计算机工程 ›› 2021, Vol. 47 ›› Issue (1): 210-216.doi: 10.19678/j.issn.1000-3428.0056723

• 图形图像处理 • 上一篇    下一篇

基于改进Faster?RCNN的自然场景人脸检测

李祥兵, 陈炼   

  1. 南昌大学 信息工程学院, 南昌 330000
  • 收稿日期:2019-11-27 修回日期:2020-01-23 发布日期:2020-02-19
  • 作者简介:李祥兵(1992-),男,硕士研究生,主研方向为深度学习、计算机视觉;陈炼,教授。
  • 基金项目:
    国家自然科学基金(61862043)。

Face Detection in Natural Scene Based on Improved Faster-RCNN

LI Xiangbing, CHEN Lian   

  1. College of Information Engineering, Nanchang University, Nanchang 330000, China
  • Received:2019-11-27 Revised:2020-01-23 Published:2020-02-19

摘要: 为实现对自然场景下小尺度人脸的准确检测,提出一种改进的Faster-RCNN模型。采用ResNet-50提取卷积特征,对不同卷积层的特征图进行多尺度融合,同时将区域建议网络产生的锚框由最初的9个改为15个,以更好地适应小尺度人脸检测场景。在此基础上,利用在线难例挖掘算法优化训练过程,采用软非极大值抑制方法解决漏检重叠人脸的问题,并在训练阶段通过多尺度训练提高模型的泛化能力。实验结果表明,该模型在Wider Face数据集上平均精度为89.0%,较原Faster-RCNN模型提升3.5%,在FDDB数据集上检出率也高达95.6%。

关键词: 人脸检测, Faster-RCNN模型, 多尺度融合, 在线难例挖掘, 软非极大值抑制

Abstract: To realize accurate detection of small-scale faces in natural scene,this paper constructs an improved Faster-RCNN model.The model uses ResNet-50 to extract convolution features,and performs multi-scale fusion for feature maps of different convolutional layers.At the same time,the number of Anchors generated by the Regional Proposal Network(RPN) has been changed from 9 to 15 to better adapt to the small-scale face detection scenes.On this basis,the Online Hard Example Mining(OHEM) algorithm is used to optimize the training process.Soft-Non-Maximum Suppression(Soft-NMS) method is used to reduce the missed detection of overlapping faces,and in the training phase the multi-scale training method is adopted to improve the generalization ability of the model.Experimental results show that the average precision of the proposed model is 89.0% on the Wider Face dataset,which is 3.5% higher than that of the original Fast-RCNN model.The relevance ratio of the proposed model reaches 95.6% on the FDDB dataset.

Key words: face detection, Faster-RCNN model, multi-scale fusion, Online Hard Example Mining(OHEM), Soft-Non-Maximum Suppression(Soft-NMS)

中图分类号: