Hand Tool Identification System

Abstract

Our team is working on a project in the interest of the Kentucky National Guard and which is sponsored by the Department of Defense. Our project supervisor is Chief Warrant Officer Christopher Kehl from the Kentucky National Guard. The project aims to tackle a problem faced by the supply and inventory management staff at the Kentucky National Guard and at other military establishments. More specifically, that there are certain tools in the inventory which cannot be properly identified by the inventory staff. Inventory staff are responsible for keeping track of which tools are needed and currently held by each technician. However, inventory staff are not always knowledgeable about the tools that mechanics and technicians use in their work. Also due to operational considerations, inventory staff and technicians are not often in direct contact. Thus, tools are often misidentified or not identified at all, leading to inventory shortages, or worse, the wrong tools being distributed to technicians’ tool boxes or workstations. Our solution consists of a stand-alone native Android mobile application which uses the device’s camera to obtain images of a tool which must be identified and which then uses artificial intelligence which implements object recognition and image classification models in order to identify the type and size of the tool for the user.