Engaging in visual search tasks entails the identification of a target object within a typically intricate background. Furthermore, the task’s complexity escalates when tracked objects are obscured or resemble items located nearby. This challenge is particularly pronounced in cluttered industrial environments, where tools of diverse sizes and shapes are often grouped in close proximity. In such settings, the swift and accurate identification of the correct tool plays a pivotal role in facilitating manual tasks like assembly or device repair. To improve visual search capabilities, we propose leveraging augmented reality (AR) head-mounted displays (HMDs) equipped with cutting-edge object detection functionalities. To substantiate this proposal, we conducted an exploratory study featuring eleven participants within a highly ecologically valid experimental environment. The study’s outcomes underscore the transformative potential of AR HMDs in efficiently locating objects within the chaotic backdrop of an industrial setting. The explored approach enhances human capabilities and sheds light on novel experimental designs. Integrating the Segment Anything Model and the {YOLOv}8 object detection model further amplifies the effectiveness of visual search tasks.