SSFYOLO: a trunk detection algorithm for complex forest scenarios

Yang Can; Fan Xijian; Zhang Jiuyu

doi:10.12171/j.1000-1522.20240145

Journal of Beijing Forestry University > 2025 > 47(2): 132-142. > DOI: 10.12171/j.1000-1522.20240145

Yang Can, Fan Xijian, Zhang Jiuyu. SSFYOLO: a trunk detection algorithm for complex forest scenarios[J]. Journal of Beijing Forestry University, 2025, 47(2): 132-142. DOI: 10.12171/j.1000-1522.20240145

Citation:

PDF (16562 KB)

SSFYOLO: a trunk detection algorithm for complex forest scenarios

College of Information Science and Technology & Artificial Intelligence, Nanjing Forestry University, Nanjing 210037, Jiangsu, China

More Information

Received Date: May 10, 2024
Revised Date: December 14, 2024
Accepted Date: December 17, 2024
Available Online: December 22, 2024

Graphical Abstract

Abstract

Abstract

Objective
Due to the large size variation and frequent occlusion of trunks in complex environments, trunk detection is susceptible to issues such as missed detections and false positives. To effectively solve these problems, a trunk detection network named SSFYOLO, based on a single-stage target detection framework, was proposed.
Method
Firstly, a spatial awareness module (SAM) was designed. The SAM module can efficiently process multi-scale and multi-resolution feature information, ensuring computational efficiency while fully integrating and accurately extracting various features, thereby improving the accuracy and efficiency of target detection. Secondly, a multi-scale feature enhancement adaptive network (FastScaleNet) was designed to replace the C2f structure in the YOLO model. By refining multi-scale feature fusion and optimization, and utilizing skip connections and feature fusion strategies, FastScaleNet effectively retained feature information at different levels, enhanced the model’s ability to express targets of different scales, and improved the model’s robustness and versatility. Finally, the WIoU mechanism was introduced to dynamically optimize the loss weights for small targets. Depending on the target size, the WIoU mechanism dynamically adjusted the loss weights, allowing the model to flexibly adjust parameters when facing small targets. This enabled the model to adapt to the detection requirements of targets of different sizes, further improving the accuracy and robustness of small object detection.
Result
Detection experiments were conducted on a trunk dataset in complex scenarios. Compared with the mainstream detection algorithm YOLOv8, the proposed SSFYOLO algorithm achieved better detection accuracy with reduced parameter volume. Specifically, the parameter volume was reduced by 20%, mAP and the recall rate represented improvements of 1.6 and 0.7 percentage points, respectively.
Conclusion
This study has designed a tree trunk detection algorithm named SSFYOLO, which is specifically tailored for complex forest scenarios. The proposed algorithm demonstrates excellent performance in trunk detection in complex environments, and has broad application prospects.
- object detection,
- deep learning,
- deep convolutional neural networks (CNN),
- feature extraction,
- computer vision,
- trunk detection,
- complex scenarios

FullText(HTML)

References (18)

References

[1]	罗海风, 谭月胜. 一种欠驱动双板分离式采摘器的研制[J]. 北京林业大学学报, 2018, 40(12): 110−116. Luo H F, Tan Y S. Research on an under actuated dual separation plate harvestor[J]. Journal of Beijing Forestry University, 2018, 40(12): 110−116.
[2]	刘慧, 朱晟辉, 沈跃, 等. 基于多特征融合的树干快速分割算法[J]. 农业机械学报, 2020, 51(1): 221−229. Liu H, Zhu S H, Shen Y, et al, Fast segmentation algorithm of tree trunks based on multi-feature fusion[J]. Transactions of the Chinese Society of Agricultural Machinery, 2020, 51(1): 221−229.
[3]	Chen X, Zhang B, Luo L. Multi-feature fusion tree trunk detection and orchard mobile robot localization using camera/ultrasonic sensors[J]. Computers and Electronics in Agriculture, 2018, 147: 91−108. doi: 10.1016/j.compag.2018.02.009
[4]	Zhao Z Q, Zheng P, Xu S, et al. Object detection with deep learning: a review[J]. IEEE Transactions on Neural Networks and Learning Systems, 2019, 30(11): 3212−3232. doi: 10.1109/TNNLS.2018.2876865
[5]	Girshick R, Donahue J, Darrell T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]// O’Conner L. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. New York: IEEE, 2014: 580−587.
[6]	Ren S, He K, Girshick R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. Advances in Neural Information Processing Systems, 2015, 28: 1137−1149.
[7]	Cai Z W, Nuno V. Cascade R-CNN: delving into high quality object detection[C]// O’Conner L. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Salt Lake City: IEEE. 2018: 6154−6162.
[8]	Huo A Q, Zhang W L, Li Y. Traffic sign recognition based on improved SSD model [C]//Cristina C. Proceedings of International Conference on Computer Network, Electronic and Automation (ICCNEA). Xi’an: IEEE, 2020: 54−58.
[9]	Redmon J, Divvala S, Girshick R, et al. You only look once: unified, real-time object detection[C]// O’Conner L. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. New York: IEEE, 2016: 779−788.
[10]	Zhang J, Tian M, Yang Z, et al. An improved target detection method based on YOLOv5 in natural orchard environments[J]. Computers and Electronics in Agriculture, 2024, 219: 108780. doi: 10.1016/j.compag.2024.108780
[11]	Lü F, Wang X, Li L, et al. Tree detection algorithm based on embedded YOLO lightweight network[J]. Journal of Shanghai Jiaotong University (Science), 2022, 29: 518−527.
[12]	Yang J F, Fu X Y, Hu Y W, et al. PanNet: a deep network architecture for pan-sharpening[C]//O’Conner L. Proceedings of the IEEE International Conference on Computer Vision. Venice: IEEE, 2017: 1753−1761.
[13]	Ghiasi G, Lin T Y. NAS-FPN: learning scalable feature pyramid architecture for object detection[C]// O’Conner L. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach: IEEE, 2019: 7029−7038.
[14]	Li Q F, Jia X, Zhou J C, et al. Rediscovering BCE loss for uniform classification[J/OL]. arXiv preprint arXiv, 2024 [2023−12−20]. https://arxiv.org/abs/2403.07289.
[15]	Tong Z J, Chen Y H, Xu Z W, et al. Wise-IoU: bounding box regression loss with dynamic focusing mechanism[J/OL]. arXiv preprint arXiv, 2023 [2024−03−20]. https://arxiv.org/abs/2301.10051.
[16]	Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need[C]// Curran Associates, Inc. Proceedings of the 31st International Conference on Neural Information Processing Systems. New York: Red Hook, 2017: 6000–6010.
[17]	Silva D, Santos F, Sousa A, et al. Visible and thermal image-based trunk detection with deep learning for forestry mobile robotics[J]. Journal of Imaging, 2021, 7(9): 176−180. doi: 10.3390/jimaging7090176
[18]	Silva D, Santos F, Sousa A, et al. Unimodal and multimodal perception for forest management: review and dataset[J]. Computation, 2021, 9(12): 127−132. doi: 10.3390/computation9120127

Cited By

Cited by

Periodical cited type(21)

1.	彭小静，黄海山，严芝银，邹星晨，贺康宁，程唱，王作枭，李睿，刘婧雯，石正阳，刘仟仟. 祁连山东部地区不同林分密度白桦天然林土壤理化性质特征. 生态学报. 2025(02): 743-756 .
2.	张佳凝，张建军，赖宗锐，赵炯昌，胡亚伟，李阳，卫朝阳. 林分密度对刺槐人工林土壤养分和微生物群落的影响. 干旱区研究. 2025(02): 274-288 .
3.	武燕，李歆玉，张奕婷，丁波，张运林，符裕红，刘讯. 西南喀斯特地区不同龄组马尾松人工林枯落物碳氮磷化学计量特征及其影响因子. 北京林业大学学报. 2024(02): 87-94 . 本站查看
4.	巩大鹏，毕华兴，王劲峰，赵丹阳，黄靖涵，宋艺琳. 晋西黄土区不同密度刺槐人工林叶片-枯落物-土壤化学计量特征. 林业科学研究. 2024(02): 156-164 .
5.	陈宇，庞涛，瞿相，彭建，杨汉波，代林利，辜云杰. 造林密度对楠木幼龄林生长、土壤理化性质与酶活性的影响. 四川林业科技. 2024(03): 9-20 .
6.	龚世豪，查同刚，张晓霞，张恒硕，高连炜，于洋. 晋西黄土区典型林分凋落物-土壤养分对降雨再分配变化的响应. 生态学报. 2024(17): 7748-7759 .
7.	窦金萍，武小钢，杨秀云，陈冠光，靳雅君，吴茜. 不同类型豆科植物群落凋落物对城市土壤质量的影响. 林业调查规划. 2024(05): 198-204 .
8.	贾亚倢，杨建英，张建军，胡亚伟，张犇，赵炯昌，李阳，唐鹏. 晋西黄土区林分密度对油松人工林生物量及土壤理化性质的影响. 浙江农林大学学报. 2024(06): 1211-1221 .
9.	陈涛，王露露，王思崇，朱学灵，叶永忠. 河南省丘陵低山区刺槐人工林立地分类及立地质量评价. 西北林学院学报. 2023(01): 153-159 .
10.	高利强，刘莹，王国梁. 人工和天然油松林表层土壤不同粒径团聚体有机碳及其组分分布特征. 水土保持学报. 2023(02): 320-328 .
11.	孙阔，袁兴中，王晓锋，袁嘉，候春丽，魏丽景. 三峡水库消落带土壤养分含量及生态化学计量特征. 长江流域资源与环境. 2023(02): 403-414 .
12.	张誉. 不同造林技术对水土保持林土壤特性的影响研究. 广东蚕业. 2023(03): 50-52 .
13.	钟欢，董文渊，浦婵，谢泽轩，张炜，郑静楠，夏莉. 滇东北4种类型筇竹林分土壤碳氮磷生态化学计量特征研究. 西南林业大学学报(自然科学). 2023(03): 111-119 .
14.	魏亚娟，刘美英，解云虎，李星. 吉兰泰盐湖防护体系建立38 a以来土壤养分特征. 干旱区研究. 2023(05): 747-755 .
15.	党记刚. 陕西黄土区典型人工林分结构与水土保持功能耦合关系研究. 科技创新与生产力. 2023(08): 47-50 .
16.	朱燕，翟博超，孙美美，罗伶书，王瑛，杜盛. 黄土丘陵区不同密度刺槐和油松人工林土壤理化性质与化学计量特征. 水土保持研究. 2023(06): 160-167 .
17.	兰道云，毕华兴，赵丹阳，王宁，云慧雅，王珊珊，崔艳红. 晋西黄土区不同密度油松人工林保育土壤功能评价. 水土保持学报. 2022(02): 189-196 .
18.	郭强，官凤英，辉朝茂，刘蔚漪，邹学明. 密度和施肥调控对巨龙竹新竹生长及生物量特征的影响. 北京林业大学学报. 2022(04): 95-106 . 本站查看
19.	张恒宇，孙树臣，吴元芝，安娟，宋红丽. 黄土高原不同植被密度条件下土壤水、碳、氮分布特征. 生态环境学报. 2022(05): 875-884 .
20.	郭鑫，魏天兴，陈宇轩，沙国良，任康，于欢. 黄土丘陵区典型退耕恢复植被土壤生态化学计量特征. 干旱区地理. 2022(06): 1899-1907 .
21.	梁广国，陶建元，郭坤，王子旗，张艳明，王金颖. 不同林型、不同林分密度植被下土壤养分及其化学计量比研究. 吉林林业科技. 2021(06): 14-21 .

Other cited types(13)

Get Citation

PDF

XML

Article views (189) PDF downloads (53) Cited by(34)

Turn off MathJax

Article Contents

Abstract

References

SSFYOLO: a trunk detection algorithm for complex forest scenarios

Abstract

References

Cited by

Periodical cited type(21)

Other cited types(13)

Catalog

Export File

Citation

Format

Content