Paper
12 December 2018 A model of point target temperature estimation based on multispectral infrared
Author Affiliations +
Proceedings Volume 10846, Optical Sensing and Imaging Technologies and Applications; 108461P (2018) https://doi.org/10.1117/12.2505149
Event: International Symposium on Optoelectronic Technology and Application 2018, 2018, Beijing, China
Abstract
At present, when the infrared sensor detects the targets remotely, the target appears as a spot in the image plane, the geometry information is difficult to obtain, and the surface brightness temperature becomes an effective feature for the target recognition. However, due to the long distance of the target, the weak signal and the complex transmission path, the temperature features are difficult to extract accurately, which brings great uncertainty to the target recognition. Based on the principle of multi-spectral infrared radiation temperature measurement, this paper establishes a BP network model to estimate the point target temperature. Experiments show that the accuracy of extracting the faint targets temperature characteristics can be effectively improved, which shows great support for target recognition.
© (2018) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Zheng Liu, Hongxia Mao, Congming Dai, and Heli Wei "A model of point target temperature estimation based on multispectral infrared", Proc. SPIE 10846, Optical Sensing and Imaging Technologies and Applications, 108461P (12 December 2018); https://doi.org/10.1117/12.2505149
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Feature extraction

Temperature metrology

Infrared radiation

Signal to noise ratio

Neural networks

Target recognition

Error analysis

Back to Top