Because of the absence of mechanical components for separation and connection, transmission lines without spacers are at a higher risk of galloping during strong winds, freezing rain, and heavy snow. However, current research has yet to achieve targeted real-time high-precision monitoring. To address this issue, this paper proposes a galloping identification method for transmission lines without spacers based on trajectory extraction and ellipse detection. First, data augmentation and low-illumination image enhancement processing are performed. Next, a lightweight improvement of the traditional DeepLabV3+ model is carried out to lock into the target line and fast segmentation is realized in the initial frame of the video. Finally, after extracting the short-term operational trajectory of the transmission line, the galloping monitoring process is transformed into ellipse detection of the trajectory based on inspection expert knowledge. The results demonstrate that the proposed method effectively balances the speed and accuracy of line trajectory extraction, and can analyze the galloping severity in transmission lines in complex environmental conditions. This has significant practical value for transmission line inspection in extreme weather.