Identification of shadows in the Atlantic Forest for preservation monitoring

Abstract

The presence of shadows in aerial images of forest areas can be useful for several tasks, such asdetection of emergent layer specimens, determination of bamboo proliferation or evaluation ofother forest structural parameters. Invariant features in images have special value since theprocess of capture of images involves considerable levels of variation. The aim of the presentstudy was to evaluate an algorithm based on an illumination invariant shadow ratio to segmentRGB aerial images of the Atlantic Forest in shadowed and non-shadowed areas. Some parametersof the algorithm were tuned with satisfactory results regarding the success rate comparing withmanual selection, and thus allowed obtaining a procedure to achieve an automatic process. Onthe other hand, the processing time was reasonably short, making this methodology a suitable optionfor rainforest conservation monitoring.