Shan Wang, Yiying Hu, Liang Feng, Linying Guo. Improved Breast Mass Recognition YOLOv3 Algorithm Based on Cross-Layer Feature Aggregation[J]. Laser & Optoelectronics Progress, 2022, 59(4): 0410003
- Laser & Optoelectronics Progress
- Vol. 59, Issue 4, 0410003 (2022)
![Network structure of YOLOv3[20]](/richHtml/lop/2022/59/4/0410003/img_1.jpg)
Fig. 1. Network structure of YOLOv3[20]
![Structural unit of YOLOv3[21]. (a) CBL unit; (b) residual unit](/richHtml/lop/2022/59/4/0410003/img_2.jpg)
Fig. 2. Structural unit of YOLOv3[21]. (a) CBL unit; (b) residual unit
![Feature pyramid network[15]](/Images/icon/loading.gif){kind=icon}
Fig. 3. Feature pyramid network[15]
Fig. 4. Structure of MCFN
Fig. 5. Pseudo code of SD-NMS algorithm
Fig. 7. Structure of proposed network
Fig. 8. Illustration of breast cancer screening cases
Fig. 9. Cropped images
Fig. 10. "Sphere" structural element[22]
Fig. 11. Result of local enhancement. (a) Original image; (b) enhanced image
Fig. 12. Training results of MCF-YOLO
Fig. 13. Comparison of Precision-Recall curves between MCF-YOLO and other object detection algorithms
Fig. 14. Comparison of masses detection results of five algorithms. (a) Ground truth of breast masses; (b) Faster RCNN; (c) SSD; (d) RetinaNet; (e) YOLOv3; (f) MCF-YOLO
|
Table 1. Quantitative comparison between MCF-YOLO and other object detection algorithms
Set citation alerts for the article
Please enter your email address
© Copyright 2018-2021 | Chinese Laser Press. All Rights Reserved 沪ICP备15018463号-20