Talukder Zaki Jubery
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Talukder Zaki Jubery has a total of 88 co-authors. Here are the top co-authors by number of shared publications:
- Baskar Ganapathysubramanian
Department of Mechanical Engineering, Iowa State University, Ames, USA
14 shared publications - Soumik Sarkar
Department of Mechanical Engineering, Iowa State University, Ames, USA
9 shared publications - Arti Singh
Department of Agronomy, Iowa State University, Ames, USA
7 shared publications - Asheesh Kumar Singh
Department of Agronomy, Iowa State University, Ames, USA
7 shared publications Baskar Ganapathysubramanian
4 shared publications- Venkata Naresh Boddepalli
Iowa State University
4 shared publications
Latest Publications
- Open Access
- 19 August 2026
- Published by Wiley in The Plant Genome
- Vol. 19 (3) , e70294
Integrative genomics and zero‐shot deep learning–based phenotyping reveal the genetic architecture of seed traits in mungbean ( Vigna radiata L.)
Improving seed size and weight is a major breeding goal in mungbean ( Vigna radiata (L.) R. Wilczek). Improved genomic resources and precision phenotyping may enable more efficient selection for seed trait improvement. In this study, we integrated a deep learning–based segment anything model phenotyping pipeline with genome‐wide association studies (GWAS), comparative mapping, and genomic prediction (GP) to dissect the genetic architecture of seed size, shape, and weight traits in the Iowa mungbean diversity panel. The zero‐shot segmentation approach reliably captured seed size traits, which exhibited high heritability ( H 2 > 0.90) and strong correlation with seed weight ( r > 0.91). A multi‐model GWAS identified 82 unique single nucleotide polymorphisms (SNPs) across all seven traits, of which 13 were major pleiotropic SNPs governing multiple seed dimensions, including high‐confidence regions on chromosomes 1, 4, and 6 that explained over 20% of the phenotypic variance. Within these SNP regions, comparative mapping highlighted candidate genes including an ABC transporter ( Virad01G0084400 ) and two colocated candidates, NPGR1 ( Virad06G0255600 ) and a RING‐type E3 ubiquitin‐protein ligase ( Virad06G0255800 ), presented as hypothesis‐generating candidates for seed size regulation. GP using genomic best linear unbiased prediction (gBLUP) produced moderate to high accuracies for seed size and weight traits ( r = 0.76–0.84). Incorporating significant GWAS SNPs (gBLUP + SNPs) yielded slight improvements, suggesting the standard gBLUP model is sufficiently robust for selection. Collectively, this study provides the most comprehensive genomic dissection of seed size and weight traits in mungbean to date, providing candidate loci and genomic prediction models that can accelerate genetic improvement for seed yield and quality.
- 19 August 2026
- Published by Wiley in The Plant Genome
- Vol. 19 (3) , e70294
Open AccessIntegrative genomics and zero‐shot deep learning–based phenotyping reveal the genetic architecture of seed traits in mungbean ( Vigna radiata L.)
Improving seed size and weight is a major breeding goal in mungbean ( Vigna radiata (L.) R. Wilczek). Improved genomic resources and precision phenotyping may enable more efficient selection for seed trait improvement. In this study, we integrated a deep learning–based segment anything model phenotyping pipeline with genome‐wide association studies (GWAS), comparative mapping, and genomic prediction (GP) to dissect the genetic architecture of seed size, shape, and weight traits in the Iowa mungbean diversity panel. The zero‐shot segmentation approach reliably captured seed size traits, which exhibited high heritability ( H 2 > 0.90) and strong correlation with seed weight ( r > 0.91). A multi‐model GWAS identified 82 unique single nucleotide polymorphisms (SNPs) across all seven traits, of which 13 were major pleiotropic SNPs governing multiple seed dimensions, including high‐confidence regions on chromosomes 1, 4, and 6 that explained over 20% of the phenotypic variance. Within these SNP regions, comparative mapping highlighted candidate genes including an ABC transporter ( Virad01G0084400 ) and two colocated candidates, NPGR1 ( Virad06G0255600 ) and a RING‐type E3 ubiquitin‐protein ligase ( Virad06G0255800 ), presented as hypothesis‐generating candidates for seed size regulation. GP using genomic best linear unbiased prediction (gBLUP) produced moderate to high accuracies for seed size and weight traits ( r = 0.76–0.84). Incorporating significant GWAS SNPs (gBLUP + SNPs) yielded slight improvements, suggesting the standard gBLUP model is sufficiently robust for selection. Collectively, this study provides the most comprehensive genomic dissection of seed size and weight traits in mungbean to date, providing candidate loci and genomic prediction models that can accelerate genetic improvement for seed yield and quality.
- Open Access
- 15 June 2026
- Published by Elsevier in Smart Agricultural Technology
- Vol. 14
FloraForge: Procedural generation of editable and analysis-ready 3D plant geometric models using LLM-assisted template design
- 15 June 2026
- Published by Elsevier in Smart Agricultural Technology
- Vol. 14
Open AccessFloraForge: Procedural generation of editable and analysis-ready 3D plant geometric models using LLM-assisted template design
- Open Access
- 28 April 2026
- Published by Oxford University Press (OUP) in G3: Genes, Genomes, Genetics
- Vol. 16 (6)
Image-based high-throughput phenotyping enables genetic analyses of pod morphological traits in mungbean ( Vigna radiata (L.) R. Wilczek)
Mungbean (Vigna radiata (L.) R. Wilczek) is a vital source of digestible proteins and is well-suited for the plant-based protein industry. In this study, we analyzed pod morphological traits in the Iowa Mungbean Diversity (IMD) panel of 372 genotypes (2022–2023) using image-analysis-based phenotyping on 2,418 pod images. Pod morphological traits were extracted using deep learning image analysis, achieving excellent agreement with manual measurements (r > 0.96 for pod length (PL) and seed-per-pod (SPP)). Four complementary genome-wide association studies models identified 65 significant SNPs (−log10(P) ≥ 5.56) associated with pod curvature, length, width, and SPP traits. A significant SNP (5_35265704) on chromosome 4 was linked to pod dimensional traits, length, width, and curvature. A candidate gene, Virad04G0076900, located 15.6 kb from this SNP, is part of the GH3 gene family and has an Arabidopsis ortholog (AT4G27260) known for influencing organ elongation, pod, and seed development. Another SNP, 5_210437 on chromosome 6, has been found to be significantly associated with both PL and SPP. A candidate gene, Virad06G0002400 (36.5 kb from this SNP), encodes a potassium transporter and shares homology with the Arabidopsis gene HAK5 (AT4G13420), known to influence pod growth. Image-based measurements achieved genomic prediction accuracies ranging from 0.61 to 0.85 across various traits, demonstrating comparable accuracy to manual methods for linear traits and up to 22% improvement for complex shape traits. These results highlight the potential of deep learning-assisted phenomics integrated with genomic tools to accelerate selection for improved pod architecture in mungbean breeding programs across the Midwestern United States and globally.
- 28 April 2026
- Published by Oxford University Press (OUP) in G3: Genes, Genomes, Genetics
- Vol. 16 (6)
Open AccessImage-based high-throughput phenotyping enables genetic analyses of pod morphological traits in mungbean ( Vigna radiata (L.) R. Wilczek)
Mungbean (Vigna radiata (L.) R. Wilczek) is a vital source of digestible proteins and is well-suited for the plant-based protein industry. In this study, we analyzed pod morphological traits in the Iowa Mungbean Diversity (IMD) panel of 372 genotypes (2022–2023) using image-analysis-based phenotyping on 2,418 pod images. Pod morphological traits were extracted using deep learning image analysis, achieving excellent agreement with manual measurements (r > 0.96 for pod length (PL) and seed-per-pod (SPP)). Four complementary genome-wide association studies models identified 65 significant SNPs (−log10(P) ≥ 5.56) associated with pod curvature, length, width, and SPP traits. A significant SNP (5_35265704) on chromosome 4 was linked to pod dimensional traits, length, width, and curvature. A candidate gene, Virad04G0076900, located 15.6 kb from this SNP, is part of the GH3 gene family and has an Arabidopsis ortholog (AT4G27260) known for influencing organ elongation, pod, and seed development. Another SNP, 5_210437 on chromosome 6, has been found to be significantly associated with both PL and SPP. A candidate gene, Virad06G0002400 (36.5 kb from this SNP), encodes a potassium transporter and shares homology with the Arabidopsis gene HAK5 (AT4G13420), known to influence pod growth. Image-based measurements achieved genomic prediction accuracies ranging from 0.61 to 0.85 across various traits, demonstrating comparable accuracy to manual methods for linear traits and up to 22% improvement for complex shape traits. These results highlight the potential of deep learning-assisted phenomics integrated with genomic tools to accelerate selection for improved pod architecture in mungbean breeding programs across the Midwestern United States and globally.
- Open Access
- 21 March 2026
- Published by Elsevier in Smart Agricultural Technology
- Vol. 14
Neural 3D reconstruction and immersive VR visualization of row crops across phenological growth stages
- 21 March 2026
- Published by Elsevier in Smart Agricultural Technology
- Vol. 14
Open AccessNeural 3D reconstruction and immersive VR visualization of row crops across phenological growth stages
- Open Access
HS-3D-NeRF: 3D Surface and Hyperspectral Reconstruction From Stationary Hyperspectral Images Using Multi-Channel NeRFs
Advances in hyperspectral imaging (HSI) and 3D reconstruction have enabled accurate, high-throughput characterization of agricultural produce quality and plant phenotypes, both essential for advancing agricultural sustainability and breeding programs. HSI captures detailed biochemical features of produce, while 3D geometric data substantially improves morphological analysis. However, integrating these two modalities at scale remains challenging, as conventional approaches involve complex hardware setups incompatible with automated phenotyping systems. Recent advances in neural radiance fields (NeRF) offer computationally efficient 3D reconstruction but typically require moving-camera setups, limiting throughput and reproducibility in standard indoor agricultural environments. To address these challenges, we introduce HSI-SC-NeRF, a stationary-camera multi-channel NeRF framework for high-throughput hyperspectral 3D reconstruction targeting postharvest inspection of agricultural produce. Multi-view hyperspectral data is captured using a stationary camera while the object rotates within a custom-built Teflon imaging chamber providing diffuse, uniform illumination. Object poses are estimated via ArUco calibration markers and transformed to the camera frame of reference through simulated pose transformations, enabling standard NeRF training on stationary-camera data. A multi-channel NeRF formulation optimizes reconstruction across all hyperspectral bands jointly using a composite spectral loss, supported by a two-stage training protocol that decouples geometric initialization from radiometric refinement. Experiments on three agricultural produce samples demonstrate high spatial reconstruction accuracy and strong spectral fidelity across the visible and near-infrared spectrum, confirming the suitability of HSI-SC-NeRF for integration into automated agricultural workflows.
Open AccessHS-3D-NeRF: 3D Surface and Hyperspectral Reconstruction From Stationary Hyperspectral Images Using Multi-Channel NeRFs
Advances in hyperspectral imaging (HSI) and 3D reconstruction have enabled accurate, high-throughput characterization of agricultural produce quality and plant phenotypes, both essential for advancing agricultural sustainability and breeding programs. HSI captures detailed biochemical features of produce, while 3D geometric data substantially improves morphological analysis. However, integrating these two modalities at scale remains challenging, as conventional approaches involve complex hardware setups incompatible with automated phenotyping systems. Recent advances in neural radiance fields (NeRF) offer computationally efficient 3D reconstruction but typically require moving-camera setups, limiting throughput and reproducibility in standard indoor agricultural environments. To address these challenges, we introduce HSI-SC-NeRF, a stationary-camera multi-channel NeRF framework for high-throughput hyperspectral 3D reconstruction targeting postharvest inspection of agricultural produce. Multi-view hyperspectral data is captured using a stationary camera while the object rotates within a custom-built Teflon imaging chamber providing diffuse, uniform illumination. Object poses are estimated via ArUco calibration markers and transformed to the camera frame of reference through simulated pose transformations, enabling standard NeRF training on stationary-camera data. A multi-channel NeRF formulation optimizes reconstruction across all hyperspectral bands jointly using a composite spectral loss, supported by a two-stage training protocol that decouples geometric initialization from radiometric refinement. Experiments on three agricultural produce samples demonstrate high spatial reconstruction accuracy and strong spectral fidelity across the visible and near-infrared spectrum, confirming the suitability of HSI-SC-NeRF for integration into automated agricultural workflows.
- Open Access
AI-assisted Image-Based Phenotyping Reveals Genetic Architecture of Pod Traits in Mungbean ( Vigna radiata L.)
Mungbean ( Vigna radiata (L.) R. Wilczek) is a vital source of digestible proteins and is well-suited for the plant-based protein industry. In this study, we analyzed pod morphological traits in the Iowa Mungbean Diversity (IMD) panel with 372 genotypes (2022-23) with AI-assisted image phenotyping using 2,418 pod images. Pod morphological traits were extracted using deep learning image analysis, achieving excellent agreement with manual measurements (r>0.96 for pod length and seed per pod). Four complementary GWAS models identified 45 significant SNPs associated with pod curvature, length, width, and seed per pod traits. Notably, a significant SNP (5_35265704) on chromosome 1 was linked to pod dimensional traits, length, width, and curvature. A candidate gene, Vradi01g00001116 , was located within the linkage disequilibrium (LD) region of this SNP, is part of the GH3 gene family, and has an Arabidopsis ortholog ( AT4G27260 ) known for influencing organ elongation, pod, and seed development. Another SNP, 5_210437 on chromosome 2, has been found to be significantly associated with both pod length and seed per pod. A candidate gene, Vradi02g00003971 , located in the LD region of this SNP, belongs to the potassium transporter family and shares homology with the HAK5 gene family ( AT4G13420 ) in Arabidopsis , which influences pod and seed growth. Image-based measurements achieved genomic prediction accuracies ranging from 0.61 to 0.85 across various traits, exhibiting an improvement of 12-22% over manual methods. These results demonstrate the potential of AI-assisted phenomics integrated with genomic tools to accelerate selection for improved pod architecture in mungbean breeding programs across the Midwestern United States and globally.
Open AccessAI-assisted Image-Based Phenotyping Reveals Genetic Architecture of Pod Traits in Mungbean ( Vigna radiata L.)
Mungbean ( Vigna radiata (L.) R. Wilczek) is a vital source of digestible proteins and is well-suited for the plant-based protein industry. In this study, we analyzed pod morphological traits in the Iowa Mungbean Diversity (IMD) panel with 372 genotypes (2022-23) with AI-assisted image phenotyping using 2,418 pod images. Pod morphological traits were extracted using deep learning image analysis, achieving excellent agreement with manual measurements (r>0.96 for pod length and seed per pod). Four complementary GWAS models identified 45 significant SNPs associated with pod curvature, length, width, and seed per pod traits. Notably, a significant SNP (5_35265704) on chromosome 1 was linked to pod dimensional traits, length, width, and curvature. A candidate gene, Vradi01g00001116 , was located within the linkage disequilibrium (LD) region of this SNP, is part of the GH3 gene family, and has an Arabidopsis ortholog ( AT4G27260 ) known for influencing organ elongation, pod, and seed development. Another SNP, 5_210437 on chromosome 2, has been found to be significantly associated with both pod length and seed per pod. A candidate gene, Vradi02g00003971 , located in the LD region of this SNP, belongs to the potassium transporter family and shares homology with the HAK5 gene family ( AT4G13420 ) in Arabidopsis , which influences pod and seed growth. Image-based measurements achieved genomic prediction accuracies ranging from 0.61 to 0.85 across various traits, exhibiting an improvement of 12-22% over manual methods. These results demonstrate the potential of AI-assisted phenomics integrated with genomic tools to accelerate selection for improved pod architecture in mungbean breeding programs across the Midwestern United States and globally.
- Open Access
Sewing the future of cotton: a multi-omics study combining nanomechanics, transcriptomics, and phenotypic traits
Cellulose microfibrils that are essential for mechanical strength and overall quality of cotton fibers. This study quantifies and compares the nanoscale structural and mechanical properties of cellulose microfibrils such as microfibril dimensions, crossover count and angles, roughness, and Young’s modulus for two popular cotton species: Gossypium hirsutum (Gh) and Gossypium barbadense (Gb) fibers across four growth stages (8, 12, 18, and 22 days post-anthesis) using atomic force microscopy (AFM). Our results revealed for the first time that Gb fibers exhibit a better alignment, finer dimensions, and higher stiffness compared to Gh fibers at nanoscale, resulting in smoother fiber surfaces, and improved quality at macroscale. We are also the first to develop machine-learning models to predict macroscale phenotypic traits specifically boll length and cellulose content using nanoscale features alone and in combination with multi-omics modalities, substantially enhancing the predictive accuracy and highlighting opportunities for robust cross-species modeling of cotton fiber traits.
Open AccessSewing the future of cotton: a multi-omics study combining nanomechanics, transcriptomics, and phenotypic traits
Cellulose microfibrils that are essential for mechanical strength and overall quality of cotton fibers. This study quantifies and compares the nanoscale structural and mechanical properties of cellulose microfibrils such as microfibril dimensions, crossover count and angles, roughness, and Young’s modulus for two popular cotton species: Gossypium hirsutum (Gh) and Gossypium barbadense (Gb) fibers across four growth stages (8, 12, 18, and 22 days post-anthesis) using atomic force microscopy (AFM). Our results revealed for the first time that Gb fibers exhibit a better alignment, finer dimensions, and higher stiffness compared to Gh fibers at nanoscale, resulting in smoother fiber surfaces, and improved quality at macroscale. We are also the first to develop machine-learning models to predict macroscale phenotypic traits specifically boll length and cellulose content using nanoscale features alone and in combination with multi-omics modalities, substantially enhancing the predictive accuracy and highlighting opportunities for robust cross-species modeling of cotton fiber traits.
- Open Access
- 7 January 2025
- Published by Wiley in Plants, People, Planet
- Vol. 7 (4) , 1011-1026
Plot‐level satellite imagery can substitute for UAVs in assessing maize phenotypes across multistate field trials
Societal Impact Statement: Maize plays a key role in agricultural profitability and food security on six continents. Successful efforts to breed higher‐yielding maize varieties depend on replicated yield trials in many environments. Capturing in‐season data can help improve and accelerate the development of regionally adapted hybrids, but collecting these data can be impractical in many locations. We demonstrate that satellite remote sensing could play a similar role in crop performance assessment as unmanned aerial vehicles (UAVs) but with far lower labor costs and the greatest ease of collection at remote field sites. This dataset and benchmarks have the potential to enable predictive models that could guide farmers and crop breeders in decision‐making.Summary: Accurate early yield estimates at fields and plots offer potential benefits to farmers in optimizing their agronomic practices and breeders in screening thousands of varieties contributing to improving agriculture and food production systems. Effective approaches to track plant growth and predict yield require large datasets of remote sensing and ground truth data collected across multiple environments. Low‐altitude drone flights are increasingly being used to collect data from field evaluations of new crop varieties, while satellite imagery is being explored to track yield and management practices at regional scales. Satellite platforms exhibit logistical and technical advantages in scalability and accessibility and could facilitate plot‐level predictions, especially with steadily improving spatial resolution. However, plot‐level, high‐resolution satellite images capturing differences in genotypes from multiple environments with ground truth measurements are not publicly available. Here, we generated, described, and evaluated over 20,000 plot‐level images of over 80 hybrid maize varieties grown across the US corn belt under various management practices collected from (near simultaneous) satellite and drone (synonym UAVs, UASs) flights integrated with ground truth yield measurement. Of the six baseline models examined, models employing data collected from satellite images often matched the performance of models employing drone images for both within and cross‐environment yield prediction.
- 7 January 2025
- Published by Wiley in Plants, People, Planet
- Vol. 7 (4) , 1011-1026
Open AccessPlot‐level satellite imagery can substitute for UAVs in assessing maize phenotypes across multistate field trials
Societal Impact Statement: Maize plays a key role in agricultural profitability and food security on six continents. Successful efforts to breed higher‐yielding maize varieties depend on replicated yield trials in many environments. Capturing in‐season data can help improve and accelerate the development of regionally adapted hybrids, but collecting these data can be impractical in many locations. We demonstrate that satellite remote sensing could play a similar role in crop performance assessment as unmanned aerial vehicles (UAVs) but with far lower labor costs and the greatest ease of collection at remote field sites. This dataset and benchmarks have the potential to enable predictive models that could guide farmers and crop breeders in decision‐making.Summary: Accurate early yield estimates at fields and plots offer potential benefits to farmers in optimizing their agronomic practices and breeders in screening thousands of varieties contributing to improving agriculture and food production systems. Effective approaches to track plant growth and predict yield require large datasets of remote sensing and ground truth data collected across multiple environments. Low‐altitude drone flights are increasingly being used to collect data from field evaluations of new crop varieties, while satellite imagery is being explored to track yield and management practices at regional scales. Satellite platforms exhibit logistical and technical advantages in scalability and accessibility and could facilitate plot‐level predictions, especially with steadily improving spatial resolution. However, plot‐level, high‐resolution satellite images capturing differences in genotypes from multiple environments with ground truth measurements are not publicly available. Here, we generated, described, and evaluated over 20,000 plot‐level images of over 80 hybrid maize varieties grown across the US corn belt under various management practices collected from (near simultaneous) satellite and drone (synonym UAVs, UASs) flights integrated with ground truth yield measurement. Of the six baseline models examined, models employing data collected from satellite images often matched the performance of models employing drone images for both within and cross‐environment yield prediction.
- Open Access
- 23 December 2024
- Published by Oxford University Press (OUP) in PNAS Nexus
- Vol. 4 (1) , pgae575
InsectNet: Real-time identification of insects using an end-to-end machine learning pipeline
Insect pests significantly impact global agricultural productivity and crop quality. Effective integrated pest management strategies require the identification of insects, including beneficial and harmful insects. Automated identification of insects under real-world conditions presents several challenges, including the need to handle intraspecies dissimilarity and interspecies similarity, life-cycle stages, camouflage, diverse imaging conditions, and variability in insect orientation. An end-to-end approach for training deep-learning models, InsectNet, is proposed to address these challenges. Our approach has the following key features: (i) uses a large dataset of insect images collected through citizen science along with label-free self-supervised learning to train a global model, (ii) fine-tuning this global model using smaller, expert-verified regional datasets to create a local insect identification model, (iii) which provides high prediction accuracy even for species with small sample sizes, (iv) is designed to enhance model trustworthiness, and (v) democratizes access through streamlined machine learning operations. This global-to-local model strategy offers a more scalable and economically viable solution for implementing advanced insect identification systems across diverse agricultural ecosystems. We report accurate identification (>96% accuracy) of numerous agriculturally and ecologically relevant insect species, including pollinators, parasitoids, predators, and harmful insects. InsectNet provides fine-grained insect species identification, works effectively in challenging backgrounds, and avoids making predictions when uncertain, increasing its utility and trustworthiness. The model and associated workflows are available through a web-based portal accessible through a computer or mobile device. We envision InsectNet to complement existing approaches, and be part of a growing suite of AI technologies for addressing agricultural challenges.
- 23 December 2024
- Published by Oxford University Press (OUP) in PNAS Nexus
- Vol. 4 (1) , pgae575
Open AccessInsectNet: Real-time identification of insects using an end-to-end machine learning pipeline
Insect pests significantly impact global agricultural productivity and crop quality. Effective integrated pest management strategies require the identification of insects, including beneficial and harmful insects. Automated identification of insects under real-world conditions presents several challenges, including the need to handle intraspecies dissimilarity and interspecies similarity, life-cycle stages, camouflage, diverse imaging conditions, and variability in insect orientation. An end-to-end approach for training deep-learning models, InsectNet, is proposed to address these challenges. Our approach has the following key features: (i) uses a large dataset of insect images collected through citizen science along with label-free self-supervised learning to train a global model, (ii) fine-tuning this global model using smaller, expert-verified regional datasets to create a local insect identification model, (iii) which provides high prediction accuracy even for species with small sample sizes, (iv) is designed to enhance model trustworthiness, and (v) democratizes access through streamlined machine learning operations. This global-to-local model strategy offers a more scalable and economically viable solution for implementing advanced insect identification systems across diverse agricultural ecosystems. We report accurate identification (>96% accuracy) of numerous agriculturally and ecologically relevant insect species, including pollinators, parasitoids, predators, and harmful insects. InsectNet provides fine-grained insect species identification, works effectively in challenging backgrounds, and avoids making predictions when uncertain, increasing its utility and trustworthiness. The model and associated workflows are available through a web-based portal accessible through a computer or mobile device. We envision InsectNet to complement existing approaches, and be part of a growing suite of AI technologies for addressing agricultural challenges.
- Open Access
Soybean Canopy Stress Classification Using 3D Point Cloud Data
Automated canopy stress classification for field crops has traditionally relied on single-perspective, two-dimensional (2D) photographs, usually obtained through top-view imaging using unmanned aerial vehicles (UAVs). However, this approach may fail to capture the full extent of plant stress symptoms, which can manifest throughout the canopy. Recent advancements in LiDAR technologies have enabled the acquisition of high-resolution 3D point cloud data for the entire canopy, offering new possibilities for more accurate plant stress identification and rating. This study explores the potential of leveraging 3D point cloud data for improved plant stress assessment. We utilized a dataset of RGB 3D point clouds of 700 soybean plants from a diversity panel exposed to iron deficiency chlorosis (IDC) stress. From this unique set of 700 canopies exhibiting varying levels of IDC, we extracted several representations, including (a) handcrafted IDC symptom-specific features, (b) canopy fingerprints, and (c) latent feature-based features. Subsequently, we trained several classification models to predict plant stress severity using these representations. We exhaustively investigated several stress representations and model combinations for the 3-D data. We also compared the performance of these classification models against similar models that are only trained using the associated top-view 2D RGB image for each plant. Among the feature-model combinations tested, the 3D canopy fingerprint features trained with a support vector machine yielded the best performance, achieving higher classification accuracy than the best-performing model based on 2D data built using convolutional neural networks. Our findings demonstrate the utility of color canopy fingerprinting and underscore the importance of considering 3D data to assess plant stress in agricultural applications.
Open AccessSoybean Canopy Stress Classification Using 3D Point Cloud Data
Automated canopy stress classification for field crops has traditionally relied on single-perspective, two-dimensional (2D) photographs, usually obtained through top-view imaging using unmanned aerial vehicles (UAVs). However, this approach may fail to capture the full extent of plant stress symptoms, which can manifest throughout the canopy. Recent advancements in LiDAR technologies have enabled the acquisition of high-resolution 3D point cloud data for the entire canopy, offering new possibilities for more accurate plant stress identification and rating. This study explores the potential of leveraging 3D point cloud data for improved plant stress assessment. We utilized a dataset of RGB 3D point clouds of 700 soybean plants from a diversity panel exposed to iron deficiency chlorosis (IDC) stress. From this unique set of 700 canopies exhibiting varying levels of IDC, we extracted several representations, including (a) handcrafted IDC symptom-specific features, (b) canopy fingerprints, and (c) latent feature-based features. Subsequently, we trained several classification models to predict plant stress severity using these representations. We exhaustively investigated several stress representations and model combinations for the 3-D data. We also compared the performance of these classification models against similar models that are only trained using the associated top-view 2D RGB image for each plant. Among the feature-model combinations tested, the 3D canopy fingerprint features trained with a support vector machine yielded the best performance, achieving higher classification accuracy than the best-performing model based on 2D data built using convolutional neural networks. Our findings demonstrate the utility of color canopy fingerprinting and underscore the importance of considering 3D data to assess plant stress in agricultural applications.