• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

计算机工程与科学 ›› 2023, Vol. 45 ›› Issue (07): 1253-1262.

• 图形与图像 • 上一篇    下一篇

复杂场景中多阶段自适应帽子检测算法

罗晓霞,邓勇,叶鸥   

  1. (西安科技大学计算机科学与技术学院,陕西 西安 710054)
  • 收稿日期:2022-05-16 修回日期:2022-07-19 接受日期:2023-07-25 出版日期:2023-07-25 发布日期:2023-07-11
  • 基金资助:
    中国博士后科学基金(2020M673446)

A multi-stage adaptive hat detection algorithm in complex scenes

LUO Xiao-xia,DENG Yong,YE Ou   

  1. (College of Computer Science and Technology,Xi’an University of Science and Technology,Xi’an 710054,China) 
  • Received:2022-05-16 Revised:2022-07-19 Accepted:2023-07-25 Online:2023-07-25 Published:2023-07-11

摘要: 针对现有目标检测算法在复杂场景中对小尺度帽子存在误检漏检等问题,提出一种多阶段自适应帽子检测算法(MAHD)。首先,构建一个基于自适应卷积的区域预测网络(MA RPN),通过多阶段对锚框的特征进行细化,提高算法在复杂背景下的目标识别能力;然后,利用自适应采样策略动态分配正负样本,并结合焦点损失函数(Focal Loss)引导MA RPN的训练,提高对小目标的检测精度;最终,在自建的HAT4.5k数据集上进行实验,结果表明,该算法相比Grid R-CNN算法AP提高了2.6%,APS提高了5.1%;并在开源的VisDrone-DET 2019数据集上进一步验证了对小目标的检测效果,所提算法具有较高的检测性能,表明了该算法的可行性和有效性。

关键词: 帽子检测, 自适应采样, 自适应卷积, Grid R-CNN, 焦点损失

Abstract: Existing object detection algorithms have problems with false positives and false negatives when detecting small hats in complex scenes. In this paper, a multi-stage adaptive hat detection algorithm (MAHD) is proposed. Firstly, a region proposal network (MA RPN) based on adaptive convolution is constructed, and the features of anchors are refined through multiple stages to improve the algorithms ability to recognize targets in complex backgrounds. Then, an adaptive sampling strategy is used to dynamically allocate positive and negative samples, and the focus loss function is combined to guide the training of MA RPN and improve the detection accuracy of small targets. Finally, experiments are conducted on a self-built HAT4.5k dataset. The results show that compared with the Grid R-CNN algorithm, the proposed algorithm improves AP by 2.6% and APS by 5.1%. The detection performance of small targets is further verified on the open-source VisDrone-DET 2019 dataset, demonstrating the feasibility and effectiveness of the proposed algorithm.

Key words: hat detection, adaptive sampling, adaptive convolution, Grid R-CNN, Focal Loss