b'Partner News Spring 2024 31CorporateThis is where, with a little creative thinkingJohn Golding, Owner of MLT, on the and partnerships with like-minded engineers,future of this technology explained the we were able to collaborate together topossibilities of the technology: BuildingNaturally, AI requires a huge solve this problem. Working with a newon the success of our recent trial, we start-up company called Machine Learninghave now incorporated other cutting- amount of training to reduce Technologies (MLT), who were creatingedge developments into our platform.the number of false positives different AI models, we collaborated to design and trial a Human DetectionThe introduction of real-time luggageby the models and to increase system using a combination of AI andIATA classification and passengerthe reliability of detection and basic imaging equipment (cameras). Thisheatmaps within the arrivals hall will required several AI models to be created:further our mission to create a seamless,identification of individuals, one for discriminating between the worldsecure, and highly efficient airportitems, and bags. This was and a human form, another for humanenvironment. These features represent gesture identification, and lastly, one toour commitment to continuous innovationcompleted over a period by determine baggage from everything else. in response to the evolving needs ofcollecting and processing modern forward-thinking airports. thousands of images of Of course, AI requires a huge amount of training the different models to reduce theWhere does it stop? With the abundancepeople and bags.number of false positives by the modelsof data collected daily by companies and and to increase the reliability of detectionthe flexibility of AI, machine learning, and and identification of individuals, items, anddeep learning machines, the applications are bags. This was completed over a periodendless when it comes to the ability of AI by collecting and processing thousands ofsupplementing our businesses and leisure images of people and bags. The initial trainingtime. So, the advent of AIs second dawn is was bolstered by a real-life trial in Londonbringing a new way of looking at complex Stansteds busy Arrivals Hall. This acceleratedengineering problems and using innovation the model to a point where the false positiveto simplify the solutions and produce alarm rate hit zero, making the AI competentbetter results of a repeatable nature.in the discerning of people and bags.Though AI cannot be solely used for safety purposes, using a layered application of deterrence, passive systems (guarding), and active systems (safety devices and AI) enables AI to be efficient in detection and verified by the other devices.Demonstrating the AI Human Presence Detection system'