A Deep Neuronal Network-Based Sensor for the Detection of Oxygen inside the Carbonization Furnace

Agricultural production generates biological surpluses that can be used to produce additional products through thermal and/or chemical transformations, such as the carbonization of vegetables. The oxygen on vegetable material during process inside a furnace is an important parameter determines not only success but also quality final product. It difficult measure in real time, partly because working environment sensor, and operating characteristics (continuous rotation its axis). goal this project develop reliable measurement system capable real-time. For continuous precise detection temperature furnaces plant, we propose method based characterization image using deep neuronal network. First all, images interior are captured digital camera front axis furnace. Then, area interest each frame video determined by processing. Oxygen content information then obtained model trained for same with pattern level sensors. Experiments conducted actual conditions demonstrate proposed estimating provides data control plant operation. scheme demonstrated high reliability against many changes present furnace, also, low computational consumption makes it viable strategy embedded implementation operation