Spectral behavior of some modal soil profiles from São Paulo State, Brazil

Remote sensing has a high potential for environmental evaluation. However, a necessity exists for a better understanding of the relations between the soil attributes and spectral data. The objective of this work was to analyze the spectral behavior of some soil profiles from the region of Piracicaba, São Paulo State, using a laboratory spectroradiometer (400 to 2500 nm). The relations between the reflected electromagnetic energy and the soil physical, chemical and mineralogical attributes were analyzed, verifying the spectral variations of soil samples in depth along the profiles with their classification and discrimination. Sandy soil reflected more, presenting a spectral curve with an ascendant form, opposite to clayey soils. The 1900 nm band discriminated soil with 2:1 mineralogy from the 1:1 and oxidic soils. It was possible to detect the presence of kaolinite, gibbsite, hematite and goethite in the soils through the descriptive aspects of curves, absorption features and reflectance intensity. A relation exists between the weathering stage and spectral data. The evaluation of the superficial and subsuperficial horizon samples allowed characterizing and discriminating the analytical variability of the profile, helping to soil distinguishing and classification.

QuMinS: fast and scalable querying, mining and summarizing multi-modal databases

Given a large image set, in which very few images have labels, how to guess labels for the remaining majority? How to spot images that need brand new labels different from the predefined ones? How to summarize these data to route the user’s attention to what really matters? Here we answer all these questions. Specifically, we propose QuMinS, a fast, scalable solution to two problems: (i) Low-labor labeling (LLL) – given an image set, very few images have labels, find the most appropriate labels for the rest; and (ii) Mining and attention routing – in the same setting, find clusters, the top-'N IND.O' outlier images, and the 'N IND.R' images that best represent the data. Experiments on satellite images spanning up to 2.25 GB show that, contrasting to the state-of-the-art labeling techniques, QuMinS scales linearly on the data size, being up to 40 times faster than top competitors (GCap), still achieving better or equal accuracy, it spots images that potentially require unpredicted labels, and it works even with tiny initial label sets, i.e., nearly five examples. We also report a case study of our method’s practical usage to show that QuMinS is a viable tool for automatic coffee crop detection from remote sensing images.