Colored three-dimensional reconstruction of vehicular thermal infrared images

Shaoyuan Sun; Henry Leung; Zhenyi Shen

doi:10.1117/1.OE.54.6.063102

11 June 2015 Colored three-dimensional reconstruction of vehicular thermal infrared images

Shaoyuan Sun, Henry Leung, Zhenyi Shen

Author Affiliations +

Optical Engineering, Vol. 54, Issue 6, 063102 (June 2015). https://doi.org/10.1117/1.OE.54.6.063102

Abstract

Enhancement of vehicular night vision thermal infrared images is an important problem in intelligent vehicles. We propose to create a colorful three-dimensional (3-D) display of infrared images for the vehicular night vision assistant driving system. We combine the plane parameter Markov random field (PP-MRF) model-based depth estimation with classification-based infrared image colorization to perform colored 3-D reconstruction of vehicular thermal infrared images. We first train the PP-MRF model to learn the relationship between superpixel features and plane parameters. The infrared images are then colorized and we perform superpixel segmentation and feature extraction on the colorized images. The PP-MRF model is used to estimate the superpixel plane parameter and to analyze the structure of the superpixels according to the characteristics of vehicular thermal infrared images. Finally, we estimate the depth of each pixel to perform 3-D reconstruction. Experimental results demonstrate that the proposed method can give a visually pleasing and daytime-like colorful 3-D display from a monochromatic vehicular thermal infrared image, which can help drivers to have a better understanding of the environment.

Citation Download Citation

Shaoyuan Sun, Henry Leung, and Zhenyi Shen "Colored three-dimensional reconstruction of vehicular thermal infrared images," Optical Engineering 54(6), 063102 (11 June 2015). https://doi.org/10.1117/1.OE.54.6.063102

Published: 11 June 2015

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available