Unlocking plant secrets: A systematic review of 3D imaging in plant phenotyping techniques

被引:15
作者
Akhtar, Muhammad Salman [1 ]
Zafar, Zuhair [1 ]
Nawaz, Raheel [2 ]
Fraz, Muhammad Moazam [1 ]
机构
[1] Natl Univ Sci & Technol NUST, Sect H-12, Islamabad 44000, Pakistan
[2] Staffordshire Univ, Stoke On Trent ST4 2DE, England
关键词
3D imaging; Point cloud; Plant phenotyping; Morphological traits; Computer vision; POINT CLOUD; SEGMENTATION; PHENOMICS; TECHNOLOGIES; BIOMASS; MODEL; TOOL;
D O I
10.1016/j.compag.2024.109033
中图分类号
S [农业科学];
学科分类号
09 ;
摘要
Phenotyping is a systematic process of quantifying and assessing a wide range of structural and physiological traits to understand the intricate interplay between an organism's genetic makeup, its surrounding environment, and management practices, often referred to as genome-to-environment (GxE) interaction. In the context of plants, these traits can include aspects such as plant height, stem diameter, leaf size, angle, and shape, chlorophyll content, biomass, leaf area, etc. 3D plant phenotyping plays a crucial role in advancing our understanding of plant biology, improving crop breeding, and agricultural practices. 3D imaging has become a powerful phenotyping tool, offering in-depth insights into plant structures and traits. In contrast to 2D imaging, 3D imaging enables precise measurement of plant traits that cannot be sufficiently evaluated in two dimensions by overcoming challenges such as partial occlusion through the utilization of depth perception and multiple viewpoints. However, even with significant recent progress, various challenges persist, including the need for well-designed experimental setups for standardized data collection, the automation of processing pipelines, and the robust analysis techniques of 3D representations, which still impede the widespread adoption of 3D plant phenotyping. To propel the progress of 3D imaging-based phenotyping, an all-encompassing assessment of existing strategies is imperative, yet there is currently a lack of specialized reviews that scrutinize and emphasize distinct facets for future enhancement. To bridge this gap, we perform a systematic survey of 81 research studies that employ 3D imaging for various trait assessments of plants. Our review thoroughly investigates the stages of data acquisition, encompassing sensing technologies, representations, preprocessing approaches, analysis methodologies, and techniques for estimating phenotypic traits. We believe that this comprehensive review will serve as a valuable guide for researchers and professionals engaged in high throughput plant phenotyping, equipping them to formulate effective experimental setups and utilize appropriate processing and analysis methods, thereby fostering its continued advancement.
引用
收藏
页数:28
相关论文
共 139 条
[41]   PlantNet: A dual-function point cloud segmentation network for multiple plant species [J].
Li, Dawei ;
Shi, Guoliang ;
Li, Jinsheng ;
Chen, Yingliang ;
Zhang, Songyin ;
Xiang, Shiyu ;
Jin, Shichao .
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING, 2022, 184 :243-263
[42]   A Leaf Segmentation and Phenotypic Feature Extraction Framework for Multiview Stereo Plant Point Clouds [J].
Li, Dawei ;
Shi, Guoliang ;
Kong, Weijian ;
Wang, Sifan ;
Chen, Yang .
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, 2020, 13 :2321-2336
[43]   An Overlapping-Free Leaf Segmentation Method for Plant Point Clouds [J].
Li, Dawei ;
Cao, Yan ;
Shi, Guoliang ;
Cai, Xin ;
Chen, Yang ;
Wang, Sifan ;
Yan, Siyuan .
IEEE ACCESS, 2019, 7 :129054-129070
[44]   Automatic Branch-Leaf Segmentation and Leaf Phenotypic Parameter Estimation of Pear Trees Based on Three-Dimensional Point Clouds [J].
Li, Haitao ;
Wu, Gengchen ;
Tao, Shutian ;
Yin, Hao ;
Qi, Kaijie ;
Zhang, Shaoling ;
Guo, Wei ;
Ninomiya, Seishi ;
Mu, Yue .
SENSORS, 2023, 23 (09)
[45]   Towards end-to-end deep RNN based networks to precisely regress of the lettuce plant height by single perspective sparse 3D point cloud [J].
Li, Jingsong ;
Wang, Ying ;
Zheng, LiHua ;
Zhang, Man ;
Wang, Minjuan .
EXPERT SYSTEMS WITH APPLICATIONS, 2023, 229
[46]   A Review of Imaging Techniques for Plant Phenotyping [J].
Li, Lei ;
Zhang, Qin ;
Huang, Danfeng .
SENSORS, 2014, 14 (11) :20078-20111
[47]   Automatic organ-level point cloud segmentation of maize shoots by integrating high-throughput data acquisition and deep learning [J].
Li, Yinglun ;
Wen, Weiliang ;
Miao, Teng ;
Wu, Sheng ;
Yu, Zetao ;
Wang, Xiaodong ;
Guo, Xinyu ;
Zhao, Chunjiang .
COMPUTERS AND ELECTRONICS IN AGRICULTURE, 2022, 193
[48]   A review of computer vision technologies for plant phenotyping [J].
Li, Zhenbo ;
Guo, Ruohao ;
Li, Meng ;
Chen, Yaru ;
Li, Guangyao .
COMPUTERS AND ELECTRONICS IN AGRICULTURE, 2020, 176
[49]   LiDAR: An important tool for next-generation phenotyping technology of high potential for plant phenomics? [J].
Lin, Yi .
COMPUTERS AND ELECTRONICS IN AGRICULTURE, 2015, 119 :61-73
[50]   Hyperspectral imaging and 3D technologies for plant phenotyping: From satellite to close-range sensing [J].
Liu, Huajian ;
Bruning, Brooke ;
Garnett, Trevor ;
Berger, Bettina .
COMPUTERS AND ELECTRONICS IN AGRICULTURE, 2020, 175