Automatic Detection of Dead Trees Based on Lightweight YOLOv4 and UAV Imagery
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Dead trees significantly impact forest production and the ecological environment and pose constraints to thesustainable development of forests. A lightweight YOLOv4 dead tree detection algorithm based on unmannedaerial vehicle images is proposed to address current limitations in dead tree detection that rely mainly oninefficient, unsafe and easy-to-miss manual inspections. An improved logarithmic transformation method wasdeveloped in data pre-processing to display tree features in the shadows. For the model structure, the originalCSPDarkNet-53 backbone feature extraction network was replaced by MobileNetV3. Some of the standardconvolutional blocks in the original extraction network were replaced by depthwise separable convolutionblocks. The new ReLU6 activation function replaced the original LeakyReLU activation function to make thenetwork more robust for low-precision computations. The K-means++ clustering method was also integratedto generate anchor boxes that are more suitable for the dataset. The experimental results show that the improvedalgorithm achieved an accuracy of 97.33%, higher than other methods. The detection speed of the proposedapproach is higher than that of YOLOv4, improving the efficiency and accuracy of the detection process.
Abstract
Dead trees significantly impact forest production and the ecological environment and pose constraints to thesustainable development of forests. A lightweight YOLOv4 dead tree detection algorithm based on unmannedaerial vehicle images is proposed to address current limitations in dead tree detection that rely mainly oninefficient, unsafe and easy-to-miss manual inspections. An improved logarithmic transformation method wasdeveloped in data pre-processing to display tree features in the shadows. For the model structure, the originalCSPDarkNet-53 backbone feature extraction network was replaced by MobileNetV3. Some of the standardconvolutional blocks in the original extraction network were replaced by depthwise separable convolutionblocks. The new ReLU6 activation function replaced the original LeakyReLU activation function to make thenetwork more robust for low-precision computations. The K-means++ clustering method was also integratedto generate anchor boxes that are more suitable for the dataset. The experimental results show that the improvedalgorithm achieved an accuracy of 97.33%, higher than other methods. The detection speed of the proposedapproach is higher than that of YOLOv4, improving the efficiency and accuracy of the detection process.
참고문헌 (24)
- Agnieszka Kamińska, "Species-related single dead tree detection using multi-temporal ALS data and CIR imagery", Remote Sensing of Environment, 219, pp.31~43 (2018), Elsevier BV[DOI]
- Kaori Otsu, "Estimating the Threshold of Detection on Tree Crown Defoliation Using Vegetation Indices from UAS Multispectral Imagery", Drones, 3(4), pp.80~ (2019), MDPI AG[DOI]
- Salim Malek, "Efficient Framework for Palm Tree Detection in UAV Images", IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 7(12), pp.4692~4703 (2014), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Weijia Li, "Deep Learning Based Oil Palm Tree Detection and Counting for High-Resolution Remote Sensing Images", Remote Sensing, 9(1), pp.22~ (2016), MDPI AG[DOI]
- María Culman, "Individual Palm Tree Detection Using Deep Learning on RGB Imagery to Support Tree Inventory", Remote Sensing, 12(21), pp.3476~ (2020), MDPI AG[DOI]
- Emilio Guirado, "Deep-learning Versus OBIA for Scattered Shrub Detection with Google Earth Imagery: Ziziphus lotus as Case Study", Remote Sensing, 9(12), pp.1220~ (2017), MDPI AG[DOI]
- Huan Tao, "Deep learning-based dead pine tree detection from unmanned aerial vehicle images", International Journal of Remote Sensing, 41(21), pp.8238~8255 (2020), Informa UK Limited[DOI]
- Run Yu, "Early detection of pine wilt disease using deep learning algorithms and UAV-based multispectral imagery", Forest Ecology and Management, 497, pp.119493~ (2021), Elsevier BV[DOI]
- Mohamad Haniff Junos, "Automatic detection of oil palm fruits from UAV images using an improved YOLO model", The Visual Computer, 38(7), pp.2341~2355 (2021), Springer Science and Business Media LLC[DOI]
- Xinni Liu, "Automatic Detection of Oil Palm Tree from UAV Images Based on the Deep Learning Method", Applied Artificial Intelligence, 35(1), pp.13~24 (2020), Informa UK Limited[DOI]
- Kanitta Yarak, "Oil Palm Tree Detection and Health Classification on High-Resolution Imagery Using Deep Learning", Agriculture, 11(2), pp.183~ (2021), MDPI AG[DOI]
- R. Girshick, "Fast R-CNN", Proceedings of the IEEE International Conference on Computer Vision, pp.1440~1448 (2015)
- Alexander Buslaev, "Albumentations: Fast and Flexible Image Augmentations", Information, 11(2), pp.125~ (2020), MDPI AG[DOI]
- Xueguo Wang, "Change detection in SAR images based on the logarithmic transformation and total variation denoising method", Remote Sensing Letters, 8(3), pp.214~223 (2016), Informa UK Limited[DOI]
- Midhun P. Mathew, "Leaf-based disease detection in bell pepper plant using YOLO v5", Signal, Image and Video Processing, 16(3), pp.841~847 (2021), Springer Science and Business Media LLC[DOI]
- Mark Everingham, "The Pascal Visual Object Classes (VOC) Challenge", International Journal of Computer Vision, 88(2), pp.303~338 (2009), Springer Science and Business Media LLC[DOI]
- "YOLOv4: optimal speed and accuracy of object detection" (2020)[URL]
- A. Howard, "Searching for MobileNetv3", Proceedings of the IEEE, 2019, 1314~1324
- "MobileNets: efficient convolutional neural networks for mobile vision applications" (2017)[URL]
- M. Sandler, "MobileNetv2: inverted residuals and linear bottlenecks", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.4510~4520 (2018)
- J. Hu, "Squeeze-and-excitation networks", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.7132~7141 (2018)
- D. Arthur, "K-means++ the advantages of careful seeding", Proceedings of the 18th Annual ACM-SIAM Symposium on Discrete Algorithms, pp.1027~1035 (2007)
- Cenk Donmez, "Computer vision-based citrus tree detection in a cultivated environment using UAV imagery", Computers and Electronics in Agriculture, 187, pp.106273~ (2021), Elsevier BV[DOI]
- W. Liu, "SSD: single shot multibox detector", Computer Vision–ECCV 2016, pp.21~37 (2016), Springer