Pub Date : 2023-08-15DOI: 10.5194/isprs-archives-xlviii-m-1-2023-587-2023
E. Sevgen, S. Abdikan
Abstract. Point-wise classification of 3D point clouds is a challenging task in point cloud processing, whereas, in particular, its application to high-density point clouds needs special attention because a large number of point clouds affect computational efficiency negatively. Although deep learning based models have been gaining popularity in recent years and have reached state-of-the-art results in accuracy for point-wise classification, their requirements of the high number of training samples and computational resources make those models inefficient for high-density 3D point clouds. However, traditional machine learning classifiers require less training samples, so they are capable of reducing computational requirements, even considering the latest machine learning classifiers, particularly in ensemble learning of gradient boosting machines, the results can compete with deep learning models. In this study, we are studying the point-wise classification of high-density UAV LiDAR data and focusing on efficient feature extraction and a recent state-of-the-art gradient boosting machine learning classifier, LightGBM. Our proposed framework includes the following steps: at first, we are using point cloud sampling for creating sub-sampled point clouds, then we are calculating the features based on those scales implemented on GPU. Finally, we are using the LightGBM classifier for training and testing. For the evaluation of our framework, we used a publicly available benchmark dataset, Hessigheim 3D. According to the results, we achieved an overall accuracy of 87.59% and an average F1 score of 75.92%. Our framework has promising results and scores closer to deep learning models. However, more distinctive features are required to obtain more accurate results.
{"title":"POINT-WISE CLASSIFICATION OF HIGH-DENSITY UAV-LIDAR DATA USING GRADIENT BOOSTING MACHINES","authors":"E. Sevgen, S. Abdikan","doi":"10.5194/isprs-archives-xlviii-m-1-2023-587-2023","DOIUrl":"https://doi.org/10.5194/isprs-archives-xlviii-m-1-2023-587-2023","url":null,"abstract":"Abstract. Point-wise classification of 3D point clouds is a challenging task in point cloud processing, whereas, in particular, its application to high-density point clouds needs special attention because a large number of point clouds affect computational efficiency negatively. Although deep learning based models have been gaining popularity in recent years and have reached state-of-the-art results in accuracy for point-wise classification, their requirements of the high number of training samples and computational resources make those models inefficient for high-density 3D point clouds. However, traditional machine learning classifiers require less training samples, so they are capable of reducing computational requirements, even considering the latest machine learning classifiers, particularly in ensemble learning of gradient boosting machines, the results can compete with deep learning models. In this study, we are studying the point-wise classification of high-density UAV LiDAR data and focusing on efficient feature extraction and a recent state-of-the-art gradient boosting machine learning classifier, LightGBM. Our proposed framework includes the following steps: at first, we are using point cloud sampling for creating sub-sampled point clouds, then we are calculating the features based on those scales implemented on GPU. Finally, we are using the LightGBM classifier for training and testing. For the evaluation of our framework, we used a publicly available benchmark dataset, Hessigheim 3D. According to the results, we achieved an overall accuracy of 87.59% and an average F1 score of 75.92%. Our framework has promising results and scores closer to deep learning models. However, more distinctive features are required to obtain more accurate results.\u0000","PeriodicalId":30634,"journal":{"name":"The International Archives of the Photogrammetry Remote Sensing and Spatial Information Sciences","volume":"3 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41274383","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-15DOI: 10.5194/isprs-archives-xlviii-m-1-2023-503-2023
Y. Dogan, A. O. Ok
Abstract. Extracting ground points from 3D point clouds is important for sustainable development goals, infrastructure planning, disaster management, and more. However, the irregularity and complexity of the data make it challenging. Deep learning techniques, particularly end-to-end and non-end-to-end approaches, have shown promise for 3D point cloud segmentation and classification, but both require a comprehensive understanding of the features and their relationship to the problem. This paper presents a study on the filtering of 3D LiDAR point clouds into ground and non-ground points using a non-end-to-end deep learning approach. The aim of this research is to investigate the effectiveness of utilizing geometric features and a binary classifier-based deep learning model in accurately classifying point clouds. The publicly available ACT benchmark datasets were employed for training, validation, and testing purposes. The study utilized a k-fold cross-validation method to address the limited availability of training data. The results demonstrated highly satisfactory performance, with validation averages reaching 96.83% for the divided Dataset-1 and an accuracy of 97% for the test set. Furthermore, an independent dataset, Dataset-2, was used to evaluate the generalizability of the trained model, achieving an accuracy of 93%. These findings highlight the potential of the proposed non-end-to-end approach to filtering point cloud data and its applicability in various domains such as DEM and DTM production, city modeling, urban planning, and disaster management. Moreover, this study emphasizes the need for accurate data to achieve sustainable development goals, positioning the proposed approach as a viable option in various studies.
{"title":"A FEATURE-BASED DEEP LEARNING APPROACH FOR THE EXTRACTION OF GROUND POINTS FROM 3D POINT CLOUDS","authors":"Y. Dogan, A. O. Ok","doi":"10.5194/isprs-archives-xlviii-m-1-2023-503-2023","DOIUrl":"https://doi.org/10.5194/isprs-archives-xlviii-m-1-2023-503-2023","url":null,"abstract":"Abstract. Extracting ground points from 3D point clouds is important for sustainable development goals, infrastructure planning, disaster management, and more. However, the irregularity and complexity of the data make it challenging. Deep learning techniques, particularly end-to-end and non-end-to-end approaches, have shown promise for 3D point cloud segmentation and classification, but both require a comprehensive understanding of the features and their relationship to the problem. This paper presents a study on the filtering of 3D LiDAR point clouds into ground and non-ground points using a non-end-to-end deep learning approach. The aim of this research is to investigate the effectiveness of utilizing geometric features and a binary classifier-based deep learning model in accurately classifying point clouds. The publicly available ACT benchmark datasets were employed for training, validation, and testing purposes. The study utilized a k-fold cross-validation method to address the limited availability of training data. The results demonstrated highly satisfactory performance, with validation averages reaching 96.83% for the divided Dataset-1 and an accuracy of 97% for the test set. Furthermore, an independent dataset, Dataset-2, was used to evaluate the generalizability of the trained model, achieving an accuracy of 93%. These findings highlight the potential of the proposed non-end-to-end approach to filtering point cloud data and its applicability in various domains such as DEM and DTM production, city modeling, urban planning, and disaster management. Moreover, this study emphasizes the need for accurate data to achieve sustainable development goals, positioning the proposed approach as a viable option in various studies.\u0000","PeriodicalId":30634,"journal":{"name":"The International Archives of the Photogrammetry Remote Sensing and Spatial Information Sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41684538","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-15DOI: 10.5194/isprs-archives-xlviii-m-1-2023-553-2023
G. Kurt, M. Külahlıoğlu, S. Berberoglu
Abstract. One of the most significant challenges in urban areas, where the process of rapid urban expansion takes place, is the loss of agricultural lands and natural habitats. The conversion of these areas into residential and commercial zones leads to a decline in urban biodiversity and the progressive loss of vital habitat areas. Analyzing habitat connectivity and conducting landscape measurements provide valuable insights for the development of land use and management strategies, enhancing our understanding of the spatial structure of the landscape, and directing conservation efforts. Incorporating measures such as green corridors and landscape connection networks into urban planning management becomes crucial in order to mitigate the adverse effects of habitat fragmentation and enhance ecosystem resilience within cities. Remote sensing techniques offer opportunities to create habitat connectivity models that enable the quantitative and qualitative identification of fragmented habitat patches. These models serve as tools to evaluate the effectiveness of conservation measures and monitor the potential impacts of future land use changes on habitat networks. Within this context, an optimized approach to habitat connectivity is presented, aiming to contribute to landscape planning and ecological-based studies in a city with undergoing rapid urbanization like Adana. By identifying degraded areas and introducing new habitat patches, a significant improvement in the connectivity of the habitat network has been observed. The findings indicated that the addition of new habitat patches to degraded areas can substantially enhance the city's overall habitat connectivity.
{"title":"OPTIMIZED ECOLOGICAL NETWORK APPROACH OF HIGHLY URBANIZED CITIES: THE CASE OF ADANA CITY","authors":"G. Kurt, M. Külahlıoğlu, S. Berberoglu","doi":"10.5194/isprs-archives-xlviii-m-1-2023-553-2023","DOIUrl":"https://doi.org/10.5194/isprs-archives-xlviii-m-1-2023-553-2023","url":null,"abstract":"Abstract. One of the most significant challenges in urban areas, where the process of rapid urban expansion takes place, is the loss of agricultural lands and natural habitats. The conversion of these areas into residential and commercial zones leads to a decline in urban biodiversity and the progressive loss of vital habitat areas. Analyzing habitat connectivity and conducting landscape measurements provide valuable insights for the development of land use and management strategies, enhancing our understanding of the spatial structure of the landscape, and directing conservation efforts. Incorporating measures such as green corridors and landscape connection networks into urban planning management becomes crucial in order to mitigate the adverse effects of habitat fragmentation and enhance ecosystem resilience within cities. Remote sensing techniques offer opportunities to create habitat connectivity models that enable the quantitative and qualitative identification of fragmented habitat patches. These models serve as tools to evaluate the effectiveness of conservation measures and monitor the potential impacts of future land use changes on habitat networks. Within this context, an optimized approach to habitat connectivity is presented, aiming to contribute to landscape planning and ecological-based studies in a city with undergoing rapid urbanization like Adana. By identifying degraded areas and introducing new habitat patches, a significant improvement in the connectivity of the habitat network has been observed. The findings indicated that the addition of new habitat patches to degraded areas can substantially enhance the city's overall habitat connectivity.\u0000","PeriodicalId":30634,"journal":{"name":"The International Archives of the Photogrammetry Remote Sensing and Spatial Information Sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48415941","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-15DOI: 10.5194/isprs-archives-xlviii-m-1-2023-537-2023
Ç. H. Karaman, Z. Akyurek
Abstract. In this study, we evaluate the performance of several reanalyses and satellite-based products of near-surface air temperature and precipitation to determine the best product in estimating daily and monthly variables across the complex terrain of Turkey. Each product’s performance was evaluated using 1120 ground-based gauge stations from 2015 to 2019, covering a range of complex topography with different climate classes according to the Köppen-Geiger classification scheme and land surface types according to the Moderate Resolution Imaging Spectroradiometer (MODIS). Furthermore, various traditional and more advanced machine learning downscaling algorithms were applied to improve the spatial resolution of the products. We used distance-based interpolation, classical Random Forest, and more innovative Random Forest Spatial Interpolation (RFSI) algorithms. We also investigated several satellite-based covariates as a proxy to downscale the precipitation and near-surface air temperature, including MODIS Land Surface Temperature, Vegetation Index (NDVI and EVI), Cloud Properties (Cloud Optical Properties, Cloud Effective Radius, Cloud Water Path), and topography-related features. The agreement between the ground observations and the different products, as well as the downscaled temperature products, was examined using a range of commonly employed measures. The results showed that AgERA5 was the best-performing product for air temperature estimation, while MSWEP V2.2 was superior for precipitation estimation. Spatial downscaling using bicubic interpolation improved air temperature product performance, and the Random Forest (RF) machine learning algorithm outperformed all other methods in certain seasons. The study suggests that combining ground-based measurements, precipitation products, and features related to topography can substantially improve the representation of spatiotemporal precipitation distribution in data-scarce regions.
{"title":"IMPROVING THE ACCURACY OF SATELLITE-BASED NEAR SURFACE AIR TEMPERATURE AND PRECIPITATION PRODUCTS","authors":"Ç. H. Karaman, Z. Akyurek","doi":"10.5194/isprs-archives-xlviii-m-1-2023-537-2023","DOIUrl":"https://doi.org/10.5194/isprs-archives-xlviii-m-1-2023-537-2023","url":null,"abstract":"Abstract. In this study, we evaluate the performance of several reanalyses and satellite-based products of near-surface air temperature and precipitation to determine the best product in estimating daily and monthly variables across the complex terrain of Turkey. Each product’s performance was evaluated using 1120 ground-based gauge stations from 2015 to 2019, covering a range of complex topography with different climate classes according to the Köppen-Geiger classification scheme and land surface types according to the Moderate Resolution Imaging Spectroradiometer (MODIS). Furthermore, various traditional and more advanced machine learning downscaling algorithms were applied to improve the spatial resolution of the products. We used distance-based interpolation, classical Random Forest, and more innovative Random Forest Spatial Interpolation (RFSI) algorithms. We also investigated several satellite-based covariates as a proxy to downscale the precipitation and near-surface air temperature, including MODIS Land Surface Temperature, Vegetation Index (NDVI and EVI), Cloud Properties (Cloud Optical Properties, Cloud Effective Radius, Cloud Water Path), and topography-related features. The agreement between the ground observations and the different products, as well as the downscaled temperature products, was examined using a range of commonly employed measures. The results showed that AgERA5 was the best-performing product for air temperature estimation, while MSWEP V2.2 was superior for precipitation estimation. Spatial downscaling using bicubic interpolation improved air temperature product performance, and the Random Forest (RF) machine learning algorithm outperformed all other methods in certain seasons. The study suggests that combining ground-based measurements, precipitation products, and features related to topography can substantially improve the representation of spatiotemporal precipitation distribution in data-scarce regions.\u0000","PeriodicalId":30634,"journal":{"name":"The International Archives of the Photogrammetry Remote Sensing and Spatial Information Sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43389880","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-15DOI: 10.5194/isprs-archives-xlviii-m-1-2023-529-2023
G. Karakas, S. Kocaman, C. Gokceoglu
Abstract. Recent Kahramanmaras earthquakes (Mw 7.7 and 7.6) occurred on 6 February 2023 have shown the importance of site selection for settlements and infrastructure considering the fact that multiple hazards may affect the same area and even interact with each other. The Kahramanmaras earthquakes triggered several landslides, which also increased the level of destruction. Here, we implemented a multi-hazard susceptibility assessment approach for Tut town of Golbasi, Adiyaman and its surroundings. Over 600 landslides were triggered in the area by the earthquakes. In addition, the region is prone to flooding and a devastating one occurred on March 15, 2023 after heavy rains. In this study, we employed co-seismic landslide inventory for landslide susceptibility assessment with random forest. Regarding flood susceptibility, a modified analytical hierarchical process was utilized based on expert opinion on factor importance. The earthquake hazard probability distribution was obtained from a distance-based interpolation of Arias intensity values. We utilized Mamdani Fuzzy Inference System for producing a multi-hazard susceptibility map from univariate maps of earthquake, landslide and flood. The result shows that the selected methods for each type of susceptibility map was suitable and the output of the study can be utilized for the site selection in Tut region, which is a crucial subject due to the need of new construction sites after the earthquakes.
{"title":"MULTI-HAZARD SUSCEPTIBILITY ASSESSMENT WITH HYBRID MACHINE LEARNING METHODS FOR TUT REGION (ADIYAMAN, TURKIYE)","authors":"G. Karakas, S. Kocaman, C. Gokceoglu","doi":"10.5194/isprs-archives-xlviii-m-1-2023-529-2023","DOIUrl":"https://doi.org/10.5194/isprs-archives-xlviii-m-1-2023-529-2023","url":null,"abstract":"Abstract. Recent Kahramanmaras earthquakes (Mw 7.7 and 7.6) occurred on 6 February 2023 have shown the importance of site selection for settlements and infrastructure considering the fact that multiple hazards may affect the same area and even interact with each other. The Kahramanmaras earthquakes triggered several landslides, which also increased the level of destruction. Here, we implemented a multi-hazard susceptibility assessment approach for Tut town of Golbasi, Adiyaman and its surroundings. Over 600 landslides were triggered in the area by the earthquakes. In addition, the region is prone to flooding and a devastating one occurred on March 15, 2023 after heavy rains. In this study, we employed co-seismic landslide inventory for landslide susceptibility assessment with random forest. Regarding flood susceptibility, a modified analytical hierarchical process was utilized based on expert opinion on factor importance. The earthquake hazard probability distribution was obtained from a distance-based interpolation of Arias intensity values. We utilized Mamdani Fuzzy Inference System for producing a multi-hazard susceptibility map from univariate maps of earthquake, landslide and flood. The result shows that the selected methods for each type of susceptibility map was suitable and the output of the study can be utilized for the site selection in Tut region, which is a crucial subject due to the need of new construction sites after the earthquakes.\u0000","PeriodicalId":30634,"journal":{"name":"The International Archives of the Photogrammetry Remote Sensing and Spatial Information Sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42653961","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-15DOI: 10.5194/isprs-archives-xlviii-m-1-2023-497-2023
S. Coşkun, Ç. Bayık, S. Abdikan, T. Gorum, F. Balik Sanli
Abstract. Landslides are prominent natural events with high destructive power. Since they affect large areas, it is important to monitor the areas they cover and analyse their movement. Remote sensing data and image processing techniques have been used to monitor landslides in different areas. Synthetic aperture radar (SAR) data, particularly with the Interferometric SAR (InSAR) method, is used to determine the velocity vector of the surface motion. This study aims to detect the landslide movements in Samsun, located in the north of Turkey, using persistent scattering InSAR method. Archived Copernicus Sentinel-1 satellite images taken between 2017 and 2022 were used in both descending and ascending directions. The results revealed surface movements in the direction of the line of sight, ranging between −6 and 6 mm/year in the study area. Persistent Scatterer (PS) points were identified mainly in human structures such as roads, coasts, ports, and golf courses, especially in settlements. While some regions exhibited similar movements in both descending and ascending results, opposite movements were observed in some regions. The results produced in both descending and ascending directions were used together and decomposed into horizontal and vertical deformation components. It was observed that the western coastal part experienced approximately 4.5 cm/year vertical deformation, while the central part there is more significant horizontal deformation, reaching up to approximately 6 cm/year.
{"title":"MONITORING THE SLOWLY DEVELOPING LANDSLIDE WITH THE INSAR TECHNIQUE IN SAMSUN PROVINCE, NORTHERN TURKEY","authors":"S. Coşkun, Ç. Bayık, S. Abdikan, T. Gorum, F. Balik Sanli","doi":"10.5194/isprs-archives-xlviii-m-1-2023-497-2023","DOIUrl":"https://doi.org/10.5194/isprs-archives-xlviii-m-1-2023-497-2023","url":null,"abstract":"Abstract. Landslides are prominent natural events with high destructive power. Since they affect large areas, it is important to monitor the areas they cover and analyse their movement. Remote sensing data and image processing techniques have been used to monitor landslides in different areas. Synthetic aperture radar (SAR) data, particularly with the Interferometric SAR (InSAR) method, is used to determine the velocity vector of the surface motion. This study aims to detect the landslide movements in Samsun, located in the north of Turkey, using persistent scattering InSAR method. Archived Copernicus Sentinel-1 satellite images taken between 2017 and 2022 were used in both descending and ascending directions. The results revealed surface movements in the direction of the line of sight, ranging between −6 and 6 mm/year in the study area. Persistent Scatterer (PS) points were identified mainly in human structures such as roads, coasts, ports, and golf courses, especially in settlements. While some regions exhibited similar movements in both descending and ascending results, opposite movements were observed in some regions. The results produced in both descending and ascending directions were used together and decomposed into horizontal and vertical deformation components. It was observed that the western coastal part experienced approximately 4.5 cm/year vertical deformation, while the central part there is more significant horizontal deformation, reaching up to approximately 6 cm/year.\u0000","PeriodicalId":30634,"journal":{"name":"The International Archives of the Photogrammetry Remote Sensing and Spatial Information Sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48322335","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-15DOI: 10.5194/isprs-archives-xlviii-m-1-2023-521-2023
S. Kara, B. Maden, B. Ercan, F. Sunar, T. Aysal, O. Saglam
Abstract. Remote sensing technology plays a crucial role in detecting and monitoring environmental issues, offering the ability to monitor large areas, diagnose problems early, and facilitate accurate interventions. By integrating in-situ data with qualitative measurements obtained from satellite images, comprehensive insights can be obtained, and statistical inferences can be established. This study focuses on analyzing the damages caused by beet webworm moths (Loxostege sticticalis) in sunflower fields located in the Ortaca neighborhood of Tekirdağ province in Thrace region, utilizing Sentinel-2 satellite images and in-situ data collected from the sunflower fields in Ortaca. The relationship between different spectral indices, such as the Enhanced Vegetation Index, Chlorophyll Index Green, and spectral transformation techniques like Tasseled Cap Greenness, derived from Sentinel-2 satellite images, and the observed damage rates in various sunflower fields' in-situ data was investigated. The results revealed a negative correlation between the variables, highlighting EVI as the most effective indicator of damage among the plant indices. Leveraging these findings, a damage map was generated using EVI, enabling visual interpretation of the damage status in other sunflower fields within the study area. These findings offer valuable insights into the impact of pests on sunflower crops, despite the accuracy evaluation results falling below the desired level, with an overall accuracy of 75% and a Kappa accuracy of 65%, attributed to the limited availability of in-situ data.
摘要遥感技术在发现和监测环境问题方面发挥着至关重要的作用，提供了监测大面积、早期诊断问题和促进准确干预的能力。通过将现场数据与卫星图像的定性测量相结合，可以获得全面的见解，并建立统计推断。本研究利用Sentinel-2卫星图像和Ortaca向日葵田现场数据，对色雷斯地区泰克达尔省Ortaca地区向日葵田甜菜网虫(Loxostege sticalis)的危害进行了分析。研究了Sentinel-2卫星影像中不同光谱指数(Enhanced Vegetation Index、叶绿素Index Green、Tasseled Cap Greenness等光谱变换技术)与不同向日葵田现场数据的毁伤率之间的关系。结果表明，各变量之间呈负相关关系，EVI是最有效的植物损伤指标。利用这些发现，利用EVI生成了损害图，可以直观地解释研究区域内其他向日葵田的损害状况。这些发现为害虫对向日葵作物的影响提供了有价值的见解，尽管准确性评估结果低于预期水平，由于原位数据的有限可用性，总体精度为75%，Kappa精度为65%。
{"title":"ASSESSING THE IMPACT OF BEET WEBWORM MOTHS ON SUNFLOWER FIELDS USING MULTITEMPORAL SENTINEL-2 SATELLITE IMAGERY AND VEGETATION INDICES","authors":"S. Kara, B. Maden, B. Ercan, F. Sunar, T. Aysal, O. Saglam","doi":"10.5194/isprs-archives-xlviii-m-1-2023-521-2023","DOIUrl":"https://doi.org/10.5194/isprs-archives-xlviii-m-1-2023-521-2023","url":null,"abstract":"Abstract. Remote sensing technology plays a crucial role in detecting and monitoring environmental issues, offering the ability to monitor large areas, diagnose problems early, and facilitate accurate interventions. By integrating in-situ data with qualitative measurements obtained from satellite images, comprehensive insights can be obtained, and statistical inferences can be established. This study focuses on analyzing the damages caused by beet webworm moths (Loxostege sticticalis) in sunflower fields located in the Ortaca neighborhood of Tekirdağ province in Thrace region, utilizing Sentinel-2 satellite images and in-situ data collected from the sunflower fields in Ortaca. The relationship between different spectral indices, such as the Enhanced Vegetation Index, Chlorophyll Index Green, and spectral transformation techniques like Tasseled Cap Greenness, derived from Sentinel-2 satellite images, and the observed damage rates in various sunflower fields' in-situ data was investigated. The results revealed a negative correlation between the variables, highlighting EVI as the most effective indicator of damage among the plant indices. Leveraging these findings, a damage map was generated using EVI, enabling visual interpretation of the damage status in other sunflower fields within the study area. These findings offer valuable insights into the impact of pests on sunflower crops, despite the accuracy evaluation results falling below the desired level, with an overall accuracy of 75% and a Kappa accuracy of 65%, attributed to the limited availability of in-situ data.\u0000","PeriodicalId":30634,"journal":{"name":"The International Archives of the Photogrammetry Remote Sensing and Spatial Information Sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47226172","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-15DOI: 10.5194/isprs-archives-xlviii-m-1-2023-559-2023
A. Polat, O. Akcay
Abstract. Atmospheric drought due to meteorological events occurring out of seasonal norms, and consequent droughts in agriculture and wetlands cause great damage to the ecological balance. The initial effects of this situation appear on a local scale, while the aftereffects, which last for years, appear on a global scale. Monitoring and detecting drought with remote sensing technologies can contribute to the management of water resources and forest areas and enable many measures to be taken to reduce the effects of drought. Within the scope of this study, a system that automatically performs the extraction of different drought parameters depending on years has been developed. Işıklı Lake was selected as the study area and the change of water areas over the years has been extracted from satellite images. With the system developed on the Google Earth Engine platform, different parameters were analyzed over a 13-year period and their consistency was tested. As a result, it is seen that the water areas in the lake decreased by 30% between 2010 and 2022. Likewise, the systematic decrease in the parameters, especially in 2015 and afterward, indicates the drought in the region. With the proposed automatic system, it is thought that early precautions can be taken for drought scenarios that may occur in larger-scale regions.
{"title":"AUTOMATIC EXTRACTION OF SURFACE DYNAMICS USING GOOGLE EARTH ENGINE FOR UNDERSTANDING DROUGHT PHENOMENON","authors":"A. Polat, O. Akcay","doi":"10.5194/isprs-archives-xlviii-m-1-2023-559-2023","DOIUrl":"https://doi.org/10.5194/isprs-archives-xlviii-m-1-2023-559-2023","url":null,"abstract":"Abstract. Atmospheric drought due to meteorological events occurring out of seasonal norms, and consequent droughts in agriculture and wetlands cause great damage to the ecological balance. The initial effects of this situation appear on a local scale, while the aftereffects, which last for years, appear on a global scale. Monitoring and detecting drought with remote sensing technologies can contribute to the management of water resources and forest areas and enable many measures to be taken to reduce the effects of drought. Within the scope of this study, a system that automatically performs the extraction of different drought parameters depending on years has been developed. Işıklı Lake was selected as the study area and the change of water areas over the years has been extracted from satellite images. With the system developed on the Google Earth Engine platform, different parameters were analyzed over a 13-year period and their consistency was tested. As a result, it is seen that the water areas in the lake decreased by 30% between 2010 and 2022. Likewise, the systematic decrease in the parameters, especially in 2015 and afterward, indicates the drought in the region. With the proposed automatic system, it is thought that early precautions can be taken for drought scenarios that may occur in larger-scale regions.\u0000","PeriodicalId":30634,"journal":{"name":"The International Archives of the Photogrammetry Remote Sensing and Spatial Information Sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46592108","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-15DOI: 10.5194/isprs-archives-xlviii-m-1-2023-609-2023
I. Yalcin, R. Can, S. Kocaman, C. Gokceoglu
Abstract. Rock mass discontinuity and orientation are among the important rock mass features. They are conventionally determined with scan-line surveys by engineering geologists in field, which can be difficult or impossible depending on site accessibility. Photogrammetry and computer vision techniques can aid to automatically perform these measurements, although variations in size, shape and appearance of rock masses make the task challenging. Here we propose an automated approach for the detection of rock mass discontinuities using deep learning and photogrammetric image processing methods. Two deep convolutional neural network (DCNNs) were implemented for this purpose and applied to basalts in Kizilcahamam Guvem Geosite near Ankara, Türkiye. Red-green-blue (RGB) band images of the site were taken from an off-the-shelf camera with 1.7 mm resolution and a 3D digital surface model and orthophotos were produced by using photogrammetric software. The discontinuities were delineated manually on the orthophoto and converted to masks. The first DCNN model was based on the open-source crack dataset consisting of a total of 11,298 road and pavement images, which were used to train the Resnet-18 model (Model-1). The second model (Model-2) was based on fine-tuning of Model-1 using the study data from Kizilcahamam. After fine-tuning, Model-2 was able to achieve high performance with a Jaccard Score of 88% on the test data. The results show high potential of the methodology for transfer learning with fine-tuning of a small amount of data that can be applied to other sites and rock mass types as well.
{"title":"ROCK MASS DISCONTINUITY DETERMINATION WITH TRANSFER LEARNING","authors":"I. Yalcin, R. Can, S. Kocaman, C. Gokceoglu","doi":"10.5194/isprs-archives-xlviii-m-1-2023-609-2023","DOIUrl":"https://doi.org/10.5194/isprs-archives-xlviii-m-1-2023-609-2023","url":null,"abstract":"Abstract. Rock mass discontinuity and orientation are among the important rock mass features. They are conventionally determined with scan-line surveys by engineering geologists in field, which can be difficult or impossible depending on site accessibility. Photogrammetry and computer vision techniques can aid to automatically perform these measurements, although variations in size, shape and appearance of rock masses make the task challenging. Here we propose an automated approach for the detection of rock mass discontinuities using deep learning and photogrammetric image processing methods. Two deep convolutional neural network (DCNNs) were implemented for this purpose and applied to basalts in Kizilcahamam Guvem Geosite near Ankara, Türkiye. Red-green-blue (RGB) band images of the site were taken from an off-the-shelf camera with 1.7 mm resolution and a 3D digital surface model and orthophotos were produced by using photogrammetric software. The discontinuities were delineated manually on the orthophoto and converted to masks. The first DCNN model was based on the open-source crack dataset consisting of a total of 11,298 road and pavement images, which were used to train the Resnet-18 model (Model-1). The second model (Model-2) was based on fine-tuning of Model-1 using the study data from Kizilcahamam. After fine-tuning, Model-2 was able to achieve high performance with a Jaccard Score of 88% on the test data. The results show high potential of the methodology for transfer learning with fine-tuning of a small amount of data that can be applied to other sites and rock mass types as well.\u0000","PeriodicalId":30634,"journal":{"name":"The International Archives of the Photogrammetry Remote Sensing and Spatial Information Sciences","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"70624662","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-08-15DOI: 10.5194/isprs-archives-xlviii-m-1-2023-455-2023
D. Alkan, L. Karasaka
Abstract. Fires damage nature and living beings. Detection of this damage is important for future. In this study, it was aimed to determine burned areas. For this purpose, Landsat-8 images and U-Net model were used. Python language was preferred. Band combinations 7,5,4; 5,3,7; 5,4,3; 4,3,2; 4,3,2,5 and 2,3,4,5,6,7 have been tried. Train and test processes were carried out separately for each band combination. After the train and test processes were completed, a probability result consisting of values between 0-1 was obtained. Then, a threshold value was used. Thus, binary results consisting of 0 and 1 values were obtained. Three different values were preferred for the threshold: 0.1, 0.5 and 0.9. Thus, the effect of threshold value selection on the test results was examined. The prediction results were evaluated using the masks. For this, general accuracy, recall, precision, F1-score and Jaccard score metrics were used. Recall, precision, and F1-score values were calculated for both burned areas and unburned areas. In addition, minimum, maximum, mean, and standard deviation values were calculated for each metric. When the results are examined, it is seen that the model gives better results when the threshold value is 0.1 and 0.5. Among the band combinations, it is seen that the 7,5,4 combination gave better results than the others. For this band combination, the highest mean accuracy is 0.9743 with the 0.5 threshold value. For this threshold mean recall, mean precision and mean F1-score for burned areas are 0.7203, 0.8411 and 0.7601, respectively. And Jaccard score is 0.6328.
{"title":"SEGMENTATION OF LANDSAT-8 IMAGES FOR BURNED AREA DETECTION WITH DEEP LEARNING","authors":"D. Alkan, L. Karasaka","doi":"10.5194/isprs-archives-xlviii-m-1-2023-455-2023","DOIUrl":"https://doi.org/10.5194/isprs-archives-xlviii-m-1-2023-455-2023","url":null,"abstract":"Abstract. Fires damage nature and living beings. Detection of this damage is important for future. In this study, it was aimed to determine burned areas. For this purpose, Landsat-8 images and U-Net model were used. Python language was preferred. Band combinations 7,5,4; 5,3,7; 5,4,3; 4,3,2; 4,3,2,5 and 2,3,4,5,6,7 have been tried. Train and test processes were carried out separately for each band combination. After the train and test processes were completed, a probability result consisting of values between 0-1 was obtained. Then, a threshold value was used. Thus, binary results consisting of 0 and 1 values were obtained. Three different values were preferred for the threshold: 0.1, 0.5 and 0.9. Thus, the effect of threshold value selection on the test results was examined. The prediction results were evaluated using the masks. For this, general accuracy, recall, precision, F1-score and Jaccard score metrics were used. Recall, precision, and F1-score values were calculated for both burned areas and unburned areas. In addition, minimum, maximum, mean, and standard deviation values were calculated for each metric. When the results are examined, it is seen that the model gives better results when the threshold value is 0.1 and 0.5. Among the band combinations, it is seen that the 7,5,4 combination gave better results than the others. For this band combination, the highest mean accuracy is 0.9743 with the 0.5 threshold value. For this threshold mean recall, mean precision and mean F1-score for burned areas are 0.7203, 0.8411 and 0.7601, respectively. And Jaccard score is 0.6328.\u0000","PeriodicalId":30634,"journal":{"name":"The International Archives of the Photogrammetry Remote Sensing and Spatial Information Sciences","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42575965","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","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：