Journal of Computer Applications ›› 2023, Vol. 43 ›› Issue (8): 2602-2610.DOI: 10.11772/j.issn.1001-9081.2022071009

• Multimedia computing and computer simulation • Previous Articles    

Leukocyte detection method based on twice-fusion-feature CenterNet

Huan LIU, Lianghong WU(), Lyu ZHANG, Liang CHEN, Bowen ZHOU, Hongqiang ZHANG   

  1. School of Information and Electrical Engineering,Hunan University of Science and Technology,Xiangtan Hunan 411201,China
  • Received:2022-07-11 Revised:2022-11-18 Accepted:2022-11-30 Online:2023-01-15 Published:2023-08-10
  • Contact: Lianghong WU
  • About author:LIU Huan, born in 1997, M. S. candidate. Her research interests include image processing.
    ZHANG Lyu, born in 1997, M. S. His research interests include computer vision, image processing.
    CHEN Liang, born in 1985, Ph. D., lecturer. His research interests include deep learning, image recognition, computer vision.
    ZHOU Bowen, born in 1983, Ph. D., senior engineer. His research interests include computer vision, image processing.
    ZHANG Hongqiang, born in 1979, Ph. D., lecturer. His research interests include swarm robot system, swarm intelligence, optimization and intelligent control.
  • Supported by:
    National Defense Basic Research Program of China(JCKY2019403D006);Natural Science Foundation of Hunan Province(2021JJ30280);Hunan Province Science and Technology Innovation Program(2017XK2302)


刘欢, 吴亮红(), 张侣, 陈亮, 周博文, 张红强   

  1. 湖南科技大学 信息与电气工程学院,湖南 湘潭 411201
  • 通讯作者: 吴亮红
  • 作者简介:刘欢(1997—),女,河南南阳人,硕士研究生,CCF会员,主要研究方向:图像处理
  • 基金资助:


Leukocyte detection is difficult due to different shapes and degrees of staining of leukocytes during real detection process in complex scenarios. To solve the problem, a dual feature fusion CenterNet based leukocyte detection method TFF-CenterNet (Twice-Fusion-Feature CenterNet) was proposed. Firstly, the features of the backbone network were fused with the features of deconvolution layers through Feature Pyramid Network (FPN). In this way, the feature extraction ability of the method was improved to solve the problems of individual differences and different degrees of staining of leukocytes. Then, aiming at the problem of severe imbalance between the image area of leukocytes and the background image area, the heatmap loss function was improved to enhance the focus on positive samples of leukocyte and improve detection mean Average Precision (mAP). Finally, for the characteristics of the tiny target, random location, and cell adhesion of leukocyte images, coordinate attention and coordinate convolution were introduced to improve the attention and sensitivity of leukocyte location information. For leukocytes in complex scenarios, TFF-CenterNet achieves the mAP of 97.01% and the detection speed of 167 frame/s, which are 3.24 percentage points higher and 42 frame/s faster than those of CenterNet respectively. Experimental results show that the proposed method can improve the mAP of leukocyte detection in complex situations while achieving real-time requirements, and improves the robustness, so that this method can provide technical support for rapid automatic leukocyte detection in complementary medical diagnosis.

Key words: leukocyte detection, CenterNet, coordinate attention, feature fusion, coordinate convolution


针对实际检测过程中复杂场景下由白细胞的形态、染色程度差异较大而导致白细胞检测困难的问题,提出一种基于特征双融合CenterNet的白细胞检测方法TFF-CenterNet(Twice-Fusion-Feature CenterNet)。首先,通过特征金字塔网络(FPN)将主干网络特征与反卷积层特征相融合,增强方法的特征提取能力,从而解决白细胞个体差异、染色程度不同等问题;然后,针对白细胞占据图像面积与图像背景面积严重不均衡的问题,改进热力图损失函数来提升对白细胞正样本的关注以提高检测平均精度均值(mAP);最后,针对白细胞图像目标微小、位置随机、细胞粘连的特点,引入坐标注意力和坐标卷积,提高对白细胞位置信息的关注度和敏感性。对于复杂场景下的白细胞,TFF-CenterNet的mAP达到97.01%,比CenterNet高3.24个百分点;检测速度达到167 frame/s,比CenterNet快42 frame/s。实验结果表明,所提方法在复杂情况下能在提高白细胞检测mAP的同时达到实时性要求,并提升了鲁棒性,可为辅助医疗诊断中白细胞的快速自动检测提供技术支持。

关键词: 白细胞检测, CenterNet, 坐标注意力, 特征融合, 坐标卷积

CLC Number: