“Researchers have made new breakthroughs in the field of infrared imaging detection technology to address the significant harm caused by gas leakage accidents. Traditional gas infrared images have low contrast and blurry edges, making recognition difficult. Therefore, experts have proposed a low contrast gas infrared image leak detection method based on mixture Gaussian background modeling. This method first uses time-domain adaptive inter frame filtering algorithm for noise reduction and detail preservation in the preprocessing stage, and then combines spatial information and gradient information constraints to construct a spatiotemporal mixture Gaussian background model to preliminarily extract the foreground of leaked gas targets. To further improve accuracy, researchers also utilized an improved fast and robust fuzzy C-means clustering method to achieve adaptive segmentation of gas regions. The experimental results show that at a leakage distance of 5 meters, the accuracy of the detection algorithm is as high as 92.3%~96.3%, effectively compensating for the problem of gas voids and reducing interference from other moving objects. This research achievement not only significantly improves the anti-interference ability and regional segmentation accuracy of gas leak detection, but also provides new ideas and directions for the development and application of infrared imaging detection technology.”