Mapping of the aquatic plants infestation in reservoirs using multiscale image and artificial neural networks

Revista Brasileira de Cartografia

关键词: Multiscale Analysis;    Artificial Neural Networks;    Mapping of Aquatic Plants in Reservoirs;    Remote Sensing.;   

DOI  :  

来源: DOAJ

【 摘 要 】

In past few years, infestations of aquatic plants in reservoirs have been studied as an effect of the environmental unbalance caused by pollution and damming of rivers. The excessive amount of plants, deriving from this unbalance, makes navigation and the production of electricity difficult. This kind of occurrence, as well as the appearance of some substances in the water, cause changes in the water radiance detected by satellite sensors. Thus, processing techniques of remote sensing and data analysis may be used as a complementary data source to give information related to the degree of infestation of these plants in reservoirs. So, this research aimed at verifying the influence of the spatial resolution of multispectral images in the detection and mapping of areas infested by aquatic plants in a small reservoir, using multiscale analysis procedures and supervised classification by artificial neural networks. Multispectral images IKONOS of the Salto Grande reservoir, in the city of Americana-SP were used. At first, a multiscale image was generated, resulting in images of 8, 16 and 32 meters of spatial resolution. In the classification of these images, the input data for artificial neural networks was constituted of multispectral images IKONOS, texture data, and a vegetation index image and allowed represent the spectral variations of the water body and infested areas of aquatic plants in the various levels of spatial resolution. Furthermore, an analysis was made comparing classified multiscale images by using cross tabulation, which permits comparing the results obtained in the multiscale levels. As result is pointed out that the thematic maps were representative for the 4 levels of spatial resolution. The method used was adequate to map the spectral variation of the water body and to detect infested areas of aquatic plants in the various levels of resolution of the image. The classification by neural network using parameters defined for the original image and applied in the training of the scheme adopted for the different resolution levels was satisfactory.

【 授权许可】

Unknown