Pub Date : 2019-07-01DOI: 10.1109/agro-geoinformatics.2019.8820558
{"title":"Agro-Geoinformatics 2019 Committees","authors":"","doi":"10.1109/agro-geoinformatics.2019.8820558","DOIUrl":"https://doi.org/10.1109/agro-geoinformatics.2019.8820558","url":null,"abstract":"","PeriodicalId":143731,"journal":{"name":"2019 8th International Conference on Agro-Geoinformatics (Agro-Geoinformatics)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114902163","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 : 2019-07-01DOI: 10.1109/Agro-Geoinformatics.2019.8820662
Zheng Sun, Di Wang, Qingbo Zhou
Dryland crops in China have a large planting area and wide spatial distribution, which contributes a lot to grain production. Accurate and timely acquisition of information on dryland crop types, planting areas and spatial distribution in dryland can provide an important basis for agricultural production and management, as well as the formulation of national food policy and economic plan, therefore, it is of great significance for promoting structural reform of agricultural supply side and national food security. The key stage of crop growth in the north of China is affected by cloud and rain weather, which makes it difficult to obtain sufficient effective optical data, and the current recognition accuracy of dryland crops based on polarized SAR data is low. In order to solve these problems, this study selected Jizhou City, Hebei Province as the research area, using the full-polarization RADARSAT -2 data of July 17, August 7 and September 24, 2018, and then selected three polarization decomposition methods (Cloude-Pottier decomposition, Freeman decomposition and Yamaguchi decomposition) and two classification methods (maximum likelihood and random forest) to construct 18 classification combinations. The identification of corn and cotton in study area was studied by using various schemes. Finally, the accuracy of dry land crop identification under various combination schemes was compared quantitatively with the ground survey data. The results showed that, Yamaguchi decomposition combined with maximum likelihood classification method was used on August 7 (flowering and boll period of cotton), and the classification accuracy was the highest (production accuracy was 78.98%). For corn, Yamaguchi decomposition combined with random forest classification method was used on September 24 (mature period of corn), and the classification accuracy was the highest (production accuracy was over 90%). I For the decomposition method, Yamaguchi decomposition has the highest classification accuracy among the three decomposition methods, followed by Freeman decomposition, Cloude-Pottier decomposition has the lowest classification accuracy; as far as the classification method is concerned, the maximum likelihood classification method has the highest recognition accuracy for cotton, but the random forest classification has the highest recognition accuracy for corn; in terms of the best identification phase, the flowering and boll period is the best recognition period for cotton, and the maturity period is the best recognition time for corn.. This study will help to improve the recognition accuracy of corn and cotton in fully polarized SAR data, and provide reference for the identification of multi-temporal dryland crops under complex planting structures.
{"title":"Dryland Crop Recognition Based on Multi-temporal Polarization SAR Data","authors":"Zheng Sun, Di Wang, Qingbo Zhou","doi":"10.1109/Agro-Geoinformatics.2019.8820662","DOIUrl":"https://doi.org/10.1109/Agro-Geoinformatics.2019.8820662","url":null,"abstract":"Dryland crops in China have a large planting area and wide spatial distribution, which contributes a lot to grain production. Accurate and timely acquisition of information on dryland crop types, planting areas and spatial distribution in dryland can provide an important basis for agricultural production and management, as well as the formulation of national food policy and economic plan, therefore, it is of great significance for promoting structural reform of agricultural supply side and national food security. The key stage of crop growth in the north of China is affected by cloud and rain weather, which makes it difficult to obtain sufficient effective optical data, and the current recognition accuracy of dryland crops based on polarized SAR data is low. In order to solve these problems, this study selected Jizhou City, Hebei Province as the research area, using the full-polarization RADARSAT -2 data of July 17, August 7 and September 24, 2018, and then selected three polarization decomposition methods (Cloude-Pottier decomposition, Freeman decomposition and Yamaguchi decomposition) and two classification methods (maximum likelihood and random forest) to construct 18 classification combinations. The identification of corn and cotton in study area was studied by using various schemes. Finally, the accuracy of dry land crop identification under various combination schemes was compared quantitatively with the ground survey data. The results showed that, Yamaguchi decomposition combined with maximum likelihood classification method was used on August 7 (flowering and boll period of cotton), and the classification accuracy was the highest (production accuracy was 78.98%). For corn, Yamaguchi decomposition combined with random forest classification method was used on September 24 (mature period of corn), and the classification accuracy was the highest (production accuracy was over 90%). I For the decomposition method, Yamaguchi decomposition has the highest classification accuracy among the three decomposition methods, followed by Freeman decomposition, Cloude-Pottier decomposition has the lowest classification accuracy; as far as the classification method is concerned, the maximum likelihood classification method has the highest recognition accuracy for cotton, but the random forest classification has the highest recognition accuracy for corn; in terms of the best identification phase, the flowering and boll period is the best recognition period for cotton, and the maturity period is the best recognition time for corn.. This study will help to improve the recognition accuracy of corn and cotton in fully polarized SAR data, and provide reference for the identification of multi-temporal dryland crops under complex planting structures.","PeriodicalId":143731,"journal":{"name":"2019 8th International Conference on Agro-Geoinformatics (Agro-Geoinformatics)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131804714","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 : 2019-07-01DOI: 10.1109/Agro-Geoinformatics.2019.8820230
Wenjiao Shi, Yiting Liu
Climate change can affect the shifts of farming-pastoral ecotone (FPE) boundaries, but previous studies have not adequately detected the climate contributions to the FPE boundary shifts. In this study, we presented gravity center analysis, boundary shifts detected in the X- and Y-coordinate direction and the direction of transects along the boundary, and spatial analysis to detect climate contributions at a 1-km scale in different ecological functional regions from the 1970s to the 2000s.Climate and land use data were used in this study. The results showed that during the 1970s–1980s and 1990s–2000s, the northeastern and southeastern parts of the FPE in northern China had similar spatial patterns with more extensive boundary shifts. In the directions of X-, Y-coordinate and the transects along boundaries, different ecological functional regions had significant differences in climate contributions to FPE boundary shifts during the three periods. In addition, during most of the periods, the results in different directions had good agreement in most of the ecological functional regions. However, the values of contributions in the directions of transects in the X- and Y-coordinate directions (4–56%) were always larger than those in the direction of transects along boundaries (1–17%), which shows that the results in the transect directions are more reliable and stable. Thus, the method of detecting the shifts in the transect directions developed by this study is an alternative one for analyzing the climate contributions to boundary shifts. Further evidences for explanation of the driving forces of climate change were given by spatial analysis of the relationship between climate change and land use change in the context of the FPE boundary shifts in northern China. Our findings provide an improved understanding of the responses of boundary shifts in farming–pastoral ecotone of northern China to climate change, which will be important for addressing adaptation and mitigation measures to climate change and regional land use management.
{"title":"Quantitative analysis of the responses of boundary shifts in farming –pastoral ecotone of northern China to climate change","authors":"Wenjiao Shi, Yiting Liu","doi":"10.1109/Agro-Geoinformatics.2019.8820230","DOIUrl":"https://doi.org/10.1109/Agro-Geoinformatics.2019.8820230","url":null,"abstract":"Climate change can affect the shifts of farming-pastoral ecotone (FPE) boundaries, but previous studies have not adequately detected the climate contributions to the FPE boundary shifts. In this study, we presented gravity center analysis, boundary shifts detected in the X- and Y-coordinate direction and the direction of transects along the boundary, and spatial analysis to detect climate contributions at a 1-km scale in different ecological functional regions from the 1970s to the 2000s.Climate and land use data were used in this study. The results showed that during the 1970s–1980s and 1990s–2000s, the northeastern and southeastern parts of the FPE in northern China had similar spatial patterns with more extensive boundary shifts. In the directions of X-, Y-coordinate and the transects along boundaries, different ecological functional regions had significant differences in climate contributions to FPE boundary shifts during the three periods. In addition, during most of the periods, the results in different directions had good agreement in most of the ecological functional regions. However, the values of contributions in the directions of transects in the X- and Y-coordinate directions (4–56%) were always larger than those in the direction of transects along boundaries (1–17%), which shows that the results in the transect directions are more reliable and stable. Thus, the method of detecting the shifts in the transect directions developed by this study is an alternative one for analyzing the climate contributions to boundary shifts. Further evidences for explanation of the driving forces of climate change were given by spatial analysis of the relationship between climate change and land use change in the context of the FPE boundary shifts in northern China. Our findings provide an improved understanding of the responses of boundary shifts in farming–pastoral ecotone of northern China to climate change, which will be important for addressing adaptation and mitigation measures to climate change and regional land use management.","PeriodicalId":143731,"journal":{"name":"2019 8th International Conference on Agro-Geoinformatics (Agro-Geoinformatics)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122941887","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 : 2019-07-01DOI: 10.1109/Agro-Geoinformatics.2019.8820541
Lida Kouhalvandi, Ece Olcay Günes, S. Özoguz
In designing an artificial network, different parameters such as activation functions, hyper-parameters, etc. are considered. Dealing with large number of parameters and also the functions that are expensive for evalualtion are very hard tasks. In this case, it is logical to find methods that results in smaller number of evaluations and improvements in performance. There are various techniques for multiobjective Bayesian optimization in deep learning structure. S-metric selection efficient global optimization (SMS-EGO) and DIRECT are one of the many techniques for multiobjective Bayesian optimization. In this paper, SMS-EGO and DIRECT techniques are applied to deep learning model and the average number of evaluations of each objective including time and error are investigated. For training and validating the deep network, a number of images present various diseases in leaves are provided from Plant Village data set. The simulation results show that by using SMSEGO technique, performance is improved and average time per iteration is faster.
{"title":"Algorithms for Speeding-Up the Deep Neural Networks For Detecting Plant Disease","authors":"Lida Kouhalvandi, Ece Olcay Günes, S. Özoguz","doi":"10.1109/Agro-Geoinformatics.2019.8820541","DOIUrl":"https://doi.org/10.1109/Agro-Geoinformatics.2019.8820541","url":null,"abstract":"In designing an artificial network, different parameters such as activation functions, hyper-parameters, etc. are considered. Dealing with large number of parameters and also the functions that are expensive for evalualtion are very hard tasks. In this case, it is logical to find methods that results in smaller number of evaluations and improvements in performance. There are various techniques for multiobjective Bayesian optimization in deep learning structure. S-metric selection efficient global optimization (SMS-EGO) and DIRECT are one of the many techniques for multiobjective Bayesian optimization. In this paper, SMS-EGO and DIRECT techniques are applied to deep learning model and the average number of evaluations of each objective including time and error are investigated. For training and validating the deep network, a number of images present various diseases in leaves are provided from Plant Village data set. The simulation results show that by using SMSEGO technique, performance is improved and average time per iteration is faster.","PeriodicalId":143731,"journal":{"name":"2019 8th International Conference on Agro-Geoinformatics (Agro-Geoinformatics)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115524314","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 : 2019-07-01DOI: 10.1109/Agro-Geoinformatics.2019.8820526
Jiahui Sheng, Peng Rao, Hongliang Ma
Soil moisture (SM) is a key variable in the study of hydrology, the environment, meteorology, and other fields. One widely used approach to retrieve soil moisture data is based on satellite remote sensing technology. However, the spatiotemporally continuous soil moisture products retrieved from microwave remote sensing data do not meet the accuracy requirements of flood prediction and irrigation management, due to their coarse spatial resolution. China's Fengyun-3B (FY3B) microwave radiation imager (MWRI) soil moisture product is one of the relatively new passive microwave products. Remotely sensed soil moisture data retrieved by the MWRI onboard the FY3B satellite is currently provided at a 25 km grid resolution. In this study, in terms of the thermal inertia theory, the FY3B soil moisture products were downscaled from 25 km to 1 km based on the North American Land Data Assimilation System (NLDAS) grid (12.5 km). For different ranges of the normalized difference vegetation index (NDVI) from the Advanced Very High Resolution Radiometer (AVHRR), the relationship of soil moisture and diurnal temperature change from the land surface model of NLDAS could be obtained. The 1 km soil moisture was then computed from this regression model using 1 km LST data from the Moderate-Resolution Imaging Spectroradiometer (MODIS) (1 km), which was then bias-corrected using FY3B 25 km soil moisture data. The algorithm was applied to every FY3B pixel in the Soil Moisture Active Passive Validation Experiment 2015 (SMAPVEX15). The downscaling results were validated using the in-situ soil moisture from SMAPVEX15. The downscaling estimates better characterize the continuity of spatial and temporal aspects and are more consistent with the soil moisture data used for validation.
{"title":"Downscaling of FY3B Soil Moisture Based on Land Surface Temperature and Vegetation Data","authors":"Jiahui Sheng, Peng Rao, Hongliang Ma","doi":"10.1109/Agro-Geoinformatics.2019.8820526","DOIUrl":"https://doi.org/10.1109/Agro-Geoinformatics.2019.8820526","url":null,"abstract":"Soil moisture (SM) is a key variable in the study of hydrology, the environment, meteorology, and other fields. One widely used approach to retrieve soil moisture data is based on satellite remote sensing technology. However, the spatiotemporally continuous soil moisture products retrieved from microwave remote sensing data do not meet the accuracy requirements of flood prediction and irrigation management, due to their coarse spatial resolution. China's Fengyun-3B (FY3B) microwave radiation imager (MWRI) soil moisture product is one of the relatively new passive microwave products. Remotely sensed soil moisture data retrieved by the MWRI onboard the FY3B satellite is currently provided at a 25 km grid resolution. In this study, in terms of the thermal inertia theory, the FY3B soil moisture products were downscaled from 25 km to 1 km based on the North American Land Data Assimilation System (NLDAS) grid (12.5 km). For different ranges of the normalized difference vegetation index (NDVI) from the Advanced Very High Resolution Radiometer (AVHRR), the relationship of soil moisture and diurnal temperature change from the land surface model of NLDAS could be obtained. The 1 km soil moisture was then computed from this regression model using 1 km LST data from the Moderate-Resolution Imaging Spectroradiometer (MODIS) (1 km), which was then bias-corrected using FY3B 25 km soil moisture data. The algorithm was applied to every FY3B pixel in the Soil Moisture Active Passive Validation Experiment 2015 (SMAPVEX15). The downscaling results were validated using the in-situ soil moisture from SMAPVEX15. The downscaling estimates better characterize the continuity of spatial and temporal aspects and are more consistent with the soil moisture data used for validation.","PeriodicalId":143731,"journal":{"name":"2019 8th International Conference on Agro-Geoinformatics (Agro-Geoinformatics)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125670558","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 : 2019-07-01DOI: 10.1109/Agro-Geoinformatics.2019.8820658
A. Çilek, S. Berberoglu, C. Donmez
This study aims to evaluate CropSyst model for second crop maize in a Mediterranean Environment under different irrigation regimes including, 40% and 60% of water consumption within one-meter deep soil profile. We examined how soil influences maize yields through a process-based crop modelling called CropSyst, and climate variables observed in the Lower Seyhan Plain, Turkey. CropSyst is a process-based simulation model consist of four stages: i) database creation; ii) model calibration; iii) validation; and iv)model results to simulate the growth and development of potential maize crop. Calibration and validation procedures were implemented using climate, soil, management practices, and rotation data previously measured in the field. Daily climate data derived from 22 meteorological stations (including TARBIL Climate station), additionally, soil series, soil classification including soil profiles, profile depth, pH values, organic matter, salinity, texture, soil volume and total porosity have been transferred into GIS environment for modelling. In the world, a significant portion of the freshwater resources (72%) is used in agricultural irrigation. The rapid increase in world population and the need for more water use across sectors increase the importance of more efficient use of irrigation water. Thus, optimum strategies for management and planning of existing water resources in agriculture become a national and global strategic issue.
{"title":"Simulating Second Crop Maize Growth under Different Irrigation Regimes in Lower Seyhan Plain using CropSyst Model","authors":"A. Çilek, S. Berberoglu, C. Donmez","doi":"10.1109/Agro-Geoinformatics.2019.8820658","DOIUrl":"https://doi.org/10.1109/Agro-Geoinformatics.2019.8820658","url":null,"abstract":"This study aims to evaluate CropSyst model for second crop maize in a Mediterranean Environment under different irrigation regimes including, 40% and 60% of water consumption within one-meter deep soil profile. We examined how soil influences maize yields through a process-based crop modelling called CropSyst, and climate variables observed in the Lower Seyhan Plain, Turkey. CropSyst is a process-based simulation model consist of four stages: i) database creation; ii) model calibration; iii) validation; and iv)model results to simulate the growth and development of potential maize crop. Calibration and validation procedures were implemented using climate, soil, management practices, and rotation data previously measured in the field. Daily climate data derived from 22 meteorological stations (including TARBIL Climate station), additionally, soil series, soil classification including soil profiles, profile depth, pH values, organic matter, salinity, texture, soil volume and total porosity have been transferred into GIS environment for modelling. In the world, a significant portion of the freshwater resources (72%) is used in agricultural irrigation. The rapid increase in world population and the need for more water use across sectors increase the importance of more efficient use of irrigation water. Thus, optimum strategies for management and planning of existing water resources in agriculture become a national and global strategic issue.","PeriodicalId":143731,"journal":{"name":"2019 8th International Conference on Agro-Geoinformatics (Agro-Geoinformatics)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128232532","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 : 2019-07-01DOI: 10.1109/Agro-Geoinformatics.2019.8820623
Ömer Vanli, A. Sabuncu, Z. Avci
Accurate and timely information about crop acreage estimation is important for agricultural management. In Turkey, wheat production is very important, and it is widely planted in Anatolia and in Southeastern Turkey. In this study, four different classification types were evaluated for wheat determination. As a study area, the region of Islahiye and Nurdagi counties of Gaziantep, Turkey was chosen. As satellite data, a Landsat 8 OLI image acquired on April 10, 2017 was used. The ground-truth points that were collected in surveying, and additionally field information taken from farmer registration system provided by local administrations were used as references. The application was done by classification of the satellite image using four different methods (Maximum Likelihood, Support Vector Machine, Condition-Based and Nearest Neighbor). After the results were obtained, the wheat classes obtained were transformed to vector format to overlay on the satellite image for visual analysis. The area of wheat class obtained from each method was presented and compared. The results were also evaluated by comparing with the data taken from Turkish Statistical Institute. All of the methods provided results close to the Turkish Statistical Institute records. Even the results were not significantly different from each other, wheat area determined using Support Vector Machine classification was better than others. The accuracy assessments were performed by calculating the total accuracy and KAPPA/KIA coefficient. The accuracy assessment analysis showed that the three supervised methods were better than the unsupervised one. As a future study, evaluation of these four classification methods using a multi-temporal dataset is planned.
{"title":"Regional Classification of Winter Wheat Using Remote Sensing Data in Southeastern Turkey","authors":"Ömer Vanli, A. Sabuncu, Z. Avci","doi":"10.1109/Agro-Geoinformatics.2019.8820623","DOIUrl":"https://doi.org/10.1109/Agro-Geoinformatics.2019.8820623","url":null,"abstract":"Accurate and timely information about crop acreage estimation is important for agricultural management. In Turkey, wheat production is very important, and it is widely planted in Anatolia and in Southeastern Turkey. In this study, four different classification types were evaluated for wheat determination. As a study area, the region of Islahiye and Nurdagi counties of Gaziantep, Turkey was chosen. As satellite data, a Landsat 8 OLI image acquired on April 10, 2017 was used. The ground-truth points that were collected in surveying, and additionally field information taken from farmer registration system provided by local administrations were used as references. The application was done by classification of the satellite image using four different methods (Maximum Likelihood, Support Vector Machine, Condition-Based and Nearest Neighbor). After the results were obtained, the wheat classes obtained were transformed to vector format to overlay on the satellite image for visual analysis. The area of wheat class obtained from each method was presented and compared. The results were also evaluated by comparing with the data taken from Turkish Statistical Institute. All of the methods provided results close to the Turkish Statistical Institute records. Even the results were not significantly different from each other, wheat area determined using Support Vector Machine classification was better than others. The accuracy assessments were performed by calculating the total accuracy and KAPPA/KIA coefficient. The accuracy assessment analysis showed that the three supervised methods were better than the unsupervised one. As a future study, evaluation of these four classification methods using a multi-temporal dataset is planned.","PeriodicalId":143731,"journal":{"name":"2019 8th International Conference on Agro-Geoinformatics (Agro-Geoinformatics)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133811643","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 : 2019-07-01DOI: 10.1109/Agro-Geoinformatics.2019.8820504
Wen Dong, Yingwei Sun, Jiancheng Luo
The rapid development of industrialized agriculture has leads to the problems of soil pollution and water pollution. In order to solve these problems, precision agriculture (PA) has been applied to achieve precise management of agricultural water and fertilizer. In PA process, fine mapping of soil nutrient is an effective technology to acquire accurate water and fertilizer distribution information and make agricultural decision. A significant progress has been made in digital soil mapping (DSM) of soil nutrient content over the past 20 years. However, the accuracy of grid-based DSM cannot meet the practical application needs of PA. This paper proposed a fine DSM method of soil nutrient content using high resolution remote sensing images and multi-scale auxiliary data for PA application. Three key technologies were studied for the implementation of this method. The automatic extraction of fine mapping units was the basis of this method. We designed different automatic extraction methods based on high resolution remote sensing images for agricultural production units in plains and mountainous areas. The auxiliary variables in different scales were chosen and converted to construct fine-scale soil nutrient-environment relationship model. Finally, machine learning methods were used to map the spatial distribution of soil nutrients. We chose Zhongning County, Ningxia Province as the study area, which includes typical plain and mountainous agriculture. The proposed method and technologies were applied for typical soil nutrients mapping. A common grid-based spatial interpolation method was implemented with the same soil sample dataset to evaluate the effect of the proposed method. The result showed that this method could reduce the number of prediction units and effectively improve the prediction efficiency in both plain and mountainous areas for fine soil mapping and precision agriculture application. This study was an attempt to realize fine soil mapping based on PA application unit in different environments. The high-resolution remote sensing images provide basic data for the realization of this idea, and the conversion technology of multi-scale data provides better support for the spatial inference of fine soil attribute information. In the future, we will carry out experiments in larger areas to further improve the efficiency of application, and plan to expand this study to consider three-dimensional soil property prediction.
{"title":"Fine mapping of key soil nutrient content using high resolution remote sensing image to support precision agriculture in Northwest China","authors":"Wen Dong, Yingwei Sun, Jiancheng Luo","doi":"10.1109/Agro-Geoinformatics.2019.8820504","DOIUrl":"https://doi.org/10.1109/Agro-Geoinformatics.2019.8820504","url":null,"abstract":"The rapid development of industrialized agriculture has leads to the problems of soil pollution and water pollution. In order to solve these problems, precision agriculture (PA) has been applied to achieve precise management of agricultural water and fertilizer. In PA process, fine mapping of soil nutrient is an effective technology to acquire accurate water and fertilizer distribution information and make agricultural decision. A significant progress has been made in digital soil mapping (DSM) of soil nutrient content over the past 20 years. However, the accuracy of grid-based DSM cannot meet the practical application needs of PA. This paper proposed a fine DSM method of soil nutrient content using high resolution remote sensing images and multi-scale auxiliary data for PA application. Three key technologies were studied for the implementation of this method. The automatic extraction of fine mapping units was the basis of this method. We designed different automatic extraction methods based on high resolution remote sensing images for agricultural production units in plains and mountainous areas. The auxiliary variables in different scales were chosen and converted to construct fine-scale soil nutrient-environment relationship model. Finally, machine learning methods were used to map the spatial distribution of soil nutrients. We chose Zhongning County, Ningxia Province as the study area, which includes typical plain and mountainous agriculture. The proposed method and technologies were applied for typical soil nutrients mapping. A common grid-based spatial interpolation method was implemented with the same soil sample dataset to evaluate the effect of the proposed method. The result showed that this method could reduce the number of prediction units and effectively improve the prediction efficiency in both plain and mountainous areas for fine soil mapping and precision agriculture application. This study was an attempt to realize fine soil mapping based on PA application unit in different environments. The high-resolution remote sensing images provide basic data for the realization of this idea, and the conversion technology of multi-scale data provides better support for the spatial inference of fine soil attribute information. In the future, we will carry out experiments in larger areas to further improve the efficiency of application, and plan to expand this study to consider three-dimensional soil property prediction.","PeriodicalId":143731,"journal":{"name":"2019 8th International Conference on Agro-Geoinformatics (Agro-Geoinformatics)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130576383","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 : 2019-07-01DOI: 10.1109/Agro-Geoinformatics.2019.8820239
C. Leng, Shanzhen Yi, Wenhao Xie
Rainfall is not only an essential parameter in hydrology and in the research of water resources, but also an important consideration for the issue of flood control, disaster mitigation, runoff forecast, irrigation, etc. However, the conventional monitoring approaches of rainfall involve many disadvantages, such as limited observing range, high cost and only-one-point rainfall observation. Consequently, how to get the rainfall of any part of the valley attracts more and more attention. In this study, the main meteorological parameters which influencing the rainfall can be extracted from the MODIS satellite cloud imagery, and these meteorological parameters are combined with the actual observed rainfall data which is obtained from ground-based rainfall site correspondingly. The remote sensing retrieval model is established respectively based on the BP neural network and GA-BP neural network, and a better effect of error precision estimation is obtained. It’s also proved that the high resolution of MODIS cloud products can be used to estimate rainfall rate.
{"title":"Estimation of rainfall based on MODIS using neural networks","authors":"C. Leng, Shanzhen Yi, Wenhao Xie","doi":"10.1109/Agro-Geoinformatics.2019.8820239","DOIUrl":"https://doi.org/10.1109/Agro-Geoinformatics.2019.8820239","url":null,"abstract":"Rainfall is not only an essential parameter in hydrology and in the research of water resources, but also an important consideration for the issue of flood control, disaster mitigation, runoff forecast, irrigation, etc. However, the conventional monitoring approaches of rainfall involve many disadvantages, such as limited observing range, high cost and only-one-point rainfall observation. Consequently, how to get the rainfall of any part of the valley attracts more and more attention. In this study, the main meteorological parameters which influencing the rainfall can be extracted from the MODIS satellite cloud imagery, and these meteorological parameters are combined with the actual observed rainfall data which is obtained from ground-based rainfall site correspondingly. The remote sensing retrieval model is established respectively based on the BP neural network and GA-BP neural network, and a better effect of error precision estimation is obtained. It’s also proved that the high resolution of MODIS cloud products can be used to estimate rainfall rate.","PeriodicalId":143731,"journal":{"name":"2019 8th International Conference on Agro-Geoinformatics (Agro-Geoinformatics)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130459315","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：