Oilfield inspection tasks are featured by complex surroundings, numerous targets and large-scale coverage. Conventional path planning methods suffer from drawbacks including low search efficiency, redundant trajectories and frequent steering. To overcome these limitations, a high-efficiency ant colony optimization path planning algorithm with multi-strategy improvements is proposed. Based on the standard ant colony framework, a target-oriented heuristic function is constructed to enhance path guidance and eliminate invalid turns. An adaptive regulation strategy for pheromone and expected heuristic factors is introduced to boost the global search ability and convergence stability in complex environments. Moreover, an adaptive evaporation coefficient based on information entropy is devised to effectively balance the global path quality and convergence rate. Simulation results illustrate that in the single-target scenario, the proposed algorithm reduces the path length by 14.7% and 10.6%, the iteration number by 25% and 45.4%, and the turning frequency by 56.1% and 41.9% respectively, in comparison with the standard ant colony algorithm and existing improved variants. In the multi-target inspection scenario, it achieves a 6.0% reduction in path length relative to the standard algorithm. The presented algorithm prominently promotes the efficiency and practicability of path planning, which can furnish reliable technical support for intelligent oilfield inspection.