Texture analysis of tissue samples using DTOCS and approximated path DTOCS

Autor:	Niina Päivinen, Jouni Paakkonen, Pekka Toivanen
Rok vydání:	2009
Předmět:	Image texture Pixel Computer science business.industry Path (graph theory) Cognitive neuroscience of visual object recognition Computer vision Artificial intelligence business Distance transform Curved space Texture (geology) Grayscale
Zdroj:	2009 IEEE 5th International Conference on Intelligent Computer Communication and Processing.
DOI:	10.1109/iccp.2009.5284765
Popis:	The aim of this paper is to apply distance transform on curved space (DTOCS) to grayscale images of human tissue samples. Also a new approximated path DTOCS (APDTOCS), which approximates distances on curved space using only the end points of the paths, is proposed. APDTOCS reveals more detail than standard DTOCS especially in parts of the image where the texture is complex and has a lot of small details. APDTOCS is faster and simpler to implement than DTOCS. Texture analysis of tissue sample, which reveals regions of interest, is needed because histopathologic recognition is currently really slow when the whole tissue sample is examined.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::f79e70651419a86072bf2eb05344d7d6 https://doi.org/10.1109/iccp.2009.5284765 Zobrazit plný text záznamu