Region-based Convolutional Neural Networks with IoT-based Alzheimer's disease detection and classifications

It is proposed that more research into the potential of thermal imaging cameras for human recognition be conducted. When visual recognition is difficult, we recommend building 3D facial reconstructions and combining speech to provide a reliable identification read. Given the numerous uses of facial reconstruction, it is critical to investigate how information theory might be used to speed up and optimised the process. Furthermore, our prototype allows us to follow the whereabouts of the individual with AD. Furthermore, our IoT prototype secures the camera photos using steganography, which allows the recipient to decode the original image using a key. Accuracy, precision, and recall were used to assess the performance of the RCNN-PSO proposed system. The experimental analysis showed that the proposed model RCNN-PSO was able to achieve high performance in multiclass classification.