Intelligent multi-spectral IR image segmentation

Published in SPIE Defense + Security, Pattern Recognition and Tracking XXVIII, 2017

Recommended citation: Thomas Lu, Andrew Luong, Stephen Heim, Maharshi Patel, Kang Chen, Tien-Hsin Chao, Edward Chow, Gilbert Torres, } "Intelligent multi-spectral IR image segmentation", Proc. SPIE 10203, Pattern Recognition and Tracking XXVIII, 1020303 (1 May 2017); doi: 10.1117/12.2262730; https://doi.org/10.1117/12.2262730 https://www.spiedigitallibrary.org/conference-proceedings-of-spie/10203/1/Intelligent-multi-spectral-IR-image-segmentation/10.1117/12.2262730.short?SSO=1

This article presents a neural network based multi-spectral image segmentation method. A neural network is trained on the selected features of both the objects and background in the longwave (LW) Infrared (IR) images. Multiple iterations of training are performed until the accuracy of the segmentation reaches satisfactory level. The segmentation boundary of the LW image is used to segment the midwave (MW) and shortwave (SW) IR images. A second neural network detects the local discontinuities and refines the accuracy of the local boundaries. This article compares the neural network based segmentation method to the Wavelet-threshold and Grab-Cut methods. Test results have shown increased accuracy and robustness of this segmentation scheme for multi-spectral IR images.

Download paper here

Recommended citation: Thomas Lu, Andrew Luong, Stephen Heim, Maharshi Patel, Kang Chen, Tien-Hsin Chao, Edward Chow, Gilbert Torres, } “Intelligent multi-spectral IR image segmentation”, Proc. SPIE 10203, Pattern Recognition and Tracking XXVIII, 1020303 (1 May 2017); doi: 10.1117/12.2262730; https://doi.org/10.1117/12.2262730

Leave a Comment