Grant: Collaborative Research and Development Grants

This project will combine advanced image recognition techniques with machine learning to develop an inventory cataloging system. Missing Link Technologies manages data centers hosting a large number of racks consisting of communication servers. In order to maintain the inventory of servers, an expert needs to physically sift through the racks and make decisions on the reconfiguration of servers. The company has developed software to simplify the decision process by making recommendations on available shelves for new equipment. The company proposes to deploy a system to acquire images of racks of networking equipment. The image acquisition can be done by a human taking images of the racks and may eventually be automated with the help of a robotic system. However, in order to further automate this process, the software needs an accurate database of the existing inventory. This project will design and implement a module that will analyze the acquired images to summarize the content of the server racks and store the summaries in a database. The database will also include an image library consisting of crisp and clear images of the equipment that is found on these racks. The contents of the racks will be matched against the equipment in the image library. The module will include a feedback mechanism that will be used by a machine learning algorithm to automatically update the system for improved performance in the future. As a result, the system will continue to improve during its operational phase without software modifications. The company's research shows that there is no other comparable product in the market. The proposed system will give them a competitive advantage with increased productivity and efficient management of equipment.