atrous Separable Convolution is used instead of convolution with the down-sampling part. Second, we propose a multiscale joint convolution layer to obtain information at multiple scales. Third, we deepen the network with the residual connection structure for higher-level features. At the final layer, we optimize the network result with the conditional random field method. To validate our approach, we test it on three glaciers and we compare the segmentation results of four different methods in parallel. The results show that the intersection over the union of the proposed method is the most efficient.
{"title":"A Multiscale Joint Deep Neural Network for Glacier Contour Extraction","authors":"Jinzhou Liu, Li Fang, Huifang Shen, Shudong Zhou","doi":"10.1080/07038992.2021.1986810","DOIUrl":"https://doi.org/10.1080/07038992.2021.1986810","url":null,"abstract":"Abstract Rapid and accurate acquisition of glacier regional changes is of great significance to the study of glaciers. Among all satellite images, Synthetic Aperture Radar (SAR) data has a great advantage in monitoring the glaciers in harsh weather conditions. Conventionally, glacier boundaries are manually delineated on images. However, this is a time-consuming process, especially in the batch process of large-area data. In this paper, we propose a Multiscale Joint Deep Neural Network (MJ-DNN) for large-scale glaciers contour extraction using single-polarimetric SAR intensity images. Based on U-Net, the proposed method has been improved in three aspects. First, Atrous Separable Convolution is used instead of convolution with the down-sampling part. Second, we propose a multiscale joint convolution layer to obtain information at multiple scales. Third, we deepen the network with the residual connection structure for higher-level features. At the final layer, we optimize the network result with the conditional random field method. To validate our approach, we test it on three glaciers and we compare the segmentation results of four different methods in parallel. The results show that the intersection over the union of the proposed method is the most efficient.","PeriodicalId":48843,"journal":{"name":"Canadian Journal of Remote Sensing","volume":"48 1","pages":"93 - 106"},"PeriodicalIF":2.6,"publicationDate":"2021-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45440223","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-13DOI: 10.1080/07038992.2021.1967732
Aaron Judah, Baoxin Hu
Abstract Methodologies were developed to classify wetlands (Open Bog, Treed Bog, Open Fen, Treed Fen, and Swamps) from remotely sensed data using advanced classification algorithms through two hierarchical approaches. The data utilized included multispectral optical and thermal data (Landsat-5, and Landsat-8), radar imagery (Sentinel-1), and a digital elevation model. Goals were to determine the best way to combine imagery to classify wetlands through hierarchically based classification approaches to produce more accurate and efficient maps compared to standard classification. Algorithms used were Random Forest (RF), and Naïve Bayes. A hierarchically based RF classification methodology produced the most accurate classification result (91.94%). The hierarchically based approaches also improved classification accuracies for low-quality data, as defined through feature analysis, when compared to a nonhierarchical classifier. The hierarchical approaches also produced a significant increase in classification accuracy for the Naïve Bayes classifier versus the standard approach (∼12% increase) while not significantly increasing computation time – comparable in accuracy to the RF tests for around 20% the computational effort. Preselection of spectral bands, polarizations and other input parameters (Normalized Difference Vegetation Index, Normalized Difference Water Index, albedo, slope, etc.) using log-normal or RF variable importance analysis was very effective at identifying low-quality features and features which were of higher quality.
{"title":"The Integration of Multi-Source Remotely Sensed Data with Hierarchically Based Classification Approaches in Support of the Classification of Wetlands","authors":"Aaron Judah, Baoxin Hu","doi":"10.1080/07038992.2021.1967732","DOIUrl":"https://doi.org/10.1080/07038992.2021.1967732","url":null,"abstract":"Abstract Methodologies were developed to classify wetlands (Open Bog, Treed Bog, Open Fen, Treed Fen, and Swamps) from remotely sensed data using advanced classification algorithms through two hierarchical approaches. The data utilized included multispectral optical and thermal data (Landsat-5, and Landsat-8), radar imagery (Sentinel-1), and a digital elevation model. Goals were to determine the best way to combine imagery to classify wetlands through hierarchically based classification approaches to produce more accurate and efficient maps compared to standard classification. Algorithms used were Random Forest (RF), and Naïve Bayes. A hierarchically based RF classification methodology produced the most accurate classification result (91.94%). The hierarchically based approaches also improved classification accuracies for low-quality data, as defined through feature analysis, when compared to a nonhierarchical classifier. The hierarchical approaches also produced a significant increase in classification accuracy for the Naïve Bayes classifier versus the standard approach (∼12% increase) while not significantly increasing computation time – comparable in accuracy to the RF tests for around 20% the computational effort. Preselection of spectral bands, polarizations and other input parameters (Normalized Difference Vegetation Index, Normalized Difference Water Index, albedo, slope, etc.) using log-normal or RF variable importance analysis was very effective at identifying low-quality features and features which were of higher quality.","PeriodicalId":48843,"journal":{"name":"Canadian Journal of Remote Sensing","volume":"48 1","pages":"158 - 181"},"PeriodicalIF":2.6,"publicationDate":"2021-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48786257","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-11DOI: 10.1080/07038992.2021.1999797
C. Barnes, C. Hopkinson
Abstract The objective of this study is to evaluate the performance of an Airborne Laser Scanning (ALS) snow sampling strategy using two distinct flight paths within a mountainous watershed. Drivers of snow depth variability (canopy, elevation, topographic position index, aspect) were used to generate a classified snow accumulation unit (SAU) raster for the Westcastle watershed, Alberta (103 km2). A “Least Cost Path” (LCP) analysis and an “expert” three-transect selection (T3) were used to create two flight path scenarios that each sampled <18% of the watershed area and maximized the number of represented SAUs. Watershed “wall-to-wall” snow depth was predicted from the T3, LCP, and combined T3 + LCP sampling data using ESRI’s Forest Based Regression. The variance was ∼ 83% for each of the three FBR scenarios. However, validation of the watershed-wide observed versus FBR predicted snow depth at watershed-scale produced R2 = 0.72 and RMSE = 0.38 m for the combined T3 + LCP flight line and R 2 = 0.66 (RMSE = 0.43 m) for T3 alone. The LCP sampling did not perform as well (R 2 = 0.34, RMSE = 0.61 m), indicating grid cell-level SAU attributes need to be supplemented by latitudinal and longitudinal sampling that captures beyond grid cell-level hydro-climatological trends across the watershed. By flying sampling corridors, that capture land surface attributes representative of the spatial variability of snow depth, watershed-scale snow volumes can be predicted.
{"title":"Comparing Mountain Snowpack Depth Model Results from Different Airborne Laser Scanning Flight Path Samples","authors":"C. Barnes, C. Hopkinson","doi":"10.1080/07038992.2021.1999797","DOIUrl":"https://doi.org/10.1080/07038992.2021.1999797","url":null,"abstract":"Abstract The objective of this study is to evaluate the performance of an Airborne Laser Scanning (ALS) snow sampling strategy using two distinct flight paths within a mountainous watershed. Drivers of snow depth variability (canopy, elevation, topographic position index, aspect) were used to generate a classified snow accumulation unit (SAU) raster for the Westcastle watershed, Alberta (103 km2). A “Least Cost Path” (LCP) analysis and an “expert” three-transect selection (T3) were used to create two flight path scenarios that each sampled <18% of the watershed area and maximized the number of represented SAUs. Watershed “wall-to-wall” snow depth was predicted from the T3, LCP, and combined T3 + LCP sampling data using ESRI’s Forest Based Regression. The variance was ∼ 83% for each of the three FBR scenarios. However, validation of the watershed-wide observed versus FBR predicted snow depth at watershed-scale produced R2 = 0.72 and RMSE = 0.38 m for the combined T3 + LCP flight line and R 2 = 0.66 (RMSE = 0.43 m) for T3 alone. The LCP sampling did not perform as well (R 2 = 0.34, RMSE = 0.61 m), indicating grid cell-level SAU attributes need to be supplemented by latitudinal and longitudinal sampling that captures beyond grid cell-level hydro-climatological trends across the watershed. By flying sampling corridors, that capture land surface attributes representative of the spatial variability of snow depth, watershed-scale snow volumes can be predicted.","PeriodicalId":48843,"journal":{"name":"Canadian Journal of Remote Sensing","volume":"48 1","pages":"81 - 92"},"PeriodicalIF":2.6,"publicationDate":"2021-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43749417","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-10-21DOI: 10.1080/07038992.2021.1971960
G. J. D. S. Carmo, R. Castoldi, G. D. Martins, A. C. Jacinto, N. D. Tebaldi, H. Charlo, R. Zampiróli
Abstract This study aimed to detect soft rot caused by Pectobacterium carotovorum subsp. carotovorum in lettuce using images obtained by multispectral sensors mounted on an unmanned aerial vehicle (UAV). A secondary objective was to identify the best sensor and determine the optimal stage after inoculation to detect infected plants. In the field, soft rot lesions and the agronomic traits of lettuce plants inoculated or not with the bacteria were assessed on different days after inoculation (DAI). Classifications were made using the Support Vector Machine (SVM) and Naive Bayes (NB) algorithms to analyze data groups consisting of spectral bands, vegetation indices and a combination of bands and indices obtained from a conventional visible camera and Mapir Survey3W multispectral camera, as well as agronomic parameters. The results confirmed the possibility of pre-symptomatic detection of P. carotovorum subsp. carotovorum in lettuce at the canopy level. With respect to identifying healthy and infected lettuce plants by supervised classification, the best results were obtained at 4 and 8 DAI, especially when using the subsets derived from the Mapir Survey3W camera (RGN sensor), for both classifiers. The subsets obtained with the conventional visible sensor (RGB sensor) produced the best results at 20 and 24 DAI.
{"title":"Detection of Lesions in Lettuce Caused by Pectobacterium carotovorum Subsp. carotovorum by Supervised Classification Using Multispectral Images","authors":"G. J. D. S. Carmo, R. Castoldi, G. D. Martins, A. C. Jacinto, N. D. Tebaldi, H. Charlo, R. Zampiróli","doi":"10.1080/07038992.2021.1971960","DOIUrl":"https://doi.org/10.1080/07038992.2021.1971960","url":null,"abstract":"Abstract This study aimed to detect soft rot caused by Pectobacterium carotovorum subsp. carotovorum in lettuce using images obtained by multispectral sensors mounted on an unmanned aerial vehicle (UAV). A secondary objective was to identify the best sensor and determine the optimal stage after inoculation to detect infected plants. In the field, soft rot lesions and the agronomic traits of lettuce plants inoculated or not with the bacteria were assessed on different days after inoculation (DAI). Classifications were made using the Support Vector Machine (SVM) and Naive Bayes (NB) algorithms to analyze data groups consisting of spectral bands, vegetation indices and a combination of bands and indices obtained from a conventional visible camera and Mapir Survey3W multispectral camera, as well as agronomic parameters. The results confirmed the possibility of pre-symptomatic detection of P. carotovorum subsp. carotovorum in lettuce at the canopy level. With respect to identifying healthy and infected lettuce plants by supervised classification, the best results were obtained at 4 and 8 DAI, especially when using the subsets derived from the Mapir Survey3W camera (RGN sensor), for both classifiers. The subsets obtained with the conventional visible sensor (RGB sensor) produced the best results at 20 and 24 DAI.","PeriodicalId":48843,"journal":{"name":"Canadian Journal of Remote Sensing","volume":"48 1","pages":"144 - 157"},"PeriodicalIF":2.6,"publicationDate":"2021-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49546221","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：