Abstract: This article proposes an intelligent path automatic optimization method based on machine vision and improved particle swarm optimization algorithm to address the problems of low efficiency, high labor costs, and insufficient obstacle avoidance performance in the traditional manual inspection mode of substations. This method obtains the initial inspection path through machine vision and human-computer interaction technology, constructs a multi constraint optimization model with the goal of minimizing the path length, and introduces a dynamic weight adjustment mechanism and kinematic constraint processing strategy to achieve automated solution of the global optimal path. Finally, taking a 220 kV substation as an example, this method was compared with ant colony optimization method and hybrid immune particle swarm optimization method in simulation experiments. The results showed that the proposed method has a shorter inspection path and less inspection time, providing a more efficient solution for intelligent inspection of substations.