Research Article

Foreground-Aware Region of Interest Cropping and CBAM Attention for Excavator Segmentation in UAV Imagery

Volume: 21 Number: 2 September 30, 2026
EN TR

Foreground-Aware Region of Interest Cropping and CBAM Attention for Excavator Segmentation in UAV Imagery

Abstract

In disaster situations like earthquakes, excavators have a crucial part to play in the rapid clearing of rubble, which is critical for conducting search and rescue efforts. Effective tracking and localization of excavators can be achieved through accurate monitoring of their presence and distribution in disaster-hit regions, enabling better coordination and resource management. This study presents a deep learning-based approach for the segmentation of excavators in high-resolution Unmanned Aerial Vehicle (UAV) imagery, addressing challenges related to small object representation and complex backgrounds. Specifically, our proposed approach involves using a foreground-aware region of interest cropping mechanism to ensure that there are sufficient excavator pixels in the training dataset, and a convolutional block attention module to improve feature refinement by taking into account channel-wise and spatial relationships between features. The proposed network utilizes a U-Net architecture based on ResNet-34. Experimental evaluations on UAV imagery reveal that the proposed approach achieved an IoU of 0.663, mIoU of 0.830, Dice score of 0.798, Precision of 0.856, Recall of 0.747 and geometric mean (G-Mean) of 0.863. These results demonstrate a notable improvement over the baseline model (U-Net with a ResNet-34 backbone), with the IoU increasing from 0.509 to 0.663 and the Dice score improving from 0.675 to 0.798.

Keywords

References

  1. Zhu X.X., Tuia D., Mou L., Xia G.S., Zhang L., Xu F., Fraundorfer F. Deep learning in remote sensing: A comprehensive review and list of resources. IEEE Geosci Remote Sens Mag 2017; 5(4): 8–36.
  2. Xia G.S., Bai X., Ding J., Zhu Z., Belongie S., Luo J., Zhang L. DOTA: A large-scale dataset for object detection in aerial images. Proc IEEE Conf Comput Vis Pattern Recognit 2018; 3974–3983.
  3. Fang W., Ding L., Luo H., Love P.E.D. Falls from heights: A computer vision-based approach for safety harness detection. Autom Constr 2018; 91: 53–61.
  4. Li Q, Sun M, Dong L, Gao X, Wang Z, Zhang H. HCD-Mask: A multi-task model for small object detection and instance segmentation in high-resolution UAV images. 2022 IEEE International Conference on Industrial Technology (ICIT); 22–25 August 2022; Shanghai, China. pp. 1–8.
  5. Yu A, Quan Y, Yu R, Guo W, Wang X, Hong D, Zhang H, Chen J ve diğerleri. Deep Learning Methods for Semantic Segmentation in Remote Sensing with Small Data: A Survey. Remote Sens 2023; 15(20): 4987.
  6. Qiu Z, Gong T, Cong R, Zhao Y, Bai H. UESDNet: UAV Visual Multitask Perception Via Expert-Guided Prior With Mutual Optimization of Segmentation and Detection. IEEE J Sel Top Appl Earth Obs Remote Sens 2026; 19: 3157–3174.
  7. Meng L., Peng Z., Zhou J., Zhang J., Lu Z., Baumann A., Du Y. Real-time detection of ground objects based on unmanned aerial vehicle remote sensing with deep learning: Application in excavator detection for pipeline safety. Remote Sens 2020; 12(1): 182.
  8. Zhang S., Zhang L. Construction site safety monitoring and excavator activity analysis system. Constr Robot 2022; 6(2): 151–161.

Details

Primary Language

English

Subjects

Deep Learning

Journal Section

Research Article

Publication Date

September 30, 2026

Submission Date

May 7, 2026

Acceptance Date

September 24, 2026

Published in Issue

Year 2026 Volume: 21 Number: 2

APA
Şeker, S. E., Değirmenci, A., & Çankaya, İ. (2026). Foreground-Aware Region of Interest Cropping and CBAM Attention for Excavator Segmentation in UAV Imagery. Turkish Journal of Science and Technology, 21(2), 437-447. https://doi.org/10.55525/tjst.1941049
AMA
1.Şeker SE, Değirmenci A, Çankaya İ. Foreground-Aware Region of Interest Cropping and CBAM Attention for Excavator Segmentation in UAV Imagery. TJST. 2026;21(2):437-447. doi:10.55525/tjst.1941049
Chicago
Şeker, Salih Enes, Ali Değirmenci, and İlyas Çankaya. 2026. “Foreground-Aware Region of Interest Cropping and CBAM Attention for Excavator Segmentation in UAV Imagery”. Turkish Journal of Science and Technology 21 (2): 437-47. https://doi.org/10.55525/tjst.1941049.
EndNote
Şeker SE, Değirmenci A, Çankaya İ (September 1, 2026) Foreground-Aware Region of Interest Cropping and CBAM Attention for Excavator Segmentation in UAV Imagery. Turkish Journal of Science and Technology 21 2 437–447.
IEEE
[1]S. E. Şeker, A. Değirmenci, and İ. Çankaya, “Foreground-Aware Region of Interest Cropping and CBAM Attention for Excavator Segmentation in UAV Imagery”, TJST, vol. 21, no. 2, pp. 437–447, Sept. 2026, doi: 10.55525/tjst.1941049.
ISNAD
Şeker, Salih Enes - Değirmenci, Ali - Çankaya, İlyas. “Foreground-Aware Region of Interest Cropping and CBAM Attention for Excavator Segmentation in UAV Imagery”. Turkish Journal of Science and Technology 21/2 (September 1, 2026): 437-447. https://doi.org/10.55525/tjst.1941049.
JAMA
1.Şeker SE, Değirmenci A, Çankaya İ. Foreground-Aware Region of Interest Cropping and CBAM Attention for Excavator Segmentation in UAV Imagery. TJST. 2026;21:437–447.
MLA
Şeker, Salih Enes, et al. “Foreground-Aware Region of Interest Cropping and CBAM Attention for Excavator Segmentation in UAV Imagery”. Turkish Journal of Science and Technology, vol. 21, no. 2, Sept. 2026, pp. 437-4, doi:10.55525/tjst.1941049.
Vancouver
1.Salih Enes Şeker, Ali Değirmenci, İlyas Çankaya. Foreground-Aware Region of Interest Cropping and CBAM Attention for Excavator Segmentation in UAV Imagery. TJST. 2026 Sep. 1;21(2):437-4. doi:10.55525/tjst.1941049