Timely and accurate flame detection is a very important and practical technology for preventing the occurrence of fire accidents effectively. However, the current methods of flame detection are still faced with many challenges in video surveillance scenarios due to issues such as varying flame shapes, imbalanced samples, and interference from flame-like objects. In this work, a real-time flame detection method based on deformable object detection and time sequence analysis is proposed to address these issues. Firstly, based on the existing single-stage object detection network YOLOv5s, the network structure is improved by introducing deformable convolution to enhance the feature extraction ability for irregularly shaped flames. Secondly, the loss function is improved by using Focal Loss as the classification loss function to solve the problems of the imbalance of positive (flames) and negative (background) samples, as well as the imbalance of easy and hard samples, and by using EIOU Loss as the regression loss function to solve the problems of a slow convergence speed and inaccurate regression position in network training. Finally, a time sequence analysis strategy is adopted to comprehensively analyze the flame detection results of the current frame and historical frames in the surveillance video, alleviating false alarms caused by flame shape changes, flame occlusion, and flame-like interference. The experimental results indicate that the average precision (AP) and the F-Measure index of flame detection using the proposed method reach 93.0% and 89.6%, respectively, both of which are superior to the compared methods, and the detection speed is 24–26 FPS, meeting the real-time requirements of video flame detection.

Smoking and calling are two typical behaviors involved in public and industrial safety that usually need to be strictly monitored and even prohibited on many occasions. To resolve the problems of missed detection and false detection in the existing traditional and deep-learning-based behavior-recognition methods, an intelligent recognition method using a multi-task YOLOv4 (MT-YOLOv4) network combined with behavioral priors is proposed. The original YOLOv4 is taken as the baseline network to be improved in the proposed method. Firstly, a K-means++ algorithm is used to re-cluster and optimize the anchor boxes, which are a set of predefined bounding boxes to capture the scale and aspect ratio of specific objects. Then, the network is divided into two branches with the same blocks but independent tasks after the shared feature extraction layer of CSPDarknet-53, i.e., the behavior-detection branch and the object-detection branch, which predict the behaviors and their related objects respectively from the input image or video frame. Finally, according to the preliminary predicted results of the two branches, comprehensive reasoning rules are established to obtain the final behavior-recognition result. A dataset on smoking and calling detection is constructed for training and testing, and the experimental results indicate that the proposed method has a 6.2% improvement in recall and a 2.4% improvement in F1 score at the cost of a slight loss in precision compared to the baseline method; the proposed method achieved the best performance among the compared methods. It can be deployed to related security surveillance systems for unsafe-behavior monitoring and early-warning management in practical scenarios.