Connected Filtering and Segmentation Using Component Trees

Real-time landing place assessment in man-made environments

“…In this section, we first provide a brief overview of component tree segmentation mostly using the formulation of [29]. We then discuss extensions we use for finding obstacle-free candidate landing sites.…”

“…To this end, we introduce a Hough-like voting scheme into the algorithm that builds our spatio temporal MSERs to guarantee that they have planar sides. Unlike other algorithms that perform global image segmentation [7,45], ours is highly efficient and runs at 5 Hz on 320×240 images using commodity hardware, in part because our MSERs are found using an efficient component tree representation [29].…”

Segmentation of Complex Images Based on Component-Trees: Methodological Tools

“…As mentioned above, component-trees have been considered for the development of image segmentation methods, mainly in the field of (bio)medical imaging, and in particular for: dermatological data [10], wood micrographs [6], cerebral MRI [4], CT/MR angiography [19], or confocal microscopy [12].…”

“…Among the methods which fully use component-trees for segmentation purpose, some can consider complex (i.e. multiple) knowledge [10] or can be run without userinteraction [6], but none of them is able to determine the correct pieces of knowledge required to perform segmentation without guidance of the user. Moreover, such methods only deal with simple-shape objects (circular or elliptical 2-D features in [6,10]).…”

“…Thanks to efforts devoted to its efficient computation [2,14,7,11,8] or its use in complex knowledge handling procedures [15], component-trees have been considered for the design of various kinds of grey-level image processing methods, including image filtering and segmentation [5,6,17,12,16,10], video segmentation [14], image registration [7], image compression [14], or image retrieval [9,1].…”