Illegal behavior detection method based on improved YOLOv8n

• Main Authors: WANG LEI, YI XI, LEI CANRU, WU JUAN, ZHAO DAN, LUO RENZE, LIAO BO, WU TAO, CAO RUI, LIN HONGYU, LIU HENG
• Format: Patent
• Language: Chinese; English
• Published: 17.11.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; PHYSICS

The invention discloses an illegal behavior detection method based on improved YOLOv8n. The method comprises the following steps: collecting illegal behavior video decoding annotations through a camera to establish a data set; dividing the marked data set into a training set, a verification set and a test set; a YOLOv8n network is improved, a training set is used for training the improved YOLOv8n, a verification set is used for verification, network parameters with the highest accuracy of the verification set are reserved, and a test set is used for testing the improved YOLOv8n network imported with the network parameters. According to the invention, by improving the C2f module and introducing the attention mechanism to construct the feature fusion module, the feature extraction capability and the detection precision of the model are improved. 本发明公开了一种基于改进YOLOv8n的违规行为检测方法，包括：通过摄像头采集违规行为视频解码标注建立数据集；将标注好的数据集划分为训练集，验证集和测试集；对YOLOv8n网络进行改进，使用训练集对改进的YOLOv8n进行训练，同时使用验证集进行验证，保留验证集准确率最高的网络参数，使用测试集对导入网络参数改进的YOLOv8n网络进行