Muck truck sealing cover detection model, training method, system, equipment and medium

• Main Authors: QIU SHUHONG, GAN HAIHUA, LU HAIRONG, WU XIANQIAO, TONG SUN
• Format: Patent
• Language: Chinese; English
• Published: 23.06.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; PHYSICS

The invention relates to the field of deep learning, and provides a muck truck sealing cover detection model, a training method, a system, equipment and a medium. The training method comprises the following steps: acquiring a real-time video stream; performing frame extraction processing on the real-time video stream to obtain a plurality of sample images; marking whether the sample image is sealed or not and key points of the carriage to obtain a training sample set; and using the training sample set to train the muck truck sealing cover detection model. The network structure of the muck truck sealing cover detection model has better delay-precision balance, can extract stronger features, can capture global information and rich context information, can effectively improve the ability to cope with errors in a small range and a medium range in a deep network training process, and can improve the accuracy of the detection of the sealing cover of the muck truck. The problem of training data balance is solved, th