Haven't been able to find anything free and/or open-source to do this, would love to be referred to any packages or libraries anyone knows about, especially if they're available through Java or Python. Thanks!
The image texels represent physical texture elements that are at a constant depth, and imaged fully or partly depending on their mutual occlusions. Since the physical elements are themselves finite size objects, the image texels are regions, whose properties – such as geometric (e.g., area and shape), photometric (e.g., spatial color distribution), and topological (recursive embeddings of subtexel regions) properties – can be stochastically characterized.
Since the notion of image texture implies a large number of texels, and therefore
a large number of samples from the underlying pdf’s, reliable statistical estimation of these pdf’s is feasible. Estimation of the pdf of intrinsic texel properties can be combined with the analysis of other aspects of texture, including properties of the textured surface (e.g., plane orientation), and stochastic rules governing the placement of physical elements on the plane, to estimate the complete texture mode.
In a study An input image is represented by the segmentation tree, T , obtained using the multiscale segmentation algorithm.
The segmentation tree T contains all segments present in the image, including texels. They can be detected as a set of disjoint, similar subtrees rooted closest to the root of T . The texel subtrees are discovered by matching all possible node pairs in the segmentation tree, and selecting those node pairs closest to the root whose attributes match, and the same holds for their descendant nodes.
See the following text as a reference
Narendra Ahuja and Sinisa Todorovic, Extracting Texels in 2.1D Natural Textures, in Proc. 11th IEEE International Conference on Computer Vision (ICCV), Rio de Janeiro, Brazil, October 2007
Dear Paulo Raposo,
Look the link. May be useful.
chrome-extension://oemmndcbldboiebfnladdacbdfmadadm/http://digitalcommons.usu.edu/cgi/viewcontent.cgi?article=1125&context=ece_facpub
Regards, Shafagat
Thank you Dr's Awad and Mahmudova!
Also happy to hear from anyone further in case they know of existing open-source software that actually detects texels.
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Software