Paper Image

Detecting tree pith from wood slices

Published on:

2 April 2024

Primary Category:

Computer Vision and Pattern Recognition

Paper Authors:

Henry Marichal,

Diego Passarella,

Gregory Randall


Key Details

Estimates local orientation of rings & other patterns in wood slices

Optimizes function to find most central point based on those patterns

Enhances technique for cases without clear concentric rings

Trains deep learning model to detect pith

Outperforms state-of-the-art methods; runs real-time

AI generated summary

Detecting tree pith from wood slices

This paper introduces methods to automatically detect the pith (center point) in images of tree slices. The methods analyze the concentric ring structure and other visual patterns to optimize a function that identifies the central point. Three approaches are presented: 1) APD analyzes local orientation of rings/patterns, 2) APD-PCL enhances APD for cases without clear rings, and 3) APD-DL uses a deep learning model. All methods are tested on diverse tree species and images. The proposed techniques outperform previous methods and can run in real-time. A new dataset of 119 gymnosperm and 9 angiosperm images is also released.

Answers from this paper


No comments yet, be the first to start the conversation...

Sign up to comment on this paper

Sign Up