Human Activity Recognition Using Conv-LSTM with Infrared Camera Surveillance

Human Activity Recognition (HAR) the usage of infrared (IR) video statistics leverages superior computer vision and deep gaining knowledge of techniques to detect and classify human activities in numerous environments, even beneath low or no-light situations. This is important for packages including safety, surveillance, and healthcare diagnostics. Deep mastering models, which includes convolutional neural networks (CNNs) and long short-term memory (LSTM) networks, play a substantial position in this technique. CNNs extract spatial features from person frames, while LSTMs model time-structured video sequences, permitting the accurate popularity of complex activities. In this have a look at, the Conv-LSTM version demonstrated superior performance, accomplishing an accuracy of ninety seven.71% on an infrared video dataset. This method outperformed other trendy models, consisting of Enhanced YOLOv5, Attention-primarily based LSTM, Deep Conv-LSTM, and movement recognition models based totally on thermal motion pictures. These outcomes highlight the effectiveness of the Conv-LSTM framework in infrared-based activity popularity, showcasing its potential for actual-time analytics in hard environments.