Development of edge computing and classification using The Internet of Things with incremental learning for object detection

The edge computing method and Internet of Things (IoT), which offers significantly shorter inactivity intervals, is one of the promising network technologies in today's generation of sys-tems. There is no need to process the data using a cloud platform whenever an edge computing technology is used; alternative ways employing offline IoT and incremental learning techniques can be used. Using IoT, the incremental learning process transfers all essential data within a specific device. Thus, edge computing, IoT and incremental learning techniques are combined in the proposed method to detect numerous objects with varying weights. An analytical model that minimizes the parametric values and has various objectives is used to carry out the object detection process. Additionally, by utilizing evaluation metrics from five different case studies that were simulated using the MATLAB computing toolkit, the proposed method was tested. The efficacy of the proposed method rises to 62% when the simulated results are compared with the current method. The suggested method can accurately identify several objects in real-time when operating in a multi-object mode.