Pub Date : 2008-09-05DOI: 10.1109/EORSA.2008.4620325
Jian Tan, Xiang Tao Fan, Jun Jie Zhu, Xiao Ping Du, Weibing Wang, Zhaoming Zhong, Chaoji Ma
Geographical GRID system is of great importance in fields such as public security, military action, emergency response etc. The homogenizing distributed geographic environment system requires the same geographical information for operations in each node. The bottle neck is how to reliably and accurately synchronize the great volume geographical data. This paper solves the problem in three ways. First, the message server queue is constructed for stable message delivery. In this way, the message server always has its alternative in preparation for breakdowns, and the whole GRID always has single working message server. Then the message server queue can be constructed and effectively works. This mode has the advantages of the other two modes that the message delivery is more reliable and less time-costing. Second, both push and pull modes are adopted to send messages in time. Push mode means the node which has altered its data is responsible for the delivery of the changed part, like ldquopushrdquo the data to the message server. While pull mode means the demand node or the message server is responsible to check the data status in other nodes and ldquopullrdquo the new data from the source. In push mode, if the network between the sponsor node and the message server break down, the message could be missing or the sponsor could be halted, when the network resumed, the update action could not be invoked again. And in pull mode, the message server needs to check the data and collect update parts in the whole grid, which is a time-costing operation that could not be executed frequently. So the combination mode is adopted. In combination mode, not only does each node has its own update trigger to invoke the delivery of the new data, but also the message server also can recurrently check the data status after an assigned interval according to the network situation and the computation ability, then the duly update can be guaranteed. Third, an extended GML is developed to wrap the geographical data. GML defines a lot of types of elements and attributes to describe the geographical entity in detail. But to synchronize geo-information in GRID-GIS, these definitions are not adequate. Because the spatial data must be wrapped into small flexible and linkable unit to cut down the time of delivering and receiving which are the most unstable periods in synchronizing course and to resend and assembly the units in unambiguous order. So our system developed the extended GML format, in which granularity level, including relation, inner string length are defined. By its help, the volume of data message is controllable and it is more reliable and accurate to resend and assembly the data fragments. These three methods are the key solutions to the geographical information synchronizing in GRID-GIS. Their validity has been proved in practice.
{"title":"Research on geographical information synchronizing in GRID-GIS","authors":"Jian Tan, Xiang Tao Fan, Jun Jie Zhu, Xiao Ping Du, Weibing Wang, Zhaoming Zhong, Chaoji Ma","doi":"10.1109/EORSA.2008.4620325","DOIUrl":"https://doi.org/10.1109/EORSA.2008.4620325","url":null,"abstract":"Geographical GRID system is of great importance in fields such as public security, military action, emergency response etc. The homogenizing distributed geographic environment system requires the same geographical information for operations in each node. The bottle neck is how to reliably and accurately synchronize the great volume geographical data. This paper solves the problem in three ways. First, the message server queue is constructed for stable message delivery. In this way, the message server always has its alternative in preparation for breakdowns, and the whole GRID always has single working message server. Then the message server queue can be constructed and effectively works. This mode has the advantages of the other two modes that the message delivery is more reliable and less time-costing. Second, both push and pull modes are adopted to send messages in time. Push mode means the node which has altered its data is responsible for the delivery of the changed part, like ldquopushrdquo the data to the message server. While pull mode means the demand node or the message server is responsible to check the data status in other nodes and ldquopullrdquo the new data from the source. In push mode, if the network between the sponsor node and the message server break down, the message could be missing or the sponsor could be halted, when the network resumed, the update action could not be invoked again. And in pull mode, the message server needs to check the data and collect update parts in the whole grid, which is a time-costing operation that could not be executed frequently. So the combination mode is adopted. In combination mode, not only does each node has its own update trigger to invoke the delivery of the new data, but also the message server also can recurrently check the data status after an assigned interval according to the network situation and the computation ability, then the duly update can be guaranteed. Third, an extended GML is developed to wrap the geographical data. GML defines a lot of types of elements and attributes to describe the geographical entity in detail. But to synchronize geo-information in GRID-GIS, these definitions are not adequate. Because the spatial data must be wrapped into small flexible and linkable unit to cut down the time of delivering and receiving which are the most unstable periods in synchronizing course and to resend and assembly the units in unambiguous order. So our system developed the extended GML format, in which granularity level, including relation, inner string length are defined. By its help, the volume of data message is controllable and it is more reliable and accurate to resend and assembly the data fragments. These three methods are the key solutions to the geographical information synchronizing in GRID-GIS. Their validity has been proved in practice.","PeriodicalId":142612,"journal":{"name":"2008 International Workshop on Earth Observation and Remote Sensing Applications","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115537965","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 : 2008-09-05DOI: 10.1109/EORSA.2008.4620318
Aixia Liu, Wang Jing, Zhengjun Liu
The Three Gorges Project (TGP) is a vital project in the development and harnessing of the Yangtze River in China. The Three Gorges Reservoir region, while with serious soil erosion for a long time, is one of the typical regions whose ecological environment is very fragile. The objective of this study is to map the inter-annual spatial changing pattern of mean soil erosion and assess the effects of land use changes on soil erosion for this area. This paper presents a comprehensive methodology that integrates an erosion model, the Revised Universal Soil Loss Equation (RUSLE) with a geographic information system (GIS) and remote sensing (RS) technology for estimating soil erosion in the Three Gorges Reservoir area. Firstly, the basic data of soil, precipitation, vegetation and crop management, and the DEM, land use map and Landsat TM images of Three Gorges Reservoir region were collected. With the aid of GIS and RS technology, the value for R, K, LS, C and P factors used by RUSLE in the estimation of soil erosion were computed respectively using suitable methods. Then based on the RUSLE the mean soil erosion amount and the soil erosion modulus in 1977 and 2005 of this region were obtained. Finally, keeping the R, K and LS factors invariant, we analyzed the effects of land use changes between 1977 and 2005 on soil erosion for the Three Gorges Reservoir region. The results show that the mean soil erosion amount and the soil erosion modulus in Three Gorges Reservoir region were 18476.27times104 t/a and 3316.53 t/(km2.a), respectively. The annual soil conservation amount was 48427633times104 t/a. The average soil conservation capacity of Three Gorges Project region is 156.27. By comparison of the soil erosion between 1977 and 2005, we can see that the total soil erosion amount decreased by 449.07times104 t and the mean soil erosion modulus increased by 13.17 t/(km2.a) from 1977 to 2005 because of the land use change. The integrated approach in this study allows for relatively easy, fast, and cost-effective estimation of spatially distributed soil erosion and analyses of the land use changespsila effects on soil erosion. It thus provides a useful and efficient tool for predicting long-term soil erosion potential and assessing erosion impacts of conservation support practices.
{"title":"Assessing the effects of land use changes on soil erosion in Three Gorges Reservoir Region of China","authors":"Aixia Liu, Wang Jing, Zhengjun Liu","doi":"10.1109/EORSA.2008.4620318","DOIUrl":"https://doi.org/10.1109/EORSA.2008.4620318","url":null,"abstract":"The Three Gorges Project (TGP) is a vital project in the development and harnessing of the Yangtze River in China. The Three Gorges Reservoir region, while with serious soil erosion for a long time, is one of the typical regions whose ecological environment is very fragile. The objective of this study is to map the inter-annual spatial changing pattern of mean soil erosion and assess the effects of land use changes on soil erosion for this area. This paper presents a comprehensive methodology that integrates an erosion model, the Revised Universal Soil Loss Equation (RUSLE) with a geographic information system (GIS) and remote sensing (RS) technology for estimating soil erosion in the Three Gorges Reservoir area. Firstly, the basic data of soil, precipitation, vegetation and crop management, and the DEM, land use map and Landsat TM images of Three Gorges Reservoir region were collected. With the aid of GIS and RS technology, the value for R, K, LS, C and P factors used by RUSLE in the estimation of soil erosion were computed respectively using suitable methods. Then based on the RUSLE the mean soil erosion amount and the soil erosion modulus in 1977 and 2005 of this region were obtained. Finally, keeping the R, K and LS factors invariant, we analyzed the effects of land use changes between 1977 and 2005 on soil erosion for the Three Gorges Reservoir region. The results show that the mean soil erosion amount and the soil erosion modulus in Three Gorges Reservoir region were 18476.27times104 t/a and 3316.53 t/(km2.a), respectively. The annual soil conservation amount was 48427633times104 t/a. The average soil conservation capacity of Three Gorges Project region is 156.27. By comparison of the soil erosion between 1977 and 2005, we can see that the total soil erosion amount decreased by 449.07times104 t and the mean soil erosion modulus increased by 13.17 t/(km2.a) from 1977 to 2005 because of the land use change. The integrated approach in this study allows for relatively easy, fast, and cost-effective estimation of spatially distributed soil erosion and analyses of the land use changespsila effects on soil erosion. It thus provides a useful and efficient tool for predicting long-term soil erosion potential and assessing erosion impacts of conservation support practices.","PeriodicalId":142612,"journal":{"name":"2008 International Workshop on Earth Observation and Remote Sensing Applications","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127059539","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 : 2008-09-05DOI: 10.1109/EORSA.2008.4620290
Guo Zhifeng, C. Yulin, Y. Li
The hydrologically based variable infiltration capacity (VIC) macroscale hydrologic model was applied to simulate streamflow for Poyang Lake Basin in China. DEM needed to get basin characteristics is from SRTM. The required soil parameters are derived from the soil classification information of global 5 min data provided by the National Atmospheric and Oceanic Administration (NOAA) Hydrology Office, the vegetation parameters are derived based on MODIS products and land data assimilation system (LDAS) and the forcing data are obtained through interpolation method based on 151 stations. All of the data (i.e. soil, vegetation, and forcings) needed by VIC-3L are compiled with at 8times8 km2 resolution. The VIC-3L model is applied to the Yellow River basin and the simulated daily runoff is routed to the outlet of two stations using ARNO model and compared to daily observed streamflow at these stations. Results show that remote sensing data can play the important role in model simulation process, though application of remote sensing data can not improve the performance of the model very much.
{"title":"A macro hydrologic model simulation based on remote sensing data","authors":"Guo Zhifeng, C. Yulin, Y. Li","doi":"10.1109/EORSA.2008.4620290","DOIUrl":"https://doi.org/10.1109/EORSA.2008.4620290","url":null,"abstract":"The hydrologically based variable infiltration capacity (VIC) macroscale hydrologic model was applied to simulate streamflow for Poyang Lake Basin in China. DEM needed to get basin characteristics is from SRTM. The required soil parameters are derived from the soil classification information of global 5 min data provided by the National Atmospheric and Oceanic Administration (NOAA) Hydrology Office, the vegetation parameters are derived based on MODIS products and land data assimilation system (LDAS) and the forcing data are obtained through interpolation method based on 151 stations. All of the data (i.e. soil, vegetation, and forcings) needed by VIC-3L are compiled with at 8times8 km2 resolution. The VIC-3L model is applied to the Yellow River basin and the simulated daily runoff is routed to the outlet of two stations using ARNO model and compared to daily observed streamflow at these stations. Results show that remote sensing data can play the important role in model simulation process, though application of remote sensing data can not improve the performance of the model very much.","PeriodicalId":142612,"journal":{"name":"2008 International Workshop on Earth Observation and Remote Sensing Applications","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121887924","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 : 2008-09-05DOI: 10.1109/EORSA.2008.4620338
Jing Xu, Bo Liu, Jinguo Yuan, Changyao Wang
GPS data of August in 2004 obtained from 4 sites in Hong Kong GPS/MET network has been used to inverse the precipitable water vapor. The inversed GPS integrated water vapor has 1.44 mm RMSE and 0.97 mm BIAS compared with that from radiosonde data in Hong Kong Observatory, which shows good agreement between them. In this paper, we analyze the temporal and spatial change of Hong Kong summer water vapor using inversed GPS PWV, and have studied the relations of integrated water vapor with the average temperature, precipitation and ground vapor pressure. The results show that: there exists obvious temporal and spatial change of PWV, and for stations with close basic geographic location, it is obvious of the impact of altitude that the PWV in mountain station is lower than that in plain point under normal circumstances; ground vapor pressure has a good correlation with PWV; there often comes forth a precipitation process when the PWV increases rapidly, but there is no obvious correlation between the amount of integrated water vapor and the size of precipitation, and so the average amount before, the increase range in short time, and the maximum value of PWV should be considered in the precipitation forecast.
利用香港GPS/MET网络4个站点2004年8月的GPS数据反演可降水量。与香港天文台探空资料相比，GPS反演综合水汽的RMSE值为1.44 mm, BIAS值为0.97 mm，两者吻合较好。本文利用GPS PWV反演资料分析了香港夏季水汽的时空变化，并研究了综合水汽与平均气温、降水和地面水汽压的关系。结果表明:PWV存在明显的时空变化，对于基础地理位置较近的站点，海拔高度的影响较为明显，一般情况下山地站点的PWV低于平原站点;地面蒸汽压与PWV有良好的相关性;当PWV快速增加时，往往会出现降水过程，但综合水汽量与降水大小之间没有明显的相关性，因此在降水预报中应考虑之前的平均值、短时内的增加幅度和PWV的最大值。
{"title":"Inversion of precipitable water vapor in Hongkong","authors":"Jing Xu, Bo Liu, Jinguo Yuan, Changyao Wang","doi":"10.1109/EORSA.2008.4620338","DOIUrl":"https://doi.org/10.1109/EORSA.2008.4620338","url":null,"abstract":"GPS data of August in 2004 obtained from 4 sites in Hong Kong GPS/MET network has been used to inverse the precipitable water vapor. The inversed GPS integrated water vapor has 1.44 mm RMSE and 0.97 mm BIAS compared with that from radiosonde data in Hong Kong Observatory, which shows good agreement between them. In this paper, we analyze the temporal and spatial change of Hong Kong summer water vapor using inversed GPS PWV, and have studied the relations of integrated water vapor with the average temperature, precipitation and ground vapor pressure. The results show that: there exists obvious temporal and spatial change of PWV, and for stations with close basic geographic location, it is obvious of the impact of altitude that the PWV in mountain station is lower than that in plain point under normal circumstances; ground vapor pressure has a good correlation with PWV; there often comes forth a precipitation process when the PWV increases rapidly, but there is no obvious correlation between the amount of integrated water vapor and the size of precipitation, and so the average amount before, the increase range in short time, and the maximum value of PWV should be considered in the precipitation forecast.","PeriodicalId":142612,"journal":{"name":"2008 International Workshop on Earth Observation and Remote Sensing Applications","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126447034","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 : 2008-09-05DOI: 10.1109/EORSA.2008.4620306
S. Han, H.T. Li, H. Gu
Conventional map-to-map comparison method is used frequently to land use change detection from remote sensing data. However, only the spectral information is used in the process of classification. The accuracy is very low. This paper applies object-oriented image analysis to the change detection and can overcome the limitation stated above. It classifies through selecting samples, using multi-scale image segmentation techniques, allowing the integration of a broad spectrum of different object features, such as spectral values, shape, texture and thematic information. In the study, it detects the land use change information of the test area during ten years through the method based on object-oriented analysis. The result is satisfying and can provide the foundation for sustaining development of land resource.
{"title":"The study of land use change detection based on object-oriented analysis","authors":"S. Han, H.T. Li, H. Gu","doi":"10.1109/EORSA.2008.4620306","DOIUrl":"https://doi.org/10.1109/EORSA.2008.4620306","url":null,"abstract":"Conventional map-to-map comparison method is used frequently to land use change detection from remote sensing data. However, only the spectral information is used in the process of classification. The accuracy is very low. This paper applies object-oriented image analysis to the change detection and can overcome the limitation stated above. It classifies through selecting samples, using multi-scale image segmentation techniques, allowing the integration of a broad spectrum of different object features, such as spectral values, shape, texture and thematic information. In the study, it detects the land use change information of the test area during ten years through the method based on object-oriented analysis. The result is satisfying and can provide the foundation for sustaining development of land resource.","PeriodicalId":142612,"journal":{"name":"2008 International Workshop on Earth Observation and Remote Sensing Applications","volume":"220 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130508385","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}
In this paper, a novel tracker for semi-automatic extraction of ribbon road centerlines from high resolution remotely sensed imagery is proposed. Actually, our approach is an integration of least squares profile matching and least squares rectangular template matching. After initialization, a road template model is built which is composed of two parts: a profile perpendicular to the road axis, and some rectangular templates of strips of road marks or strips of vegetation parallel to road moving direction. In tracking process, least squares matching is employed to search road centerline points, and parabola is deployed to model the road trajectory to predict the position of subsequent road points and to guide the tracking go through bad road conditions. Extensive experiments demonstrate that our proposed algorithm can fast and reliably trace roads with road marks or strips of vegetation despite of appearance of much occlusion from trees, building or vehicles.
{"title":"Integration method of profile matching and template matching for road extraction from high resolution remotely sensed imagery","authors":"Xiangguo Lin, Jixian Zhang, Zhengjun Liu, Jing Shen","doi":"10.1109/EORSA.2008.4620317","DOIUrl":"https://doi.org/10.1109/EORSA.2008.4620317","url":null,"abstract":"In this paper, a novel tracker for semi-automatic extraction of ribbon road centerlines from high resolution remotely sensed imagery is proposed. Actually, our approach is an integration of least squares profile matching and least squares rectangular template matching. After initialization, a road template model is built which is composed of two parts: a profile perpendicular to the road axis, and some rectangular templates of strips of road marks or strips of vegetation parallel to road moving direction. In tracking process, least squares matching is employed to search road centerline points, and parabola is deployed to model the road trajectory to predict the position of subsequent road points and to guide the tracking go through bad road conditions. Extensive experiments demonstrate that our proposed algorithm can fast and reliably trace roads with road marks or strips of vegetation despite of appearance of much occlusion from trees, building or vehicles.","PeriodicalId":142612,"journal":{"name":"2008 International Workshop on Earth Observation and Remote Sensing Applications","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131013843","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 : 2008-09-05DOI: 10.1109/EORSA.2008.4620299
Peijun Du, Yan Luo, W. Cao, Huapeng Zhang
Some analytical approaches have been developed and widely used for vegetation remote sensing, among which four popular methods are vegetation analysis via NDVI and other VIs, vegetation analysis using the vegetation abundance derived from unmixing, vegetation analysis by land cover classification, and the greenness component derived from K-T transform. There four approaches are used to extract vegetation information from Landsat TM image taking Xuzhou City as an example, and their performance is compared. Association analysis among vegetation types, NDVI values, vegetation abundance and greenness is conducted at first. It is found that the association among NDVI, vegetation abundance and greenness is quite obvious. Vegetation coverage ratio is derived based on different vegetation extraction approaches, and their consistency is analyzed. It is found that the unmixing-based approach outperforms others in terms of vegetation coverage ration estimation. By comparing the performance and effectiveness of four approaches, some suggestions are given for selecting suitable analytical approaches.
{"title":"A comparison and evaluation of four vegetation analysis approaches based on remote sensing imagery","authors":"Peijun Du, Yan Luo, W. Cao, Huapeng Zhang","doi":"10.1109/EORSA.2008.4620299","DOIUrl":"https://doi.org/10.1109/EORSA.2008.4620299","url":null,"abstract":"Some analytical approaches have been developed and widely used for vegetation remote sensing, among which four popular methods are vegetation analysis via NDVI and other VIs, vegetation analysis using the vegetation abundance derived from unmixing, vegetation analysis by land cover classification, and the greenness component derived from K-T transform. There four approaches are used to extract vegetation information from Landsat TM image taking Xuzhou City as an example, and their performance is compared. Association analysis among vegetation types, NDVI values, vegetation abundance and greenness is conducted at first. It is found that the association among NDVI, vegetation abundance and greenness is quite obvious. Vegetation coverage ratio is derived based on different vegetation extraction approaches, and their consistency is analyzed. It is found that the unmixing-based approach outperforms others in terms of vegetation coverage ration estimation. By comparing the performance and effectiveness of four approaches, some suggestions are given for selecting suitable analytical approaches.","PeriodicalId":142612,"journal":{"name":"2008 International Workshop on Earth Observation and Remote Sensing Applications","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128506526","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 : 2008-09-05DOI: 10.1109/EORSA.2008.4620334
M. Wong, Wai Yeung Yan
Multiple Classifier System (MCS) has attracted increasing interest in the field of pattern recognition and machine learning where this technique has also been introduced in remote sensing. The importance of classifier diversity in MCS has been raised recently; nevertheless, only a few of the researches have been studied in land cover classification problem. In this paper, a SPOT IV satellite image covering the Hong Kong Island and Kowloon Peninsula with six land cover classes were classified with four base classifiers: Minimum Distance Classifier, Maximum Likelihood Classifier, Mahalanobis Classifier and K-Nearest Neighbor Classifier. Same training and testing data sets were applied throughout the experiments and five Bayesian decision rules, including product rule, sum rule, max rule, min rule, and median rule, were utilized to construct different ensemble of classifiers. Performance of MCS was measured using the overall accuracy and kappa statistics, and three statistical tests including McNemarpsilas test, Cochranpsilas Q test and F-test were introduced to examine the dependence of the classification results. The experimental comparison reveals that (i) increasing the number of base classifiers may not improve the overall accuracy in MCS, (ii) significant diversity in base classifiers cannot enhance the overall performance and vice versa. These findings are noted with the condition in using the same data set and the same training set.
{"title":"Investigation of diversity and accuracy in ensemble of classifiers using Bayesian decision rules","authors":"M. Wong, Wai Yeung Yan","doi":"10.1109/EORSA.2008.4620334","DOIUrl":"https://doi.org/10.1109/EORSA.2008.4620334","url":null,"abstract":"Multiple Classifier System (MCS) has attracted increasing interest in the field of pattern recognition and machine learning where this technique has also been introduced in remote sensing. The importance of classifier diversity in MCS has been raised recently; nevertheless, only a few of the researches have been studied in land cover classification problem. In this paper, a SPOT IV satellite image covering the Hong Kong Island and Kowloon Peninsula with six land cover classes were classified with four base classifiers: Minimum Distance Classifier, Maximum Likelihood Classifier, Mahalanobis Classifier and K-Nearest Neighbor Classifier. Same training and testing data sets were applied throughout the experiments and five Bayesian decision rules, including product rule, sum rule, max rule, min rule, and median rule, were utilized to construct different ensemble of classifiers. Performance of MCS was measured using the overall accuracy and kappa statistics, and three statistical tests including McNemarpsilas test, Cochranpsilas Q test and F-test were introduced to examine the dependence of the classification results. The experimental comparison reveals that (i) increasing the number of base classifiers may not improve the overall accuracy in MCS, (ii) significant diversity in base classifiers cannot enhance the overall performance and vice versa. These findings are noted with the condition in using the same data set and the same training set.","PeriodicalId":142612,"journal":{"name":"2008 International Workshop on Earth Observation and Remote Sensing Applications","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114404041","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 : 2008-09-05DOI: 10.1109/EORSA.2008.4620326
X. Tong, Shijie Liu
Along with a wide spread application of high resolution satellite imagery (HRSI) in urban mapping and change detection, the demand is increasing for higher accuracy metric products from HRSI. However, due to the inherent sensor orientation bias, the rational polynomial coefficients (RPCs) provided by the vendors can not produce accurate coordinates in ground point determination. Based on the QuickBird stereo HRSI, this paper has presented the performance of two schemes for bias elimination, including RPCs modification and RPCs regeneration. The experimental results show that, for the testing area, modified RPCs yields low positioning accuracy of 2 - 3 meters in planimetry and 3 meters in elevation since it is responsible for correcting only the shift bias. However, with minimal 2 control points, regenerated RPCs with shift and drift bias compensated produces accuracy of about 0.6 meters in planimetry and 1 meter in elevation. If more control points are available, RPCs regeneration with affine model bias compensation is more recommended. The bias-corrected RPCs provide a steady accuracy of half meter in planimetry. Furthermore, the bias-eliminated RPCs facilitate bias-free application. They can be used as replacements of the originals producing high accuracy in photogrammetric system for further processing such as ortho-rectification and DEM generation to provide cost advantage.
{"title":"Bias-corrected RPCs for QuickBird stereo satellite imagery: A case study in Shanghai","authors":"X. Tong, Shijie Liu","doi":"10.1109/EORSA.2008.4620326","DOIUrl":"https://doi.org/10.1109/EORSA.2008.4620326","url":null,"abstract":"Along with a wide spread application of high resolution satellite imagery (HRSI) in urban mapping and change detection, the demand is increasing for higher accuracy metric products from HRSI. However, due to the inherent sensor orientation bias, the rational polynomial coefficients (RPCs) provided by the vendors can not produce accurate coordinates in ground point determination. Based on the QuickBird stereo HRSI, this paper has presented the performance of two schemes for bias elimination, including RPCs modification and RPCs regeneration. The experimental results show that, for the testing area, modified RPCs yields low positioning accuracy of 2 - 3 meters in planimetry and 3 meters in elevation since it is responsible for correcting only the shift bias. However, with minimal 2 control points, regenerated RPCs with shift and drift bias compensated produces accuracy of about 0.6 meters in planimetry and 1 meter in elevation. If more control points are available, RPCs regeneration with affine model bias compensation is more recommended. The bias-corrected RPCs provide a steady accuracy of half meter in planimetry. Furthermore, the bias-eliminated RPCs facilitate bias-free application. They can be used as replacements of the originals producing high accuracy in photogrammetric system for further processing such as ortho-rectification and DEM generation to provide cost advantage.","PeriodicalId":142612,"journal":{"name":"2008 International Workshop on Earth Observation and Remote Sensing Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116786131","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 : 2008-09-05DOI: 10.1109/EORSA.2008.4620339
Limin Yang
An accurate, up-to-date and spatially-explicate national land cover database is required for monitoring the status and trends of the nationpsilas terrestrial ecosystem, and for managing and conserving land resources at the national scale. With all the challenges and resources required to develop such a database, an innovative and scientifically sound planning must be in place and a partnership be formed among users from government agencies, research institutes and private sectors. In this paper, we summarize major scientific and technical issues regarding the development of the NLCD 1992 and 2001. Experiences and lessons learned from the project are documented with regard to project design, technical approaches, accuracy assessment strategy, and project implementation. Future improvements in developing next generation NLCD beyond 2001 are suggested, including: (1) enhanced satellite data preprocessing in correction of atmospheric and adjacency effect and the topographic normalization; (2) improved classification accuracy through comprehensive and consistent training data and new algorithm development; (3) multi-resolution and multitemporal database targeting major land cover changes and land cover database updates; (4) enriched database contents by including additional biophysical parameters and/or more detailed land cover classes through synergizing multi-sensor, multi-temporal, and multi-spectral satellite data and ancillary data, and (5) transform the NLCD project into a national land cover monitoring program.
{"title":"United States national land cover data base development 1992–2001 and beyond","authors":"Limin Yang","doi":"10.1109/EORSA.2008.4620339","DOIUrl":"https://doi.org/10.1109/EORSA.2008.4620339","url":null,"abstract":"An accurate, up-to-date and spatially-explicate national land cover database is required for monitoring the status and trends of the nationpsilas terrestrial ecosystem, and for managing and conserving land resources at the national scale. With all the challenges and resources required to develop such a database, an innovative and scientifically sound planning must be in place and a partnership be formed among users from government agencies, research institutes and private sectors. In this paper, we summarize major scientific and technical issues regarding the development of the NLCD 1992 and 2001. Experiences and lessons learned from the project are documented with regard to project design, technical approaches, accuracy assessment strategy, and project implementation. Future improvements in developing next generation NLCD beyond 2001 are suggested, including: (1) enhanced satellite data preprocessing in correction of atmospheric and adjacency effect and the topographic normalization; (2) improved classification accuracy through comprehensive and consistent training data and new algorithm development; (3) multi-resolution and multitemporal database targeting major land cover changes and land cover database updates; (4) enriched database contents by including additional biophysical parameters and/or more detailed land cover classes through synergizing multi-sensor, multi-temporal, and multi-spectral satellite data and ancillary data, and (5) transform the NLCD project into a national land cover monitoring program.","PeriodicalId":142612,"journal":{"name":"2008 International Workshop on Earth Observation and Remote Sensing Applications","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124945734","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：