Scientia Silvae Sinicae ›› 2024, Vol. 60 ›› Issue (3): 121-130.doi: 10.11707/j.1001-7488.LYKX20220597
• Research papers • Previous Articles Next Articles
Wenhan Yang(),Tianyu Liu*,Junchi Zhou,Wenwu Hu,Ping Jiang
Received:
2022-09-03
Online:
2024-03-25
Published:
2024-04-08
Contact:
Tianyu Liu
E-mail:Yangwenhan@stu.hunau.edu.cn
CLC Number:
Wenhan Yang,Tianyu Liu,Junchi Zhou,Wenwu Hu,Ping Jiang. CNN-Swin Transformer Detection Algorithm of Forest Wildlife Images Based on Improved YOLOv5s[J]. Scientia Silvae Sinicae, 2024, 60(3): 121-130.
Table 1
Distribution of the number of nine forest wildlife species in the dataset"
种类 Species | 原图数量 Number of original images | 融合图像数量 Number of image fusions | 总数 Total |
亚洲黑熊 Ursus thibetanus | 525 | 525 | 10 072 |
豹猫 Prionailurus bengalensis | 327 | 327 | |
猕猴 Macaca mulatta | 162 | 162 | |
毛冠鹿 Elaphodus cephalophus | 93 | 93 | |
短尾猫 Lynx rufus | 647 | 647 | |
黑尾鹿 Odocoileus hemionus | 1 065 | 1 065 | |
浣熊 Procyon lotor | 655 | 655 | |
红松鼠 Tamiasciurus hudsonicus | 804 | 804 | |
赤狐 Vulpes vulpes | 758 | 758 |
Table 2
Results of ablation experiments"
组别 Group | 网络模型 Model | 平均准确率 P | 平均回归率 R | 均值平均精度 mAP@0.5(%) | 检测速度 Detection speed/FPS |
1 | YOLOv5s | 0.83 | 0.66 | 74.1 | 45 |
2 | YOLOv5s+CutMixE | 0.81 | 0.68 | 76.7 | 45 |
3 | YOLOv5s+CutMixE+ConcatE | 0.85 | 0.74 | 80.4 | 45 |
4 | YOLOv5s+CutMixE+SwinTR | 0.83 | 0.75 | 84.5 | 36 |
5 | YOLOv5s+CutMixE+ConcatE+SwinTR | 0.85 | 0.82 | 86.8 | 36 |
6 | YOLOv5s+CutMixE+ConcatE+SwinTR+α-DIoU | 0.86 | 0.83 | 88.4 | 36 |
Table 3
Comparison of test results for different animal species"
类别 Sort | YOLOv5s AP AP of YOLOv5s(%) | 改进算法P P of improved algorithm | 改进算法R R of improved algorithm | 改进算法AP AP of improved algorithm (%) |
亚洲黑熊Ursus thibetanus | 83.0 | 0.95 | 0.96 | 93.6 |
豹猫Prionailurus bengalensis | 69.4 | 0.88 | 0.84 | 85.9 |
猕猴Macaca mulatta | 69.1 | 0.88 | 0.85 | 86.1 |
毛冠鹿Elaphodus cephalophus | 68.8 | 0.72 | 0.82 | 80.4 |
短尾猫Lynx rufus | 72.1 | 0.89 | 0.86 | 87.0 |
黑尾鹿Odocoileus hemionus | 81.5 | 0.88 | 0.90 | 90.6 |
浣熊Procyon lotor | 78.5 | 0.95 | 0.90 | 91.4 |
红松鼠Tamiasciurus hudsonicus | 72.0 | 0.94 | 0.89 | 90.7 |
赤狐Vulpes vulpes | 72.5 | 0.90 | 0.89 | 89.9 |
李 佳, 刘 芳, 李迪强, 等. 2017. 基于红外相机监测分析的红腹角雉日活动节律. 林业科学, 53(7): 170−176. | |
Li J, Liu F, Li D Q, et al. 2017. Daily activity rhythm of Temminick’s Tragopan (Trgopan temminckii) based on infrared camera monitoring. Journal of Computer Applications. [in Chinese] | |
李 果, 李俊生, 关 潇, 等. 2014. 生物多样性监测技术手册. 北京: 中国环境科学出版社. | |
Li G, Li J S, Guan X, et al. 2014. Biodiversity monitoring technical manuals. Beijing: China Environmental Science Press. [in Chinese] | |
Ali H, Vishwesh N, Yucheng T, et al. 2022. Swin unetr: swin transformers for semantic segmentation of brain tumors in mri images. International MICCAI Brainlesion Workshop, Springer, 272–284. | |
Bochkovskiy A, Wang C Y, Liao H Y M. 2020. YOLOv4: optimal speed and accuracy of object detection. arXiv Preprint arXiv: 2004.10934. | |
Chen R, Little R, Mihaylova L, et al. Forest wildlife surveillance using deep learning methods. Ecology and Evolution, 2019, 9 (17): 9453- 9466.
doi: 10.1002/ece3.5410 |
|
Chen G, Han T X, He Z, et al. 2014. Deep convolutional neural network based species recognition for wild animal monitoring. 2014 IEEE International Conference on Image Processing (ICIP), IEEE, 858−862. | |
Carion N, Massa F, Synnaeve G, et al. 2020. End-to-end object detection with transformers. European Conference on Computer Vision (ECCV), 12346: 213–229. | |
DeVries T, Taylor G W. 2017. Improved regularization of convolutional neural networks with cutout. arXiv Preprint arXiv: 1708.04552. | |
Dosovitskiy A, Beyer L, Kolesnikov A, et al. 2020. An image is worth 16×16 words: transformers for image recognition at scale. arXiv Preprint arXiv: 2010.11929. | |
Girshick R. 2015. Fast R-CNN. Proceedings of the IEEE International Conference on Computer Vision, 1440−1448. | |
Girshick R, Donahue J, Darrell T, et al. 2014. Rich feature hierarchies for accurate object detection and semantic segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 580−587. | |
He K, Zhang X, Ren S, et al. Spatial pyramid pooling in deep convolutional networks for visual recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37 (9): 1904- 1916.
doi: 10.1109/TPAMI.2015.2389824 |
|
He J, Erfani S, Ma X, et al. 2021. Alpha-IoU: a family of power intersection over union losses for bounding box regression. arXiv Preprint arXiv: 2110.13675. | |
Han K, Wang Y, Chen H, et al. A survey on vision transformer. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022, 45 (1): 87- 110. | |
Jannat F E, Willis A R. 2022. Improving classification of remotely sensed images with the Swin transformer. SoutheastCon 2022, IEEE, 611–618. | |
Li Y, Mao H, Girshick R, et al. 2022. Exploring plain vision transformer backbones for object detection. Computer Vision–ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23–27, 2022. Proceedings, Part IX. Cham: Springer Nature Switzerland, 280−296. | |
Lin T Y, Goyal P, Girshick R, et al. 2017. Focal loss for dense object detection. Proceedings of the IEEE International Conference on Computer Vision, 2980−2988. | |
Lin T Y, Maire M, Belongie S, et al. 2014. Microsoft coco: common objects in context. European Conference on Computer Vision, Springer, 740–755. | |
Liu T, Ma Y, Yang W, et al. Spatial-temporal interaction learning based two-stream network for action recognition. Information Sciences, 2022, 606, 864- 876. | |
Liu W, Anguelov D, Erhan D, et al. Ssd: single shot multibox detector. European Conference on Computer Vision. Springer, 2016, Cham, 21- 37. | |
Liu Z, Tan Y, He Q, et al. Swinnet: Swin transformer drives edge-aware rgb-d and rgb-t salient object detection. IEEE Transactions on Circuits and Systems for Video Technology, 2021, 32 (7): 4486- 4497. | |
Liu Z, Lin Y, Cao Y, et al. 2021. Swin transformer: Hierarchical vision transformer using shifted windows. Proceedings of the IEEE/CVF International Conference on Computer Vision, 10012−10022. | |
Naseer M M, Ranasinghe K, Khan S H, et al. Intriguing properties of vision transformers. Advances in Neural Information Processing Systems, 2021, 34, 23296- 23308. | |
Khan S, Naseer M, Hayat M, et al. Transformers in vision: a survey. ACM Computing Surveys (CSUR), 2022, 54 (10s): 1- 41. | |
Norouzzadeh M S, Nguyen A, Kosmala M, et al. Automatically identifying, counting, and describing wild animals in camera-trap images with deep learning. Proceedings of the National Academy of Sciences, 2018, 115 (25): E5716- E5725. | |
Redmon J, Divvala S, Girshick R, et al. 2016. You only look once: unified, real-time object detection. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 779−788. | |
Redmon J, Farhadi A. 2017. YOLO9000: better, faster, stronger. Proceedings of the IEEE Conference on Computer vision and Pattern Recognition. 7263−7271. | |
Redmon J, Farhadi A. 2018. Yolov3: an incremental improvement. arXiv Preprint arXiv: 1804.02767. | |
Ren S, He K, Girshick R, et al. 2015. Faster r-cnn: towards real-time object detection with region proposal networks. Advances in Neural Information Processing Systems, 28. | |
Sermanet P, Eigen D, Zhang X, et al. 2013. Overfeat: integrated recognition, localization and detection using convolutional networks. arXiv Preprint arXiv: 1312.6229. | |
Schneider T C, Kowalczyk R, Köhler M. 2013.Resting site selection by large herbivores―the case of European bison (Bison bonasus) in Biaowieza Primeval Forest. Mammalian Biology, 78(6): 438−445. | |
Villa A G, Salazar A, Vargas F. Towards automatic wild animal monitoring: identification of animal species in camera-trap images using very deep convolutional neural networks. Ecological Informatics, 2017, 41, 24- 32.
doi: 10.1016/j.ecoinf.2017.07.004 |
|
Yun S, Han D, Oh S J, et al. 2019. Cutmix: Regularization strategy to train strong classifiers with localizable features. Proceedings of the IEEE/CVF International Conference on Computer Vision, 6023−6032. | |
Zheng Z, Wang P, Liu W, et al. 2020. Distance-IoU loss: faster and better learning for bounding box regression. Proceedings of the AAAI Conference on Artificial Intelligence. 34(7): 12993−13000. |
[1] | Yadong Xue,Diqiang Li,Jia Li. Habitat Selection and Migration Pattern of Wild Bactrian Camel (Camelus ferus) in the Kumtag Desert, China Based on Satellite Tracking and Positioning Technology [J]. Scientia Silvae Sinicae, 2020, 56(10): 192-198. |
[2] | Jinhua Mo,Jia Li,Fang Liu,Xiaoguan Li,Diqiang Li. A Survey of Mammals and Birds Diversity in Jianfengling District of Hainan Province by Using Camera-Trapping [J]. Scientia Silvae Sinicae, 2019, 55(10): 203-210. |
[3] | Kong Weiyao, Sun Quan, Liu Xinxin, Qu Li, Wang Fuyou, Yao Mingyuan, Zou Hongfei. Population Dynamic of Far Eastern Leopard(Panthera pardus orientalis) in Wangqing Nature Reserve Based on Infrared Camera Monitoring [J]. Scientia Silvae Sinicae, 2019, 55(5): 188-196. |
[4] | Huang Heqing, Chu Hongjun, Cao Jie, Bu Lan, Hu Defu, Zhang Dong, Li Kai. Distribution of Gasterophilus (Diptera, Gasterophilidae) Myiasis Foci in Arid Desert Steppe:A Case Study of Kalamaili Mountain Ungulate Nature Reserve [J]. Scientia Silvae Sinicae, 2017, 53(11): 142-149. |
[5] | Wang Wengting, Xiao Sa, Huang Heqing, Li Kai, Zhang Dong, Chu Hongjun, Guo Youqing, Gao Wanli. Diversity and Infection of Gasterophilus spp. in Mongol-Xinjiang Region and Qinghai Tibet Region [J]. Scientia Silvae Sinicae, 2016, 52(2): 134-139. |
[6] | Wang Wenting, Zhang Dong, Hu Defu, Chu Hongjun, Cao Jie, Ge Yan, Aierken Jilili, Li Kai. Analysis of the Main Etiology of Gasterophilosis in Przewalski's Horse in Xinjiang [J]. Scientia Silvae Sinicae, 2014, 50(11): 90-95. |
[7] | Sun Feixiang;Dang Kunliang;Chen Junxian. Relationship between Habitat Selection of Giant Panda and Forest Community Characters in Qinling Mountains [J]. , 2013, 49(5): 147-153. |
[8] | Qi Lei;Hu Defu;Ding Changqing;Sui Jinling;Zhang Dong;Yang Liang;Wu Jigui;Jiang Wanjie. Rats Community Structure and Diversity in the Songshan National Nature Reserve, Beijing [J]. Scientia Silvae Sinicae, 2012, 48(9): 181-185. |
[9] | Wang Yanying;Wang Cheng;Qie Guangfa;Dong Jianhua;Jiang Jihong. Effect of VOCs from Branch and Leaf of Platycladus orientalis on Locomotor Activity in Mice [J]. Scientia Silvae Sinicae, 2011, 47(12): 97-100. |
[10] | Tie Jun;Zhang Jing;Peng Linpeng;Zhao Benyuan;Zhang Zhixiang;. Analysis of Main Factors Influencing Summer and Autumn Feeding of Rhinopithecus roxellana in Shennongjia Nature Reserve [J]. Scientia Silvae Sinicae, 2011, 47(7): 108-115. |
[11] | Han Zongxian;Wang Wei;Hu Jinchu. Habitat Selection by Francois' Langur in Jinfo Mountain in Spring [J]. Scientia Silvae Sinicae, 2011, 47(4): 121-128. |
Viewed | ||||||
Full text |
|
|||||
Abstract |
|
|||||