Due to the different imaging mechanisms, thermal infrared images are significantly different from the most widely used visible light images, which are mainly reflected in the following aspects.

(1) The noise of the thermal infrared image is relatively strong. The thermal motion of molecules in nature and the imperfection of infrared imaging system will introduce many kinds of noise, such as thermal noise, shot noise, 1/f noise, photon and electronic fluctuation noise and so on. These complex noises make the signal-to-noise ratio of thermal infrared images lower than that of ordinary TV images, so image denoising is often an important part of thermal infrared image processing.

(2) The contrast of the thermal infrared image is low and the image is blurred. Infrared detectors can detect a wide temperature range, but the actual scene temperature range is relatively small, so the gray distribution of the image is relatively concentrated, and the contrast with the background is low. Due to the thermal balance of the scene, long wavelength, long transmission distance, and atmospheric attenuation, thermal infrared images have strong spatial correlation and blurred visual effects.

(3) Thermal infrared images lack color and rich texture details. Thermal infrared images are grayscale images with no color information. In terms of texture details, the infrared images of the daytime scene are better than those of the night scene, because different objects in the daytime scene absorb and reflect solar radiation differently, and the emissivity and temperature distribution are quite different.

(4) The detection capability and spatial resolution of the thermal infrared imaging system are lower than those of the visible light charge coupled device (CCD) array, making the resolution of the thermal infrared image lower than that of the visible light image.

(5) Due to the inconsistency of the response characteristics of the infrared detection unit and the defects of the optical-mechanical scanning system, etc., the thermal infrared image will produce non-uniformity, which is reflected in the fixed pattern noise, crosstalk, and distortion of the image.