With rapid development in Industries and automated chemical plants leakage of gas is common issue. Gas leakage may lead to chemical and radioactive damage to the surrounding environment. Due to the nature of hazardous gases, human intervention is not possible, and it is difficult to identify the gas emission as most of the gases are colourless and odourless. Thus, leakage of gas emissions at earlier stage is most important to prevent from explosions. Thermal Imaging technology can be used to determine the severity level of the gas leakage. This paper presents the Pre-Processing techniques used for thermal images to remove unnecessary noise and to refine the image quality. The accuracy and efficiency of determining the source of gas emissions can be enhanced by pre-processing. The quality of the pre-processed image is evaluated using metrics like Peak Signal to Noise Ratio (PSNR), Structural Similarity Index (SSIM), Normalized Cross-Correlation (NCC) and Mean Square Error (MSE) for various filtering techniques.