
Intelligent Recognition and Three-Dimensional Reconstruction of Crop Phenotypes: A Review
ZHANGJianhua, YAOQiong, ZHOUGuomin, WUWendi, XIUXiaojie, WANGJian
Intelligent Recognition and Three-Dimensional Reconstruction of Crop Phenotypes: A Review
Significance The crop phenotype represents the external expression of the interaction between crop genes and the environment. It is the manifestation of the physiological, ecological and dynamic characteristics of crop growth and development and represents a core link within the field of intelligent breeding. Systematic analysis of crop phenotypes can not only provide insight into the function of genes and reveal the genetic factors that affect the key characteristics of crops, but also can be used to effectively utilize germplasm resources and breed varieties with major breakthroughs. The utilization of data-driven, intelligent, dynamic and non-contact crop phenotypic measurement enables the acquisition of key traits and phenotypic parameters of crop growth, thereby furnishing crucial data support for the breeding and identification of breeding materials throughout the entire growth cycle of crops. Progress Crop phenotype acquisition equipment represents the fundamental basis for the acquisition, analysis, measurement and identification of crop phenotypes. Such equipment can be employed monitor the growth status of crops in detail. The functions, performance and applications of the dominant high-throughput crop phenotyping platform, along with an analysis of the characteristics of various sensing and imaging devices employed obtain crop phenotypic information are presented. The rapid development of high-throughput crop phenotyping platforms and perceptual imaging equipment has led to the integration of advanced imaging technology, spectroscopy technology and deep learning algorithms. These technologies enable the automatic and high-throughput acquisition of yield, resistance, quality and other related traits of large-scale crops, as well as the generation of large-scale multi-dimensional, multi-scale and multi-modal crop phenotypic data. This supports the rapid development of crop phenomics. The research progress of deep learning in the intelligent perception of crop agronomic traits and morphological structure, with respect to various phenotypes such as crop plant height, leaf area index, and crop organ detection is presented. Additionally, the main challenges associated with this field are outlined, namely the complexity of environmental influences, the difficulty of large-scale data processing due to data diversity, model generalization issues, and the need for lightweight algorithms. The analysis of crop phenotypes and morphological characteristics based on three-dimensional reconstruction technology is considered to be more accurate than that based on two-dimensional images. A summary and discussion of the three-dimensional reconstruction method for crops is provided, and the main challenges encountered are also outlined, including the complexity of crop structures, the necessity of algorithm optimization and the cost and practicability of the method. Conclusions and Prospects It is devoted to the examination of the difficulties and challenges associated with the intelligent identification of crop phenotypes based on deep learning, from the perspective of research and development of innovative field equipment for the acquisition and analysis of phenotypic data, the establishment of a unified data acquisition and data sharing platform with the objective of improving the efficiency of data utilization, the enhancement of the generality aforementioned approach. A field crop phenotype intelligent identification model must consider multiple perspectives, modalities and points in time. This necessitates a continuous, multi-faceted analysis. It is achieved to identify characteristics in a spatiotemporal context through the fusion of various data sources, such as images, spectral data and weather information. In terms of interpretability models, it explores the potential of deep learning in crop phenotype intelligent recognition. It will be necessary for future research to break through the current bottleneck of high-throughput crop phenomics technology. It is vital to conduct further research into the field of visual perception and deep learning methods. This will allow for the realization of the intelligent acquisition of crop phenotypic information, as well as an intelligent management of phenotypic data.
crop intelligent perception / phenotypic recognition / organ detection and technology / deep learning / 3D reconstruction / morphometry / large models {{custom_keyword}} /
Table 1 Research on intelligent perception of crop morphology based on deep learning in recent years表1 近年作物三维点云作物形态智能感知研究情况 |
编号 | 作者 | 作物种类 | 测量指标 | 重构方法 | 准确率R2/% |
---|---|---|---|---|---|
1 | Ma 等[51] | 大豆 | 株高,叶面积指数 | beer-lambert law | > 94.00 |
2 | Wu等[52] | 小麦 | 株高,叶面积,投影面积,枝条体积和致密度 | MVS-Pheno V2 | > 96.00 |
3 | Zhang等[53] | 玉米、大豆 | 叶片倾斜角度和方向测量 | 3D calibration approach | 76.00、94.00 |
4 | Ao等[54] | 玉米 | 茎位置精度,株高,冠宽和叶面积 | PointCNN | > 85.00 |
5 | Li等[55] | 玉米 | 叶长,叶宽,叶倾角,叶长高,株高和茎高 | PointNet | > 82 |
6 | Gong等[56] | 水稻 | 茎的直径,茎的长度,穗的长、高、宽,主穗 | Panicle-3D | 93.4 |
7 | 杨琳等[57] | 油菜 | 叶片,角果 | 超体素算法 | 93.38 |
Table 2 Researches on three-dimensional reconstruction of crops in recent years表2 近年作物三维重建研究成果 |
编号 | 作者 | 采集设备 | 作物种类 | 重构方法 | 准确率R 2/% |
---|---|---|---|---|---|
1 | Ma等[66] | RGB相机 | 大豆 | 形状特征--相干点漂移(Intrinsic Shape Signatures-Coherent Point Drift, ISS-CPD)算法和迭代最接近点(Iterative Closest Point, ICP)算法 | 96.54 |
2 | Li等[67] | 相机 | 玉米 | 欧氏聚类算法、颜色滤波算法和点云体素滤波算法 | >92.60 |
3 | 魏天翔等[68] | 相机 | 水稻 | DeepLabv3+ | > 89.00 |
4 | Zhu 等[69] | 相机 | 玉米 | 三维辐射模型 | >97.00 |
5 | 史维杰[70] | 相机 | 谷子 | 运动恢复结构的稀疏重建、多视角立体几何 | > 94.00 |
6 | Ma等[71] | 三维激光扫描仪 | 玉米 | 网格法 | >89.07 |
7 | Sun等[72] | 相机 | 大豆 | U-net | 99.53 |
1 |
王晓鸣, 邱丽娟, 景蕊莲, 等. 作物种质资源表型性状鉴定评价:现状与趋势[J]. 植物遗传资源学报, 2022, 23(1): 12-20.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
2 |
潘朝阳, 陆展华, 刘维, 等. 表型组学研究进展及其在作物研究中的应用[J]. 广东农业科学, 2022, 49(9): 105-113.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
3 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
4 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
5 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
6 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
7 |
周济, TardieuFrancois, PridmoreTony, 等. 植物表型组学: 发展、现状与挑战[J]. 南京农业大学学报, 2018, 41(4): 580-588.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
8 |
温维亮, 郭新宇, 张颖, 等. 作物表型组大数据技术及装备发展研究[J]. 中国工程科学, 2023, 25(4): 227-238.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
9 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
10 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
11 |
左超, 冯世杰, 张翔宇, 等. 深度学习下的计算成像: 现状、挑战与未来[J]. 光学学报, 2020, 40(1): ID 0111003.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
12 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
13 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
14 |
梁永检, 吴文志, 施泽升, 等. 基于无人机RGB遥感的甘蔗株高估测[J]. 作物杂志, 2023(1): 226-232.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
15 |
郭涛, 颜安, 耿洪伟. 基于无人机影像的小麦株高与LAI预测研究[J]. 麦类作物学报, 2020, 40(9): 1129-1140.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
16 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
17 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
18 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
19 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
20 |
杨进, 明博, 杨飞, 等. 利用无人机影像监测不同生育阶段玉米群体株高的精度差异分析[J]. 智慧农业(中英文), 2021, 3(3): 129-138.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
21 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
22 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
23 |
聂志刚, 李广, 王钧, 等. 基于APSIM模型旱地小麦叶面积指数相关参数的优化[J]. 中国农业科学, 2019, 52(12): 2056-2068.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
24 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
25 |
张霞, 陶诗语, 张茂. 基于LSTM算法的FY-3B卫星春玉米叶面积指数反演[J]. 吉林大学学报(地球科学版), 2022, 52(6): 2071-2080.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
26 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
27 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
28 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
29 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
30 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
31 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
32 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
33 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
34 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
35 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
36 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
37 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
38 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
39 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
40 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
41 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
42 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
43 |
王文广, 王永红.作物株型与产量研究进展与展望[J].中国科学:生命科学,2021,51(10):1366-1375.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
44 |
米湘成, 邹应斌, 赵炳然, 等. 超级杂交水稻理想株型图像识别的探讨[J]. 湖南农业大学学报(自然科学版), 2001, 27(2): 151-153.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
45 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
46 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
47 |
贾奥博, 董天浩, 张彦, 等. 基于三维点云和集成学习的大田烟草株型特征识别[J]. 浙江大学学报(农业与生命科学版), 2022, 48(3): 393-402.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
48 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
49 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
50 |
许安宁. 基于深度学习的三维点云语义分割方法综述[J]. 长江信息通信, 2021, 34(1): 59-62.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
51 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
52 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
53 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
54 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
55 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
56 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
57 |
杨琳, 翟瑞芳, 阳旭, 等. 结合超体素和区域增长的植物器官点云分割[J]. 计算机工程与应用, 2019, 55(16): 197-203.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
58 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
59 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
60 |
周静静. 基于多视角图像的玉米植株三维重建技术研究[D]. 上海: 上海海洋大学, 2019.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
61 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
62 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
63 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
64 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
65 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
66 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
67 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
68 |
魏天翔, 汪小旵, 施印炎, 等. 基于三维重建的秸秆地单株水稻生长形态检测研究[J]. 南京农业大学学报, 2021, 44(6): 1197-1208.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
69 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
70 |
史维杰. 基于三维模型的谷子表型特征分析研究[D]. 太谷: 山西农业大学, 2020. SHI W J. Analysis and research of millet phenotype features based on 3D model[D].Taigu: Shanxi Agricultural University, 2020.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
71 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
72 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_ref.label}} |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
Collection(s)
/
〈 |
|
〉 |