Caracterização de cicatrizes de deslizamentos por processamento de dados TM Landsat em Caraguatatuba - SP

Landslide inventory using image fusion techniques in Brazil

“…6 confirms that these bands, when used together, are quite capable of differentiating soil from vegetation due to the high contrast existing between the two targets in the correspondent intervals from these bands in the electromagnetic spectrum (Bowker et al, 1985;Sestini and Florenzano, 2004).…”

“…The RMSE found in the satellite images registration were 23.53 and 36.14 m, for the X and Y directions, respectively, while the total and Sestini and Florenzano (2004). Silva (1988), that is, the excessive concentration of points in a given area and the absence in others is the main source of geometric distortions in the application of geometric correction methods.…”

“…Marcelino et al (2003) also found similar results in landslides areas at Sierra General, in the state of Santa Catarina, pointing bands 3 and 4 as the most appropriated to the identification of landslides. Sestini and Florenzano (2004) mention that the ETM+/Landsat composition 473 is also appropriated in the classification of landslide scars, giving that a great contrast can be observed between areas of exposed soil adjacent to vegetation areas. Moreover, Florenzano (1993) highlights the ability of the composition TM 453 in emphasizing the morphology, the wetlands and drainage network.…”

“…Therefore, based on the prior researches performed by Florenzano (1993), Sestini (1999), Marcelino et al (2003) and Sestini and Florenzano (2004), the band 3 (Red 630-690 nm), band 4 (NIR 760-900 nm), and band 5 (MIR 1550-1750 nm) of the ETM+ sensor were selected to identify landscape scars for the Caraguatatuba inventory. Table 4 shows that all six fusion techniques under assessment performed with correlations above 80%.…”

Remote Sensing of Land Degradation: Experiences from Latin America and the Caribbean

“…Photo‐interpretation, geometric and radiometric corrections of satellite imagery, and contrast enhancement for visual interpretation are techniques commonly implemented, as evidenced in the works of Sestini and Florenzano (2004), who applied image transforms like Landsat TM band ratios and principal component analysis before visual interpretation of landslides; Marcelino et al (2003) working with image transformations (e.g., RGB to IHS, principal components analysis and wavelet fusion) of Landsat TM data to discriminate landslide scars; and Ochoa Tejeda and Parrot (2007) using spectral indices from IKONOS imagery and a DTM for automated detection of landslide traces.…”