External damage transmission line accidents severely threaten the safe and stable operation of power systems. A primary cause of such accidents is the insufficient safety distance between engineering vehicles and transmission lines during construction operations. To address the critical need for accurate distance monitoring in this scenario, this paper makes three key contributions: First, it optimizes the YOLOv8-based vehicle detection algorithm to improve the accuracy and speed of engineering vehicle identification, ad-dressing the challenge of real-time target localization in complex construction environments. Second, the method employs an improved YOLOv8 model integrated with the bounding box prompting mechanism of the Segment Anything Model to achieve precise segmentation of transmission lines, effectively overcoming the limitations of traditional segmentation approaches in handling complex background interference and slender linear structures. Third, it derives a distance calculation formula tailored to practical engineering requirements, enabling accurate measurement of the spatial distance between detected engineering vehicles and segmented transmission lines. Experimental results verify that the proposed method achieves superior performance in engineering vehicle detection, transmission line segmentation, and distance calculation. This work effectively supports the intelligent monitoring and early warning capabilities of transmission line external damage prevention devices, providing a reliable technical solution for mitigating accident risks from engineering vehicle-related external damage.