计算机与现代化 ›› 2020, Vol. 0 ›› Issue (07): 90-96.doi: 10.3969/j.issn.1006-2475.2020.07.018

• 人工智能 • 上一篇    下一篇

基于改进YOLOv3网络的学生特定行为识别

  

  1. (北京工业大学信息学部,北京100020)

  • 出版日期:2020-07-06 发布日期:2020-07-15
  • 作者简介:王春辉(1994-),女,安徽宿州人,硕士研究生,研究方向:图像处理,应用研发,E-mail: 1151468968@qq.com; 王全民(1963-),男,教授,博士,研究方向:信息安全,应用研发,E-mail: 1578841971@qq.com。

Student Specific Behavior Recognition Based on Improved YOLOv3 Network

  1. (Information Department, Beijing University of Technology, Beijing 100020, China)
  • Online:2020-07-06 Published:2020-07-15

摘要: 为了提高卷积神经网络在学生行为识别应用的检测精度,本文使用K-means聚类对特有数据集进行聚类得到更适应的anchor box,并且提出一种基于改进损失函数的YOLOv3网络模型。该网络模型将原有的平方和损失函数权重进行动态转化,注重计算连续变量的损失。新的损失函数能有效减低Sigmoid函数梯度消失的影响,使模型收敛更加快速。实验结果表明,基于改进损失函数的深度卷积神经网络应用对“抬头”“低头”“说话”3种姿态的识别均有提高。

关键词: K-means, 图像增强, 损失函数, YOLOv3网络, 姿态识别

Abstract: In order to improve the detection accuracy of convolutional neural networks in student behavior recognition applications, this paper uses K-means clustering to cluster the unique data sets to obtain more adaptive anchor box, and proposes a YOLOv3 network based on improved loss function. The network model dynamically transforms the original squared loss function weights, focusing on the calculation of the loss of continuous variables. The new loss function can effectively reduce the influence of the gradient disappearance of the sigmoid function, making the model converge more quickly. The experimental results show that the deep convolutional neural network based on the improved loss function has improved the recognition of the three poses of “lookup”, “lookdown” and “talk”.

Key words:  , K-means; image enhancement; loss function; YOLOv3 network; gesture recognition

中图分类号: