Human behavior and abnormality detection is an emerging technology that utilizes advanced algorithms such as YOLO (You Only Look Once) and Convolutional Neural Networks (CNN) to extract features, handle temporal dependencies, and enhance the accuracy and efficiency of human behavior detection systems. This technology is specifically designed to be applicable in real-world scenarios, particularly for detecting abnormal activities in surveillance videos captured by closed circuit television (CCTV) cameras. The model leverages YOLOv3, an object detection technology, to detect human behaviors and abnormalities in the video data. Subsequently, a Convolutional Neural Network (CNN) is employed to extract action characteristics from each tracked trajectory. Finally, a Long Short-Term Memory Network (LSTM) is utilized to construct a model for identifying anomalous behaviors, thereby predicting abnormal activities performed by humans. In summary, this technology operates by taking video footage as input from surveillance cameras and applies a series of advanced algorithms to detect and classify abnormal activities. By utilizing YOLO and CNN for feature extraction and LSTM for anomaly identification, the system can accurately identify and predict abnormal human behavior.