Pub Date : 2011-07-12DOI: 10.1109/MULTI-TEMP.2011.6005090
R. R. V. Gonçalves, J. Zullo, C. S. Ferraresso, E. P. M. Sousa, L. A. Romani, A. J. Traina
The purpose of this work is to assess the sugarcane yield variation in regional scale through NDVI images from a low resolution spatial satellite. We have used Principal Component Analysis (PCA) and Cluster Analysis to correlate sugarcane cultivated land with multitemporal NDVI images also verifying the influence of climate conditions to them. According to both techniques (PCA and clustering), clusters for different set of variables are distinct only when cultivated land was included in the dataset. On the contrary, climate variables determine the clustering formation. Exploring multitemporal images from high resolution satellites through data mining techniques, such as cluster analysis, is a valuable way to improve crops monitoring specially at a time when it becomes increasingly important to understand the impact of climate change on agriculture.
{"title":"Analysis of NOAA/AVHRR multitemporal images, climate conditions and cultivated land of sugarcane fields applied to agricultural monitoring","authors":"R. R. V. Gonçalves, J. Zullo, C. S. Ferraresso, E. P. M. Sousa, L. A. Romani, A. J. Traina","doi":"10.1109/MULTI-TEMP.2011.6005090","DOIUrl":"https://doi.org/10.1109/MULTI-TEMP.2011.6005090","url":null,"abstract":"The purpose of this work is to assess the sugarcane yield variation in regional scale through NDVI images from a low resolution spatial satellite. We have used Principal Component Analysis (PCA) and Cluster Analysis to correlate sugarcane cultivated land with multitemporal NDVI images also verifying the influence of climate conditions to them. According to both techniques (PCA and clustering), clusters for different set of variables are distinct only when cultivated land was included in the dataset. On the contrary, climate variables determine the clustering formation. Exploring multitemporal images from high resolution satellites through data mining techniques, such as cluster analysis, is a valuable way to improve crops monitoring specially at a time when it becomes increasingly important to understand the impact of climate change on agriculture.","PeriodicalId":254778,"journal":{"name":"2011 6th International Workshop on the Analysis of Multi-temporal Remote Sensing Images (Multi-Temp)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129657622","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 : 2011-07-12DOI: 10.1109/MULTI-TEMP.2011.6005074
A. Nielsen, O. Andersen, P. Svendsen
This paper gives an introductory analysis of gravity data from the GRACE (Gravity Recovery And Climate Experiment) twin satellites. The data consist of gravity data in the form of 10-day maximum values of 1° by 1° equivalent water height (EWH) in meters starting at 29 July 2002 and ending at 25 August 2010. Results focussing on Greenland show statistically significant mass loss interpreted as inland ice melt-off to the SE and NW with an acceleration in the melt-off occurring to the NW and a possible deceleration to the SE. Also, there are strong indications of a transition taking place in the mass loss in Greenland from mid-2004 to early 2006.
本文对GRACE (gravity Recovery And Climate Experiment，重力恢复与气候实验)双卫星的重力数据进行了初步分析。数据包括重力数据，从2002年7月29日开始到2010年8月25日结束，以米为单位的10天最高值为1°× 1°等效水高(EWH)。集中在格陵兰岛的结果显示，统计上显著的质量损失被解释为内陆冰在东南和西北方向的融化，其中西北方向的融化加速，东南方向的融化可能减速。此外，有强烈的迹象表明，从2004年中到2006年初，格陵兰岛的大规模损失正在发生转变。
{"title":"Greenland inland ice melt-off: Analysis of global gravity data from the GRACE satellites","authors":"A. Nielsen, O. Andersen, P. Svendsen","doi":"10.1109/MULTI-TEMP.2011.6005074","DOIUrl":"https://doi.org/10.1109/MULTI-TEMP.2011.6005074","url":null,"abstract":"This paper gives an introductory analysis of gravity data from the GRACE (Gravity Recovery And Climate Experiment) twin satellites. The data consist of gravity data in the form of 10-day maximum values of 1° by 1° equivalent water height (EWH) in meters starting at 29 July 2002 and ending at 25 August 2010. Results focussing on Greenland show statistically significant mass loss interpreted as inland ice melt-off to the SE and NW with an acceleration in the melt-off occurring to the NW and a possible deceleration to the SE. Also, there are strong indications of a transition taking place in the mass loss in Greenland from mid-2004 to early 2006.","PeriodicalId":254778,"journal":{"name":"2011 6th International Workshop on the Analysis of Multi-temporal Remote Sensing Images (Multi-Temp)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114252905","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 : 2011-07-12DOI: 10.1109/MULTI-TEMP.2011.6005082
C. Bustos, Osvaldo Campanella, K. Kpalma, F. Magnago, J. Ronsin
Currently remote sensing, based on satellite images is one of the most important source of information for multitemporal change detection. From all types of satellite images, the multispectral images present the advantage of characterizing the earth surface in different bands; each band provides different and useful information. In this work we propose a new methodology based on linear PCA to extract useful and meaningful information from signals provided by the remote sensing, and based on it, detect temporal changes Experiments based on images of the satellite CBERS-2B corresponding to the urban and peri urban region of Rio Cuarto of Córdoba state in Argentina have given satisfactory results in change detection.
{"title":"A method for change detection with multi-temporal satellite images based on Principal Component Analysis","authors":"C. Bustos, Osvaldo Campanella, K. Kpalma, F. Magnago, J. Ronsin","doi":"10.1109/MULTI-TEMP.2011.6005082","DOIUrl":"https://doi.org/10.1109/MULTI-TEMP.2011.6005082","url":null,"abstract":"Currently remote sensing, based on satellite images is one of the most important source of information for multitemporal change detection. From all types of satellite images, the multispectral images present the advantage of characterizing the earth surface in different bands; each band provides different and useful information. In this work we propose a new methodology based on linear PCA to extract useful and meaningful information from signals provided by the remote sensing, and based on it, detect temporal changes Experiments based on images of the satellite CBERS-2B corresponding to the urban and peri urban region of Rio Cuarto of Córdoba state in Argentina have given satisfactory results in change detection.","PeriodicalId":254778,"journal":{"name":"2011 6th International Workshop on the Analysis of Multi-temporal Remote Sensing Images (Multi-Temp)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120902244","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 : 2011-07-12DOI: 10.1109/MULTI-TEMP.2011.6005036
G. Satalino, D. Impedovo, A. Balenzano, F. Mattia
The objective of this paper is to report on the crop classification activities carried out during the first year of the Italian project “Use of COSMO-SkyMed data for LANDcover classification and surface parameters retrieval over agricultural sites” (COSMOLAND), funded by the Italian Space Agency. The project intends to contribute to the COSMO-SkyMed mission objectives in the agriculture and hydrology application domains. In particular, the objective of the classification activities is to assess the potential of multi-temporal series of X-band COSMO-SkyMed SAR data for crop classification. The selected agricultural site is located in the Capitanata plain close to the Foggia town (Puglia region, Southern Italy). Over this area, 8 Stripmap PingPong COSMO Sky-Med images at HH/HV polarization and at low incidence angle were acquired from April to August 2010. In the paper, a classification scheme based on the Maximum Likelihood algorithm is applied to the multi-temporal data set and its accuracy is assessed with respect to a reference map obtained by means of SPOT data.
{"title":"Land cover classification by using multi-temporal COSMO-SkyMed data","authors":"G. Satalino, D. Impedovo, A. Balenzano, F. Mattia","doi":"10.1109/MULTI-TEMP.2011.6005036","DOIUrl":"https://doi.org/10.1109/MULTI-TEMP.2011.6005036","url":null,"abstract":"The objective of this paper is to report on the crop classification activities carried out during the first year of the Italian project “Use of COSMO-SkyMed data for LANDcover classification and surface parameters retrieval over agricultural sites” (COSMOLAND), funded by the Italian Space Agency. The project intends to contribute to the COSMO-SkyMed mission objectives in the agriculture and hydrology application domains. In particular, the objective of the classification activities is to assess the potential of multi-temporal series of X-band COSMO-SkyMed SAR data for crop classification. The selected agricultural site is located in the Capitanata plain close to the Foggia town (Puglia region, Southern Italy). Over this area, 8 Stripmap PingPong COSMO Sky-Med images at HH/HV polarization and at low incidence angle were acquired from April to August 2010. In the paper, a classification scheme based on the Maximum Likelihood algorithm is applied to the multi-temporal data set and its accuracy is assessed with respect to a reference map obtained by means of SPOT data.","PeriodicalId":254778,"journal":{"name":"2011 6th International Workshop on the Analysis of Multi-temporal Remote Sensing Images (Multi-Temp)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122858200","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 : 2011-07-12DOI: 10.1109/MULTI-TEMP.2011.6005046
W. Philpot, T. Ansty
Invariant features are needed for atmospheric normalization of images pairs. Powerful statistical approaches now exist, designed to isolate unchanged pixels based on quantitatively evaluating the spectral correlation of pixels in image pairs. This suggests that it should be possible to reach similar results following an analytical path, and our hypothesis is that the derivation of an analytical procedure will yield some physical insight that is not directly accessible with a stochastic approach. In this paper we derive an analytical formula that relates PIFs to the radiometric properties of the scenes. The formula is then inverted to yield an estimate of the ratio of transmission spectra of the two images given the path radiance for each scene and a set of invariant features.
{"title":"Analytical description of pseudo-invariant features (PIFs)","authors":"W. Philpot, T. Ansty","doi":"10.1109/MULTI-TEMP.2011.6005046","DOIUrl":"https://doi.org/10.1109/MULTI-TEMP.2011.6005046","url":null,"abstract":"Invariant features are needed for atmospheric normalization of images pairs. Powerful statistical approaches now exist, designed to isolate unchanged pixels based on quantitatively evaluating the spectral correlation of pixels in image pairs. This suggests that it should be possible to reach similar results following an analytical path, and our hypothesis is that the derivation of an analytical procedure will yield some physical insight that is not directly accessible with a stochastic approach. In this paper we derive an analytical formula that relates PIFs to the radiometric properties of the scenes. The formula is then inverted to yield an estimate of the ratio of transmission spectra of the two images given the path radiance for each scene and a set of invariant features.","PeriodicalId":254778,"journal":{"name":"2011 6th International Workshop on the Analysis of Multi-temporal Remote Sensing Images (Multi-Temp)","volume":"114 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122070652","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 : 2011-07-12DOI: 10.1109/MULTI-TEMP.2011.6005100
C. Toté, Katia Beringhs, E. Swinnen, Gerard Govers
Environmental change is an important issue in the Andes region. The objectives of this research are to study NDVI dynamics in the Andes region based on time series analysis of SPOT-Vegetation and NOAA-AVHRR, and to recognize to which extent this variability can be attributed to either climatic variability or human induced impacts. Correlation analysis between NDVI and SPI were performed in order to identify the best lag per pixel. Trends in SDVI and SPI were investigated using linear least square regression. Significant vegetation trends are found in 46% of the area. Both NDVI time series lead to different results, but the coupling of vegetation and precipitation is more pronounced for the SPOT-Vegetation data.
{"title":"Monitoring environmental change in the Andes based on SPOT-VGT and NOAA-AVHRR time series analysis","authors":"C. Toté, Katia Beringhs, E. Swinnen, Gerard Govers","doi":"10.1109/MULTI-TEMP.2011.6005100","DOIUrl":"https://doi.org/10.1109/MULTI-TEMP.2011.6005100","url":null,"abstract":"Environmental change is an important issue in the Andes region. The objectives of this research are to study NDVI dynamics in the Andes region based on time series analysis of SPOT-Vegetation and NOAA-AVHRR, and to recognize to which extent this variability can be attributed to either climatic variability or human induced impacts. Correlation analysis between NDVI and SPI were performed in order to identify the best lag per pixel. Trends in SDVI and SPI were investigated using linear least square regression. Significant vegetation trends are found in 46% of the area. Both NDVI time series lead to different results, but the coupling of vegetation and precipitation is more pronounced for the SPOT-Vegetation data.","PeriodicalId":254778,"journal":{"name":"2011 6th International Workshop on the Analysis of Multi-temporal Remote Sensing Images (Multi-Temp)","volume":"141 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131944054","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 : 2011-07-12DOI: 10.1109/MULTI-TEMP.2011.6005034
F. Bovolo, C. Marín, L. Bruzzone
This paper proposes an approach to change detection in very high geometrical resolution (VHR) multitemporal SAR images for hot spot surveillance. The proposed approach is based on two concepts: i) the use of backscattering information extracted at different resolution levels; and ii) the use of prior information usually available on hot spots. Here the proposed approach is designed for the solution of a surveillance problem in port areas. To this end a data set was used made up of a pair of multitemporal VHR SAR images acquired by the COSMO-SkyMed (CSK®) constellation in spotlight mode over the commercial port of Livorno (Italy). These images define a complex change-detection problem due to the different kinds of changes on the ground, the high spatial resolution and the complexity of object backscattering in the considered area. Experimental results point out the effectiveness of the proposed approach.
{"title":"A multilevel approach to change detection for port surveillance with very high resolution SAR images","authors":"F. Bovolo, C. Marín, L. Bruzzone","doi":"10.1109/MULTI-TEMP.2011.6005034","DOIUrl":"https://doi.org/10.1109/MULTI-TEMP.2011.6005034","url":null,"abstract":"This paper proposes an approach to change detection in very high geometrical resolution (VHR) multitemporal SAR images for hot spot surveillance. The proposed approach is based on two concepts: i) the use of backscattering information extracted at different resolution levels; and ii) the use of prior information usually available on hot spots. Here the proposed approach is designed for the solution of a surveillance problem in port areas. To this end a data set was used made up of a pair of multitemporal VHR SAR images acquired by the COSMO-SkyMed (CSK®) constellation in spotlight mode over the commercial port of Livorno (Italy). These images define a complex change-detection problem due to the different kinds of changes on the ground, the high spatial resolution and the complexity of object backscattering in the considered area. Experimental results point out the effectiveness of the proposed approach.","PeriodicalId":254778,"journal":{"name":"2011 6th International Workshop on the Analysis of Multi-temporal Remote Sensing Images (Multi-Temp)","volume":"2018 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130633429","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 : 2011-07-12DOI: 10.1109/MULTI-TEMP.2011.6005068
C. Vaduva, T. Costachioiu, C. Patrascu, I. Gavat, V. Lazarescu, M. Datcu
With a continuous increase in the number of Earth Observation satellites, leading to the development of satellitar image time series (SITS), the number of algorithms for land cover analysis and monitoring has greatly expanded. This paper offers a new perspective in dynamic classification for SITS. Four similarity measures (correlation coefficient, Kullback-Leibler (KL) divergence, conditional information, normalized compression distance (NCD)) based on image pairs from the data are employed, resulting in a series of maps describing different types of changes observed in the original series. The proposed algorithm performs a classification of the newly developed time series using a Latent Dirichlet Allocation model (LDA). This statistical method was originally used for text classification, thus requiring a word, document, corpus analogy with the elements inside the image. The experimental results were computed using 11 Landsat images over the city of Bucharest and surrounding areas.
{"title":"Classification of dynamic evolutions from satellitar image time series based on similarity measures","authors":"C. Vaduva, T. Costachioiu, C. Patrascu, I. Gavat, V. Lazarescu, M. Datcu","doi":"10.1109/MULTI-TEMP.2011.6005068","DOIUrl":"https://doi.org/10.1109/MULTI-TEMP.2011.6005068","url":null,"abstract":"With a continuous increase in the number of Earth Observation satellites, leading to the development of satellitar image time series (SITS), the number of algorithms for land cover analysis and monitoring has greatly expanded. This paper offers a new perspective in dynamic classification for SITS. Four similarity measures (correlation coefficient, Kullback-Leibler (KL) divergence, conditional information, normalized compression distance (NCD)) based on image pairs from the data are employed, resulting in a series of maps describing different types of changes observed in the original series. The proposed algorithm performs a classification of the newly developed time series using a Latent Dirichlet Allocation model (LDA). This statistical method was originally used for text classification, thus requiring a word, document, corpus analogy with the elements inside the image. The experimental results were computed using 11 Landsat images over the city of Bucharest and surrounding areas.","PeriodicalId":254778,"journal":{"name":"2011 6th International Workshop on the Analysis of Multi-temporal Remote Sensing Images (Multi-Temp)","volume":"16 9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125398926","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 : 2011-07-12DOI: 10.1109/MULTI-TEMP.2011.6005056
J. Lambert, A. Jacquin, J. Denux, V. Chéret
In Europe, the 2003 summer heat wave damaged forested areas. The purpose of this study is to compare two methods to analyse time series of NDVI images for monitoring forest decline. The first method is based on phenological indicator linked to spring vegetation activity, and on the analysis of its trend. The second method (BFAST) allows extracting the trend by decomposition of NDVI time series into trend, seasonal and remainder components. The two approaches show similar results for trends estimates. The main advantage of BFAST is its capability to detect breakpoints in the linear trend which highlights the impact of the exceptional climatic conditions in 2003 on forest stands development.
{"title":"Comparison of two remote sensing time series analysis methods for monitoring forest decline","authors":"J. Lambert, A. Jacquin, J. Denux, V. Chéret","doi":"10.1109/MULTI-TEMP.2011.6005056","DOIUrl":"https://doi.org/10.1109/MULTI-TEMP.2011.6005056","url":null,"abstract":"In Europe, the 2003 summer heat wave damaged forested areas. The purpose of this study is to compare two methods to analyse time series of NDVI images for monitoring forest decline. The first method is based on phenological indicator linked to spring vegetation activity, and on the analysis of its trend. The second method (BFAST) allows extracting the trend by decomposition of NDVI time series into trend, seasonal and remainder components. The two approaches show similar results for trends estimates. The main advantage of BFAST is its capability to detect breakpoints in the linear trend which highlights the impact of the exceptional climatic conditions in 2003 on forest stands development.","PeriodicalId":254778,"journal":{"name":"2011 6th International Workshop on the Analysis of Multi-temporal Remote Sensing Images (Multi-Temp)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125306343","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 : 2011-07-12DOI: 10.1109/MULTI-TEMP.2011.6005081
Stien Heremans, J. Van Orshoven
Timely and accurate information on the location and the extent of land use types is high up the agenda of several governmental and scientific organizations. Remote sensing, through image classification at the sub-pixel level, is an attractive source of this type of information. The remote sensing community has recognized the multilayer perceptron (MLP) as a popular machine learning technique for performing land use classifications, both at the pixel and at the sub-pixel level. However, theoretical advances in the machine learning community are not easily adopted by the classification practice. An example is the continued use of the gradient descent algorithm for MLP training. In this paper, the accuracy of this standard first order learning algorithm was compared to that of five alternative, second order learning algorithms for performing a sub-pixel classification of land use in Flanders. The result are clear: all second order algorithms perform markedly better than gradient descent, thereby illustrating the importance of translating theoretical advances in MLP training to the classification practice.
{"title":"Effect of the learning algorithm on the accuracy of sub-pixel land use classifications with multilayer perceptrons","authors":"Stien Heremans, J. Van Orshoven","doi":"10.1109/MULTI-TEMP.2011.6005081","DOIUrl":"https://doi.org/10.1109/MULTI-TEMP.2011.6005081","url":null,"abstract":"Timely and accurate information on the location and the extent of land use types is high up the agenda of several governmental and scientific organizations. Remote sensing, through image classification at the sub-pixel level, is an attractive source of this type of information. The remote sensing community has recognized the multilayer perceptron (MLP) as a popular machine learning technique for performing land use classifications, both at the pixel and at the sub-pixel level. However, theoretical advances in the machine learning community are not easily adopted by the classification practice. An example is the continued use of the gradient descent algorithm for MLP training. In this paper, the accuracy of this standard first order learning algorithm was compared to that of five alternative, second order learning algorithms for performing a sub-pixel classification of land use in Flanders. The result are clear: all second order algorithms perform markedly better than gradient descent, thereby illustrating the importance of translating theoretical advances in MLP training to the classification practice.","PeriodicalId":254778,"journal":{"name":"2011 6th International Workshop on the Analysis of Multi-temporal Remote Sensing Images (Multi-Temp)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2011-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127178718","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：