README.md

imtools

Tools and techniques for low-level image processing.

The source code provided herein implements various basic low-level image processing tools used in the experiments/analyses reported in the references listed below.

Description

Matlab tools for low-level image processing:

• derive: Differentiation functions
• feature: Feature extraction
• filter: Filtering functions
• graph: Graph and network analysis and representation
• surface: Surface analysis functions
• pyramid: Hierarchical pyramid analysis and decomposition
• sharpen: Sharpening and enhancing functions
• geometry: Basic geometrical tools
• kernel: Kernel operators
• morphology: Mathematical morphology
• propagation: Propagation and Fast Marching methods
• segmentation: Segmentation functions
• statistics: Basic statistical tools and operators
• texture: Texture analysis and representations
• misc: Miscellaneous functions

See Contents.m for the whole list of functions. See file Contents.html for browsable version.

Applications

Some examples of (replicable) applications are described on the following pages:

• amoeba superpixels for superpixel partitioning of multispectral images into elementary, connected, nearly uniform units.
• pseudo DEM for line networks extraction in digital images using geodesic propagation and flow simulation.

About

documentation	available at: https://gjacopo.github.io/imtools/
version	1.0 (non-active development)
since	2007
license	EUPL (cite the source code or any of the references below!)

References

• Soille P. and Grazzini J. (2011): IMAGE-2006 Mosaic: Geometric and radiometric consistency of input imagery, Publications Office of the European Union, doi:10.2788/50967.
• Grazzini J., Dillard S., and Soille P. (2010): A new generic method for the semi-automatic extraction of river and road networks in low and mid-resolution satellite images, in Proc. of SPIE - Image and Signal Processing for Remote Sensing XVI, vol. 7830, pp. 7830071-10, doi:10.1117/12.865052.
• Grazzini J., Dillard S., and Prasad L. (2010): Simultaneous hierarchical segmentation and vectorization of satellite images through combined non-uniform data sampling and anisotropic triangulation, in Proc. of SPIE - Image and Signal Processing for Remote Sensing XVI, vol. 7830, pp. 78300F-13, doi:10.1117/12.865047.
• Dillard S., Prasad L., and Grazzini J. (2010): Region and edge-adaptive sampling and boundary completion for segmentation, in Proc AVC, Lecture Notes in Computer Science, vol. 6454, pp.64-74, doi:10.1007/978-3-642-17274-8_7.
• Grazzini J., Dillard S., and Soille P. (2010): Multichannel image regularisation using anisotropic geodesic filtering, in Proc. IEEE ICPR, pp. 2664-2667, doi:10.1109/ICPR.2010.653.
• Grazzini J. and Soille P. (2010): Iterative ramp sharpening for structure/signature-preserving simplification of images, in Proc. IEEE ICPR, pp. 4586-4589, doi:10.1109/ICPR.2010.1120.
• Grazzini J. and Soille P. (2009): Image filtering based on locally estimated geodesic functions, in Proc. VISIGRAPP, Communications in Computer and Information Science, vol. 24, pp.123-124, doi:10.1007/978-3-642-10226-4_10.
• Soille P. and Grazzini J. (2009): Constrained connectivity and transition regions, in Proc. ISMM, Lecture Notes in Computer Science, vol. 5720, pp.59-69, doi:10.1007/978-3-642-03613-2_6.
• Grazzini J. and Soille P. (2009): Edge-preserving smoothing using a similarity measure in adaptive geodesic neighbourhoods, Pattern Recognition, 42(10):2306-2316, doi:10.1016/j.patcog.2008.11.004.
• Grazzini J. and Soille P. (2008): Adaptive morphological filters using similarities based on geodesic time, in Proc. DGCI, Lecture Notes in Computer Science, vol. 4992, pp.519-528, doi:10.1007/978-3-540-79126-3_46.
• Grazzini J., Soille P., and Bielski C. (2007): On the use of geodesic distances for spatial interpolation, in Proc GeoComputation, 7C3, uri:http://publications.jrc.ec.europa.eu/repository/handle/JRC37427.
• Bielski C., Grazzini J., and Soille P. (2007): The Little algorithm that grew: Scaling the morphological image compositing algorithm to meet the chanllenges of processing large image data sets, in Proc GeoComputation, 7C3, uri:http://publications.jrc.ec.europa.eu/repository/handle/JRC44083.
• Bielski C., Grazzini J., and Soille P. (2007): Automated morphological image composition for mosaicing large image data sets, in Proc IEEE IGARSS, doi:10.1109/IGARSS.2007.4423743.
• Grazzini J. and Soille P. (2007): Improved morphological interpolation of elevation contour data with generalised geodesic propagations, in Proc. CAIP, Lecture Notes in Computer Science, vol. 4673, pp.742-750, doi:10.1007/978-3-540-74272-2_92.
• Soille P. and Grazzini J. (2007): Extraction of river networks from satellite images by combining morphology and hydrology, in Proc. CAIP, Lecture Notes in Computer Science, vol. 4673, pp.636-644, doi:10.1007/978-3-540-74272-2_79.