Classication of Agricultural Fields in Satellite Images Using Two-Dimensional Hidden Markov Models
Image segmentation is a key competence for many real life applications such as precision agriculture. In this work we present an approach to classify agricultural fields in noisy satellite images. We start with the Markovian neighborhood hypothesis from where on we derive a general two-dimension...
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | conferenceObject |
| Lenguaje: | Inglés |
| Publicado: |
2021
|
| Materias: | |
| Acceso en línea: | http://hdl.handle.net/11086/21531 |
| Aporte de: |
| Sumario: | Image segmentation is a key competence for many real life
applications such as precision agriculture. In this work we present an
approach to classify agricultural fields in noisy satellite images. We start
with the Markovian neighborhood hypothesis from where on we derive a
general two-dimensional hidden Markov model (2D-HMM). To make the
2D-HMM feasible we apply the Path-Constrained Variable-State Viterbi
Algorithm (PCVSVA) which allows us to approximate the optimal hidden state map. We evaluate the PCVSVA for a Landsat image of the
province of C´ordoba, Argentina and a synthetic satellite image. In both
cases we use Cohen’s κb coefficient to compare the PCVSVA and the solution obtained by maximum likelihood (ML) to show the effectiveness
of 2D-HMM of solving image segmentation tasks. |
|---|