Image foresting transform
In the practice of Image processing Alexandre X. Falca˜ o, Jorge Stolfi, and Roberto de Alencar Lotufo have created and proven that the Image Foresting Transform (IFT) can be used as a time saver in processing 2-D and 3-D images[1]. The transform uses a method used in machine learning that is known as Random forest which finds the "cheapest" way to go about applying image operators. The transform is a tweaked version of Dijkstra’s shortest-path algorithm[2] maximized for more than one input and operator. Method of using image processing operators in a way that depends on how they are connected and prioritizes them by the shortest graphical path between these operators being used based on the pixel characteristics such as gray scale, position, brightness or luminescence. These priorities could change depending on the region of the image being processed depending upon the characteristics of that region. This would make the priority of the operators to be changed. The robustness of the transform does come at a cost and uses a lot of storage for the code and the data being processed. The attractive aspect of this transform is that multiple operators can be used and run at the same time where otherwise they would not be able to[3].
Related topics
Watershed (image processing) Random forest Image processing
References
- ^ Falcao, A.X. Stolfi, J. de Alencar Lotufo, R. : "The image foresting transform: theory, algorithms, and applications", In IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 26, NO. 1, JANUARY 2004
- ^ E.W. Dijkstra, “A Note on Two Problems in Connexion with Graphs,” Numerische Mathematik, vol. 1, pp. 269-271, 1959
- ^ Jean Cousty, Gilles Bertrand, Laurent Najman, and Michel Couprie. Watershed cuts: thinnings, shortest-path forests and topological watersheds. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 32, NO. 5, MAY 2010 pp. 925–939.