3D reconstruction of citrus seedlings based on SLAM and NeRF

GUO Jun; YANG Dacheng; MO Zhenjie; LAN Yubin; ZHANG Yali

doi:10.7671/j.issn.1001-411X.202405030

Journal of South China Agricultural University > 2025 > 46(3): 429-438. > DOI: 10.7671/j.issn.1001-411X.202405030

GUO Jun, YANG Dacheng, MO Zhenjie, et al. 3D reconstruction of citrus seedlings based on SLAM and NeRF[J]. Journal of South China Agricultural University, 2025, 46(3): 429-438. DOI: 10.7671/j.issn.1001-411X.202405030

Citation:

PDF (1941 KB)

3D reconstruction of citrus seedlings based on SLAM and NeRF

1.
College of Engineering, South China Agricultural University, Guangzhou 510642, China
2.
National Center for International Collaboration Research on Precision Agricultural Aviation Pesticide Spraying Technology, Guangzhou 510642, China

More Information

Received Date: May 19, 2024
Available Online: March 02, 2025
Published Date: February 23, 2025

Graphical Abstract

Abstract

Abstract

Objective
Aiming at the problem that it is difficult to obtain the accurate 3D point cloud of citrus seedlings and their 3D phenotypic parameters to characterize the state of seedlings with the existing 3D reconstruction techniques, this paper proposes a method based on the simultaneous localization and mapping (SLAM) and neural radiance fields (NeRF) for 3D reconstruction of citrus seedlings.
Method
One-year old citrus seedlings were taken as the research object. Firstly, a depth sensor was used to capture the RGB map and depth map of the citrus seedling. Secondly, SLAM was employed to obtain the poses of the depth sensor in each frame of the image. Then, NeRF was trained for citrus seedlings, and the multi-view images with attached positional pose were fed into the multilayer erceptron (MLP). Finally, through supervised training with volume rendering, a high-precision 3D realistic point cloud model of citrus seedlings was reconstructed.
Result
The 3D model of citrus seedlings reconstructed by this method was highly realistic in terms of color and texture, with clear contours and distinct layers, and had real-world level accuracy. Based on this model, the 3D phenotypic parameters of citrus seedlings could be effectively extracted with the accuracy of 97.94% for plant height, 93.95% for breadth length, 94.11% for breadth width and 97.62% for stem thickness.
Conclusion
This study helps to accelerate the selection and nursery process of excellent citrus seedlings and provides a technical support for the sustainable development of the citrus industry.
- Citrus seedling,
- Plant 3D phenotype,
- 3D reconstruction,
- Neural radiance fields (NeRF),
- Simultaneous localization and mapping (SLAM),
- Deep learning

FullText(HTML)

References (21)

References

[1]	柴宏红, 邵科, 于超, 等. 基于三维点云的甜菜根表型参数提取与根型判别[J]. 农业工程学报, 2020, 36(10): 181-188. doi: 10.11975/j.issn.1002-6819.2020.10.022
[2]	PAULUS S, SCHUMANN H, KUHLMANN H, et al. High-precision laser scanning system for capturing 3D plant architecture and analysing growth of cereal plants[J]. Biosystems Engineering, 2014, 121: 1-11.
[3]	LADANIYA M S, MARATHE R A, DAS A K, et al. High density planting studies in acid lime (Citrus aurantifolia Swingle)[J]. Scientia Horticulturae, 2020, 261: 108935. doi: 10.1016/j.scienta.2019.108935
[4]	CONTI G, XOCONOSTLE-CÁZARES B, MARCELINO-PÉREZ G, et al. Citrus genetic transformation: An overview of the current strategies and insights on the new emerging technologies[J]. Frontiers in Plant Science, 2021, 12: 768197. doi: 10.3389/fpls.2021.768197
[5]	BISI R B, ALBRECHT U, BOWMAN K D. Seed and seedling nursery characteristics for 10 USDA Citrus rootstocks[J]. HortScience, 2020, 55(4): 528-532. doi: 10.21273/HORTSCI14598-19
[6]	SHAKOOR N, LEE S, MOCKLER T C. High throughput phenotyping to accelerate crop breeding and monitoring of diseases in the field[J]. Current Opinion in Plant Biology, 2017, 38: 184-192. doi: 10.1016/j.pbi.2017.05.006
[7]	黄成龙, 李曜辰, 骆树康, 等. 基于结构光三维点云的棉花幼苗叶片性状解析方法[J]. 农业机械学报, 2019, 50(8): 243-248.
[8]	何海清, 严椰丽, 凌梦云, 等. 结合三维密集点云的无人机影像大豆覆盖度提取[J]. 农业工程学报, 2022, 38(2): 201-209. doi: 10.11975/j.issn.1002-6819.2022.02.023
[9]	MA H, ZHU X, JI J, et al. Rapid estimation of apple phenotypic parameters based on 3D reconstruction[J]. International Journal of Agricultural and Biological Engineering, 2021, 14(5): 180-188. doi: 10.25165/j.ijabe.20211405.6258
[10]	CHAWADE A, VAN HAM J, BLOQUIST H, et al. High-throughput field-phenotyping tools for plant breeding and precision agriculture[J]. Agronomy, 2019, 9(5): 258. doi: 10.3390/agronomy9050258
[11]	杨会君, 王瑞萍, 王增莹, 等. 基于多视角图像的作物果实三维表型重建[J]. 南京师范大学学报(自然科学版), 2021, 44(2): 92-103.
[12]	梁秀英, 周风燃, 陈欢, 等. 基于运动恢复结构的玉米植株三维重建与性状提取[J]. 农业机械学报, 2020, 51(6): 209-219. doi: 10.6041/j.issn.1000-1298.2020.06.022
[13]	HE W, YE Z, LI M, et al. Extraction of soybean plant trait parameters based on SfM-MVS algorithm combined with GRNN[J]. Frontiers in Plant Science, 2023, 14: 1181322. doi: 10.3389/fpls.2023.1181322
[14]	LEI S, LIU L, XIE Y, et al. 3D visualization technology for rubber tree forests based on a terrestrial photogrammetry system[J]. Frontiers in Forests and Global Change, 2023, 6: 1206450. doi: 10.3389/ffgc.2023.1206450
[15]	MILDENHALL B, SRINIVASAN P P, TANCIK M, et al. NeRF: Representing scenes as neural radiance fields for view synthesis[J]. Communications of the ACM, 2021, 65(1): 99-106.
[16]	宫金良, 刘镔霄, 魏鹏, 等. 基于相机位姿恢复与神经辐射场理论的果树三维重建方法[J]. 农业工程学报, 2023, 39(22): 157-165. doi: 10.11975/j.issn.1002-6819.202307275
[17]	KANG H, HU K, YING W, et al. High-fidelity 3d reconstruction of plants using neural radiance field[J]. Computer and Electronics in Agriculture, 2024, 220: 108848. doi: 10.1016/j.compag.2024.108848
[18]	IGLHAUT J, CABO C, PULITI S, et al. Structure from motion photogrammetry in forestry: A review[J]. Current Forestry Reports, 2019, 5(3): 155-168. doi: 10.1007/s40725-019-00094-3
[19]	ZHANG Z. A flexible new technique for camera calibration[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 22(11): 1330-1334.
[20]	ZHAO C, FAN C, ZHAO Z. A binocular camera calibration method based on circle detection[J]. Heliyon, 2024, 10(19): e38347. doi: 10.1016/j.heliyon.2024.e38347.
[21]	彭程, 苗艳龙, 汪刘洋, 等. 基于三维点云的田间香蕉吸芽形态参数获取[J]. 农业工程学报, 2022, 38(S1): 193-200. doi: 10.11975/j.issn.1002-6819.2022.z.022

Cited By

Cited by

Periodical cited type(6)

1.	贾鹿，赵磊，凌飞，李广亚. 基于归一化RBFNN的油井动液面测量数据异常辨识. 电子测量技术. 2024(24): 188-194 .
2.	李志臣，凌秀军，李鸿秋，李志军. 基于改进ShuffleNet的板栗分级方法. 山东农业大学学报(自然科学版). 2023(02): 299-307 .
3.	郑显润，郑鹏，王文秀，程亚红，苏宇锋. 基于多尺度特征提取深度残差网络的水稻害虫识别. 华南农业大学学报. 2023(03): 438-446 . 本站查看
4.	邓雪阳，邓达平，苏万靖. 基于并行深度卷积神经网络的舰船通信异常数据检测研究. 舰船科学技术. 2023(15): 119-122 .
5.	杨亚琦，李博雄，杨东霞，刘燕. 基于信息熵的异常数据判别方法. 科学技术创新. 2023(24): 194-199 .
6.	杜肖鹏，李恺，王春辉，侯永，尹义蕾，潘守江，张凌风. 国内温室空气温湿度检测及传输技术研究进展. 农业工程技术. 2022(34): 28-34 .

Other cited types(1)

Get Citation

PDF

XML

Article views PDF downloads Cited by(7)

Turn off MathJax

Article Contents

Abstract

References

3D reconstruction of citrus seedlings based on SLAM and NeRF

Abstract

References

Cited by

Periodical cited type(6)

Other cited types(1)

Catalog

Export File

Citation

Format

Content