In the environment of smart examination rooms, it is important to quickly and accurately detect abnormal behavior(human standing) for the construction of a smart campus. Based on deep learning, we propose an intellige...In the environment of smart examination rooms, it is important to quickly and accurately detect abnormal behavior(human standing) for the construction of a smart campus. Based on deep learning, we propose an intelligentstanding human detection (ISHD) method based on an improved single shot multibox detector to detect thetarget of standing human posture in the scene frame of exam room video surveillance at a specific examinationstage. ISHD combines the MobileNet network in a single shot multibox detector network, improves the posturefeature extractor of a standing person, merges prior knowledge, and introduces transfer learning in the trainingstrategy, which greatly reduces the computation amount, improves the detection accuracy, and reduces the trainingdifficulty. The experiment proves that the model proposed in this paper has a better detection ability for the smalland medium-sized standing human body posture in video test scenes on the EMV-2 dataset.展开更多
基金supported by the Natural Science Foundation of China 62102147National Science Foundation of Hunan Province 2022JJ30424,2022JJ50253,and 2022JJ30275+2 种基金Scientific Research Project of Hunan Provincial Department of Education 21B0616 and 21B0738Hunan University of Arts and Sciences Ph.D.Start-Up Project BSQD02,20BSQD13the Construct Program of Applied Characteristic Discipline in Hunan University of Science and Engineering.
文摘In the environment of smart examination rooms, it is important to quickly and accurately detect abnormal behavior(human standing) for the construction of a smart campus. Based on deep learning, we propose an intelligentstanding human detection (ISHD) method based on an improved single shot multibox detector to detect thetarget of standing human posture in the scene frame of exam room video surveillance at a specific examinationstage. ISHD combines the MobileNet network in a single shot multibox detector network, improves the posturefeature extractor of a standing person, merges prior knowledge, and introduces transfer learning in the trainingstrategy, which greatly reduces the computation amount, improves the detection accuracy, and reduces the trainingdifficulty. The experiment proves that the model proposed in this paper has a better detection ability for the smalland medium-sized standing human body posture in video test scenes on the EMV-2 dataset.
