Índice de cobertura verde para imagens de altíssima resolução.

The low altitude aerial images are becoming more common every day due to low cost and ease of use of platforms such as remotely piloted aircraft. The potential application of this type of data is very high. One example is the precision agriculture, a farming management concept based on observing, measuring and responding to inter and intra-field variability in crops, an activity than can greatly benefit from this technology. The low altitude of image acquisition allows very high level of scene details but aggravates problems such as lighting variation and image deformation. In addition, often common cameras are used in different situations altitude, inclination, lighting and camera setup. These specific characteristics in relation to the orbital data require development of new methods and approaches to exploit the potential of data and to mitigate problems and limitations. In this work, we present a proposal for a method that provides a green coverage index. It reflects the green pixels density in an area. The proposed index has similar applicability to vegetation indices but does not require near-infrared data, not available in common cameras. We show problems especially related to agriculture applications, present initial test results discuss the possibilities and limitations of the method.

Saved in:
Bibliographic Details
Main Authors: NEVES, M. C., NEVES JÚNIOR, O. R., LUIZ, A. J. B., SANCHES, I. D.
Other Authors: MARCOS CORREA NEVES, CNPMA; OTHON DA ROCHA NEVES JUNIOR, UFSC; ALFREDO JOSE BARRETO LUIZ, CNPMA; IEDA DEL'ARCO SANCHES, INPE.
Format: Anais e Proceedings de eventos biblioteca
Language:pt_BR
por
Published: 2018-01-08
Subjects:Image processing, Processamento de imagem, Visão computacional, Índice de área foliar, Imagem de satélite., Sensoriamento remoto, Agricultura., Remote sensing, Image analysis, Computer vision, Agriculture, Leaf area index,
Online Access:http://www.alice.cnptia.embrapa.br/alice/handle/doc/1084549
Tags: Add Tag
No Tags, Be the first to tag this record!