Research on the denoising method of infrared thermogram during rock fracture

Qiangqiang Gao,Liqiang Ma,Liqiang Ma,Wei Liu,Naseer Muhammad Khan,Naseer Muhammad Khan,Guanghui Cao,Yumiao Fang,Hui Wang

DOI: https://doi.org/10.1016/j.infrared.2023.104651

IF: 2.997

2023-06-01

Infrared Physics & Technology

Abstract:In order to overcome the defects of low contrast and low signal-to-noise ratio of the infrared thermal image of rock (ITIR), a joint denoising method of additive and multiplicative noise is proposed for infrared radiation thermal image of rock bearing fracture process. ITIR is de-noised using background phase subtraction, adaptive median filtering, and wavelet multi-scale decomposition threshold denoising Firstly, the noise types in ITIR are analyzed, the basic principles and implementation steps of the denoising method are described, and the denoising effects of different combinations of wavelet basis functions with different decomposition scales in the wavelet threshold denoising process are simulated. Finally, the denoising effect was evaluated in terms of peak signal-to-noise ratio (PSNR), root mean square error (RMSE), and blind/referenceless image spatial quality evaluator (BRISQE). The results show that the PSNR of ITIR is increased by 15.01 % based on removing additive noise. The image quality scores increased by 49.54 % before and after overall denoising. The studies have significance for IR signal extraction and denoising in rocks and solid materials.

optics,physics, applied,instruments & instrumentation