A novel framework for detection of motion and appearance-based Anomaly using ensemble learning and LSTMs
作者：
Highlights：
• A novel supervised methodology to detect motion and appearance-based Anomaly.
• An ensemble learning strategy to boost the models’ performance.
• Weighted loss is used to encounter the heavy data imbalance.
摘要
•A novel supervised methodology to detect motion and appearance-based Anomaly.•An ensemble learning strategy to boost the models’ performance.•Weighted loss is used to encounter the heavy data imbalance.
论文关键词：Optical flow,Bidirectional LSTM,Ensemble learning,Weighted loss
论文评审过程：Received 20 April 2021, Revised 22 August 2021, Accepted 9 December 2021, Available online 18 December 2021, Version of Record 22 December 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2021.116394
