A Hybrid Thermal-Visible Fusion for Outdoor Human Detection

• Main Authors: Ezrinda, Mohd Zahidee, Kamarul Hawari, Ghazali, Ren, Jinchang, Mohd Zuki, Salleh
• Format: Article
• Language: English
• Published: Universiti Teknikal Malaysia 2018
• Subjects: QA Mathematics; QA75 Electronic computers. Computer science; TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://umpir.ump.edu.my/id/eprint/21565/1/JTEC.pdf; http://umpir.ump.edu.my/id/eprint/21565/; http://journal.utem.edu.my/index.php/jtec/article/view/3580

Multisensory image fusion can be used to improve the visual interpretability of an image for further processing task. A hybrid thermal-visible image fusion is proposed in this paper to detect the target and produce an output image that had all the information from both sensors. The thermal target region was extracted using Niblack algorithm and some morphological operators. Then, the source images were decomposed at pixel level using Stationary Wavelet Transform (SWT). The appropriate fusion rule were chosen for low and high frequency components and finally the fused image was obtained from inverse SWT. Results show that the proposed method achieve better to include figure of merit than the other three methods.