In view of the limitations of the low efficiency of gasoline pipeline leak detection and the inability to accurately segment the edge of the leak region, a gasoline pipeline leak detection method is proposed based on hyperspectral image and deep learning. Firstly, the characteristic spectral bands of the two types of gasoline under the background of soil and water were extracted. The continuous projection algorithm was used to reduce the dimensionality of gasoline hyperspectral image data. The gasoline reflectivity was taken as input, and the root-mean-square error was the regression parameter used to obtain 18 characteristic bands near the gasoline reflection peak. Image rotation Angle, horizontal or vertical inversion, and random noise injection into the image are used to expand the dataset sample. Secondly, the Unet hyperspectral image semantic segmentation model is improved, and the network encoder part of Unet is replaced with a dense connection module to strengthen the information exchange between different levels, reduce the computational load, and improve the model detection speed. The spectral attention mechanism module is introduced to make the model pay attention to gasoline image space and spectral features and improve the model detection accuracy. The concept of an inactivation layer is introduced to reduce the complexity of the network by temporarily shutting down some neurons in the network. At the same time, an appropriate time point is set in the training process to implement the early stop strategy to prevent overfitting. Finally, the ablation experiment and comparison experiment were carried out. The results of ablation experiments validate the effectiveness of the dense connection module and the spectral attention mechanism module in improving the network's segmentation accuracy and recall rate. Quantitative comparison experiments on self-built data sets show that the segmentation accuracy of the proposed model for dripping gasoline is 90.34%, and the average detection time of each image is 0.23 s. Compared with Unet, PSE-Unet, and HLCA-Unet models, the average accuracy is increased by 14.39%, 8.01%, and 2.73%, respectively. The recall rate was increased by 8.95%, 8.02%, and 6.55%, and the test time was reduced by 10.83% and 16.97%, respectively, compared with Unet and PSE-Unet models. The qualitative superiority of detection was reflected in the intersection profile of gasoline and background being more consistent with the original image, and the model in this paper could obtain more accurate analysis information of gasoline characteristics. It provides a new technical scheme for gasoline pipeline leakage detection. In addition, compared with the detection of current Unet, PSE-Unet, and HLCA-Unet models on the open Pavia University remote sensing data set, the proposed model still shows better segmentation effect and strong universality and generalization ability; it can be used for many types of hyperspectral image semantic segmentation.