Scale estimation of objects using template matching

Leopoldo N. Gaxiola; Victor H. Diaz-Ramirez; Juan J. Tapia

doi:10.1117/12.2238016

Translator Disclaimer

14 September 2016 Scale estimation of objects using template matching

Leopoldo N. Gaxiola, Victor H. Diaz-Ramirez, Juan J. Tapia

Author Affiliations +

Proceedings Volume 9970, Optics and Photonics for Information Processing X; 99701K (2016) https://doi.org/10.1117/12.2238016
Event: SPIE Optical Engineering + Applications, 2016, San Diego, California, United States

Abstract

Scale estimation of objects is a challenging problem in image processing. This work presents a novel method to detect and estimate the scaling factor of a target in an observed scene corrupted with additive noise and clutter. Given a set of available views of the target the proposed method is able to detect the target and estimate its scaling factor using a template matched filters and a scale pyramidal representation. The performance of the proposed method is evaluated in synthetic and real-life scenes in different pattern recognition applications. The obtained results are characterized in terms of objective metrics.

Citation Download Citation

Leopoldo N. Gaxiola, Victor H. Diaz-Ramirez, and Juan J. Tapia "Scale estimation of objects using template matching", Proc. SPIE 9970, Optics and Photonics for Information Processing X, 99701K (14 September 2016); https://doi.org/10.1117/12.2238016

ACCESS THE FULL ARTICLE