Aperfeiçoamento do monitoramento do uso e cobertura do solo com dados MODIS a partir da utilização de um diagrama de transição de estados.
This paper describes a methodology for the improvement of current capabilities for long-term monitoring of land use. A previously published methodology for the classification of land use and land cover using data from the Moderate Resolution Imaging Spectroradiometer (MODIS) is presented, which takes advantage of that sensor’s high temporal frequency, extensive coverage, and extremely low costs for data acquisition. Then, a state-transition diagram is proposed in order to enhance that methodology by incorporating extra information in the form of a priori knowledge of land use dynamics, thus making it possible for the classification algorithm to take advantage of the temporal context of each point within the region of analysis. Finally, it is argued how such an approach can significantly improve the classification process not only in terms of overall accuracy but also in terms of better describing the land use and land cover classes found within the area of study.
| Main Authors: | , , , , |
|---|---|
| Other Authors: | |
| Format: | Artigo em anais e proceedings biblioteca |
| Language: | por |
| Published: |
2008-01-17
|
| Subjects: | Sensoriamento Remoto, Uso da Terra, Remote sensing, Land use change, |
| Online Access: | http://www.alice.cnptia.embrapa.br/alice/handle/doc/339188 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This paper describes a methodology for the improvement of current capabilities for long-term monitoring of land use. A previously published methodology for the classification of land use and land cover using data from the Moderate Resolution Imaging Spectroradiometer (MODIS) is presented, which takes advantage of that sensor’s high temporal frequency, extensive coverage, and extremely low costs for data acquisition. Then, a state-transition diagram is proposed in order to enhance that methodology by incorporating extra information in the form of a priori knowledge of land use dynamics, thus making it possible for the classification algorithm to take advantage of the temporal context of each point within the region of analysis. Finally, it is argued how such an approach can significantly improve the classification process not only in terms of overall accuracy but also in terms of better describing the land use and land cover classes found within the area of study. |
|---|