ASF-YOLO: A novel YOLO model with attentional scale sequence fusion for cell instance segmentation

被引:64
|
作者
Kang, Ming [1 ]
Ting, Chee-Ming [1 ]
Ting, Fung Fung [1 ]
Phan, Raphael C. -W. [1 ]
机构
[1] Monash Univ, Sch Informat Technol, Malaysia Campus, Subang Jaya 47500, Malaysia
关键词
Medical image analysis; Small object segmentation; You only look once (YOLO); Sequence feature fusion; Attention mechanism;
D O I
10.1016/j.imavis.2024.105057
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We propose a novel Attentional Scale Sequence Fusion based You Only Look Once (YOLO) framework (ASFYOLO) which combines spatial and scale features for accurate and fast cell instance segmentation. Built on the YOLO segmentation framework, we employ the Scale Sequence Feature Fusion (SSFF) module to enhance the multiscale information extraction capability of the network, and the Triple Feature Encoder (TFE) module to fuse feature maps of different scales to increase detailed information. We further introduce a Channel and Position Attention Mechanism (CPAM) to integrate both the SSFF and TFE modules, which focus on informative channels and spatial position-related small objects for improved detection and segmentation performance. Experimental validations on two cell datasets show remarkable segmentation accuracy and speed of the proposed ASF-YOLO model. It achieves a box mAP of 0.91, mask mAP of 0.887, and an inference speed of 47.3 FPS on the 2018 Data Science Bowl dataset, outperforming the state-of-the-art methods. The source code is available at https://github. com/mkang315/ASF-YOLO.
引用
收藏
页数:9
相关论文
共 27 条
  • [1] AM YOLO: adaptive multi-scale YOLO for ship instance segmentation
    Yuan, Ming
    Meng, Hao
    Wu, Junbao
    JOURNAL OF REAL-TIME IMAGE PROCESSING, 2024, 21 (04)
  • [2] Multi-scale adaptive YOLO for instance segmentation of grape pedicels
    Shen, Qian
    Zhang, Xuyao
    Shen, Mengcheng
    Xu, Dayu
    COMPUTERS AND ELECTRONICS IN AGRICULTURE, 2025, 229
  • [3] YUSEG:Yolo and Unet is all you need for cell instance segmentation
    Bai, Bizhe
    Tian, Jie
    Luo, Sicong
    Wang, Tao
    Lyu, Sisuo
    COMPETITIONS IN NEURAL INFORMATION PROCESSING SYSTEMS, VOL 212, 2022, 212
  • [4] BHI-YOLO: A Lightweight Instance Segmentation Model for Strawberry Diseases
    Hu, Haipeng
    Chen, Mingxia
    Huang, Luobin
    Guo, Chi
    APPLIED SCIENCES-BASEL, 2024, 14 (21):
  • [5] PSC-YOLO: a lightweight model for urban road instance segmentation
    Gu, Xiaolin
    Zhang, Guofeng
    JOURNAL OF REAL-TIME IMAGE PROCESSING, 2025, 22 (02)
  • [6] A Multi-Scale Sequence Fusion YOLO Model for Surface Defect Detection of Shock Absorber Connecting Rods
    Zhou, Yueqi
    Cui, Jialin
    Lian, Bin
    Zhao, Xianghong
    Zhang, Ziliang
    Wan, Zhiyi
    2024 IEEE 19TH CONFERENCE ON INDUSTRIAL ELECTRONICS AND APPLICATIONS, ICIEA 2024, 2024,
  • [7] GHA-Inst: a real-time instance segmentation model utilizing YOLO detection framework
    Dong, Chengang
    Tang, Yuhao
    Zhang, Liyan
    CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS, 2024, 27 (06): : 7401 - 7415
  • [8] TW-YOLO: An Innovative Blood Cell Detection Model Based on Multi-Scale Feature Fusion
    Zhang, Dingming
    Bu, Yangcheng
    Chen, Qiaohong
    Cai, Shengbo
    Zhang, Yichi
    SENSORS, 2024, 24 (19)
  • [9] Lightweight Underwater Object Detection Based on YOLO v4 and Multi-Scale Attentional Feature Fusion
    Zhang, Minghua
    Xu, Shubo
    Song, Wei
    He, Qi
    Wei, Quanmiao
    REMOTE SENSING, 2021, 13 (22)
  • [10] MSTA-YOLO: A novel retinal ganglion cell instance segmentation method using a task-aligned coupled head and efficient multi-scale attention for glaucoma analysis
    Lin, Hao
    Lin, Meimin
    Chang, Weijie
    Li, Xinlei
    Li, Yang
    Tang, Lewei
    Zhuang, Zhengfei
    Liang, Yuanbo
    BIOMEDICAL SIGNAL PROCESSING AND CONTROL, 2025, 106