High-throughput soybean seeds phenotyping with convolutional neural networks and transfer learning

Deep learning-based elaiosome detection in milk thistle seed for efficient high-throughput phenotyping

Milk thistle, Silybum marianum (L.), is a well-known medicinal plant used for the treatment of liver diseases due to its high content of silymarin. The seeds contain elaiosome, a fleshy structure attached to the seeds, which is believed to be a rich source of many metabolites including silymarin. Segmentation of elaiosomes using only image analysis is difficult, and this makes it impossible to quantify the elaiosome phenotypes. This study proposes a new approach for semi-automated detection and segmentation of elaiosomes in milk thistle seed using the Detectron2 deep learning algorithm. One hundred manually labeled images were used to train the initial elaiosome detection model. This model was used to predict elaiosome from new datasets, and the precise predictions were manually selected and used as new labeled images for retraining the model. Such semi-automatic image labeling, i.e., using the prediction results of the previous stage for retraining the model, allowed the production of sufficient labeled data for retraining. Finally, a total of 6,000 labeled images were used to train Detectron2 for elaiosome detection and attained a promising result. The results demonstrate the effectiveness of Detectron2 in detecting milk thistle seed elaiosomes with an accuracy of 99.9%. The proposed method automatically detects and segments elaiosome from the milk thistle seed. The predicted mask images of elaiosome were used to analyze its area as one of the seed phenotypic traits along with other seed morphological traits by image-based high-throughput phenotyping in ImageJ. Enabling high-throughput phenotyping of elaiosome and other seed morphological traits will be useful for breeding milk thistle cultivars with desirable traits.

Adherent Peanut Image Segmentation Based on Multi-Modal Fusion

Aiming at the problem of the difficult segmentation of adherent images due to the not fully convex shape of peanut pods, their complex surface texture, and their diverse structures, a multimodal fusion algorithm is proposed to achieve a 2D segmentation of adherent peanut images with the assistance of 3D point clouds. Firstly, the point cloud of a running peanut is captured line by line using a line structured light imaging system, and its three-dimensional shape is obtained through splicing and combining it with a local surface-fitting algorithm to calculate a normal vector and curvature. Seed points are selected based on the principle of minimum curvature, and neighboring points are searched using the KD-Tree algorithm. The point cloud is filtered and segmented according to the normal angle and the curvature threshold until achieving the completion of the point cloud segmentation of the individual peanut, and then the two-dimensional contour of the individual peanut model is extracted by using the rolling method. The search template is established, multiscale feature matching is implemented on the adherent image to achieve the region localization, and finally, the segmentation region is optimized by an opening operation. The experimental results show that the algorithm improves the segmentation accuracy, and the segmentation accuracy reaches 96.8%.

DEKR-SPrior: An Efficient Bottom-Up Keypoint Detection Model for Accurate Pod Phenotyping in Soybean

The pod and seed counts are important yield-related traits in soybean. High-precision soybean breeders face the major challenge of accurately phenotyping the number of pods and seeds in a high-throughput manner. Recent advances in artificial intelligence, especially deep learning (DL) models, have provided new avenues for high-throughput phenotyping of crop traits with increased precision. However, the available DL models are less effective for phenotyping pods that are densely packed and overlap in in situ soybean plants; thus, accurate phenotyping of the number of pods and seeds in soybean plant is an important challenge. To address this challenge, the present study proposed a bottom-up model, DEKR-SPrior (disentangled keypoint regression with structural prior), for in situ soybean pod phenotyping, which considers soybean pods and seeds analogous to human people and joints, respectively. In particular, we designed a novel structural prior (SPrior) module that utilizes cosine similarity to improve feature discrimination, which is important for differentiating closely located seeds from highly similar seeds. To further enhance the accuracy of pod location, we cropped full-sized images into smaller and high-resolution subimages for analysis. The results on our image datasets revealed that DEKR-SPrior outperformed multiple bottom-up models, viz., Lightweight-OpenPose, OpenPose, HigherHRNet, and DEKR, reducing the mean absolute error from 25.81 (in the original DEKR) to 21.11 (in the DEKR-SPrior) in pod phenotyping. This paper demonstrated the great potential of DEKR-SPrior for plant phenotyping, and we hope that DEKR-SPrior will help future plant phenotyping.

Image-based phenotyping of seed architectural traits and prediction of seed weight using machine learning models in soybean

Among seed attributes, weight is one of the main factors determining the soybean harvest index. Recently, the focus of soybean breeding has shifted to improving seed size and weight for crop optimization in terms of seed and oil yield. With recent technological advancements, there is an increasing application of imaging sensors that provide simple, real-time, non-destructive, and inexpensive image data for rapid image-based prediction of seed traits in plant breeding programs. The present work is related to digital image analysis of seed traits for the prediction of hundred-seed weight (HSW) in soybean. The image-based seed architectural traits (i-traits) measured were area size (AS), perimeter length (PL), length (L), width (W), length-to-width ratio (LWR), intersection of length and width (IS), seed circularity (CS), and distance between IS and CG (DS). The phenotypic investigation revealed significant genetic variability among 164 soybean genotypes for both i-traits and manually measured seed weight. Seven popular machine learning (ML) algorithms, namely Simple Linear Regression (SLR), Multiple Linear Regression (MLR), Random Forest (RF), Support Vector Regression (SVR), LASSO Regression (LR), Ridge Regression (RR), and Elastic Net Regression (EN), were used to create models that can predict the weight of soybean seeds based on the image-based novel features derived from the Red-Green-Blue (RGB)/visual image. Among the models, random forest and multiple linear regression models that use multiple explanatory variables related to seed size traits (AS, L, W, and DS) were identified as the best models for predicting seed weight with the highest prediction accuracy (coefficient of determination, R2=0.98 and 0.94, respectively) and the lowest prediction error, i.e., root mean square error (RMSE) and mean absolute error (MAE). Finally, principal components analysis (PCA) and a hierarchical clustering approach were used to identify IC538070 as a superior genotype with a larger seed size and weight. The identified donors/traits can potentially be used in soybean improvement programs.

Automatic Evaluation of Soybean Seed Traits Using RGB Image Data and a Python Algorithm

Soybean (Glycine max) is a crucial legume crop known for its nutritional value, as its seeds provide large amounts of plant protein and oil. To ensure maximum productivity in soybean farming, it is essential to carefully choose high-quality seeds that possess desirable characteristics, such as the appropriate size, shape, color, and absence of any damage. By studying the relationship between seed shape and other traits, we can effectively identify different genotypes and improve breeding strategies to develop high-yielding soybean seeds. This study focused on the analysis of seed traits using a Python algorithm. The seed length, width, projected area, and aspect ratio were measured, and the total number of seeds was calculated. The OpenCV library along with the contour detection function were used to measure the seed traits. The seed traits obtained through the algorithm were compared with the values obtained manually and from two software applications (SmartGrain and WinDIAS). The algorithm-derived measurements for the seed length, width, and projected area showed a strong correlation with the measurements obtained using various methods, with R-square values greater than 0.95 (p < 0.0001). Similarly, the error metrics, including the residual standard error, root mean square error, and mean absolute error, were all below 0.5% when comparing the seed length, width, and aspect ratio across different measurement methods. For the projected area, the error was less than 4% when compared with different measurement methods. Furthermore, the algorithm used to count the number of seeds present in the acquired images was highly accurate, and only a few errors were observed. This was a preliminary study that investigated only some morphological traits, and further research is needed to explore more seed attributes.

A novel transfer learning framework for sorghum biomass prediction using UAV-based remote sensing data and genetic markers

Yield for biofuel crops is measured in terms of biomass, so measurements throughout the growing season are crucial in breeding programs, yet traditionally time- and labor-consuming since they involve destructive sampling. Modern remote sensing platforms, such as unmanned aerial vehicles (UAVs), can carry multiple sensors and collect numerous phenotypic traits with efficient, non-invasive field surveys. However, modeling the complex relationships between the observed phenotypic traits and biomass remains a challenging task, as the ground reference data are very limited for each genotype in the breeding experiment. In this study, a Long Short-Term Memory (LSTM) based Recurrent Neural Network (RNN) model is proposed for sorghum biomass prediction. The architecture is designed to exploit the time series remote sensing and weather data, as well as static genotypic information. As a large number of features have been derived from the remote sensing data, feature importance analysis is conducted to identify and remove redundant features. A strategy to extract representative information from high-dimensional genetic markers is proposed. To enhance generalization and minimize the need for ground reference data, transfer learning strategies are proposed for selecting the most informative training samples from the target domain. Consequently, a pre-trained model can be refined with limited training samples. Field experiments were conducted over a sorghum breeding trial planted in multiple years with more than 600 testcross hybrids. The results show that the proposed LSTM-based RNN model can achieve high accuracies for single year prediction. Further, with the proposed transfer learning strategies, a pre-trained model can be refined with limited training samples from the target domain and predict biomass with an accuracy comparable to that from a trained-from-scratch model for both multiple experiments within a given year and across multiple years.

Artificial Intelligence in Food Safety: A Decade Review and Bibliometric Analysis

Artificial Intelligence (AI) technologies have been powerful solutions used to improve food yield, quality, and nutrition, increase safety and traceability while decreasing resource consumption, and eliminate food waste. Compared with several qualitative reviews on AI in food safety, we conducted an in-depth quantitative and systematic review based on the Core Collection database of WoS (Web of Science). To discover the historical trajectory and identify future trends, we analysed the literature concerning AI technologies in food safety from 2012 to 2022 by CiteSpace. In this review, we used bibliometric methods to describe the development of AI in food safety, including performance analysis, science mapping, and network analysis by CiteSpace. Among the 1855 selected articles, China and the United States contributed the most literature, and the Chinese Academy of Sciences released the largest number of relevant articles. Among all the journals in this field, PLoS ONE and Computers and Electronics in Agriculture ranked first and second in terms of annual publications and co-citation frequency. The present character, hot spots, and future research trends of AI technologies in food safety research were determined. Furthermore, based on our analyses, we provide researchers, practitioners, and policymakers with the big picture of research on AI in food safety across the whole process, from precision agriculture to precision nutrition, through 28 enlightening articles.

Machine learning bridges omics sciences and plant breeding

Some of the biological knowledge obtained from fundamental research will be implemented in applied plant breeding. To bridge basic research and breeding practice, machine learning (ML) holds great promise to translate biological knowledge and omics data into precision-designed plant breeding. Here, we review ML for multi-omics analysis in plants, including data dimensionality reduction, inference of gene-regulation networks, and gene discovery and prioritization. These applications will facilitate understanding trait regulation mechanisms and identifying target genes potentially applicable to knowledge-driven molecular design breeding. We also highlight applications of deep learning in plant phenomics and ML in genomic selection-assisted breeding, such as various ML algorithms that model the correlations among genotypes (genes), phenotypes (traits), and environments, to ultimately achieve data-driven genomic design breeding.

Study on Rice Grain Mildewed Region Recognition Based on Microscopic Computer Vision and YOLO-v5 Model

This study aims to develop a high-speed and nondestructive mildewed rice grain detection method. First, a set of microscopic images of rice grains contaminated by Aspergillus niger, Penicillium citrinum, and Aspergillus cinerea are acquired to serve as samples, and the mildewed regions are marked. Then, three YOLO-v5 models for identifying regions of rice grain with contamination of Aspergillus niger, Penicillium citrinum, and Aspergillus cinerea in microscopic images are established. Finally, the relationship between the proportion of mildewed regions and the total number of colonies is analyzed. The results show that the proposed YOLO-v5 models achieve accuracy levels of 89.26%, 91.15%, and 90.19% when detecting mildewed regions with contamination of Aspergillus niger, Penicillium citrinum, and Aspergillus cinerea in the microscopic images of the verification set. The proportion of the mildewed region area of rice grain with contamination of Aspergillus niger/Penicillium citrinum/Aspergillus cinerea is logarithmically correlated with the logarithm of the total number of colonies (TVC). The corresponding determination coefficients are 0.7466, 0.7587, and 0.8148, respectively. This study provides a reference for future research on high-speed mildewed rice grain detection methods based on MCV technology.

Semantic segmentation-based mechanized harvesting soybean quality detection

Crushing rate and impurity rate are important quality indicators of mechanically harvested soybeans. Intelligent quality detection of mechanically harvested soybeans based on machine vision is of great significance to evaluate soybean quality accurately and rapidly. This study proposes an improved U-Net method for identifying intact soybean grains, crushing soybean grains, and impurities. Based on the accurate identification of soybean components and using the quantitative model of soybean crushing rate and impurity rate, the quality of soybean samples can be detected in real-time. To this end, a soybean quality inspection system is designed to realize the dynamic collection and detection of soybean samples. The test results show that the comprehensive evaluation index values of the improved U-Net segmentation algorithm in identifying intact soybean grains, crushing soybean grains, and impurities are 93.04%, 89.40%, and 96.49%, respectively. Compared with the traditional U-Net model, the performance of the indicators is improved by 3.23%, 0.17% and 0.72%, respectively. Compared with manual detection, the maximum absolute error of the crushing rate detection of the soybean quality detection system is 0.57%, and the maximum absolute error of the impurity rate detection is 0.69%. The proposed soybean quality inspection system can be used as an effective tool for real-time online inspection of soybean quality.

Banana plant counting and morphological parameters measurement based on terrestrial laser scanning

BACKGROUND

The number of banana plants is closely related to banana yield. The diameter and height of the pseudo-stem are important morphological parameters of banana plants, which can reflect the growth status and vitality. To address the problems of high labor intensity and subjectivity in traditional measurement methods, a fast measurement method for banana plant count, pseudo-stem diameter, and height based on terrestrial laser scanning (TLS) was proposed.

RESULTS

First, during the nutritional growth period of banana, three-dimensional (3D) point cloud data of two measured fields were obtained by TLS. Second, the point cloud data was preprocessed. And the single plant segmentation of the canopy closed banana plant point cloud was realized furtherly. Finally, the number of banana plants was obtained by counting the number of pseudo-stems, and the diameter of pseudo-stems was measured using a cylindrical segmentation algorithm. A sliding window recognition method was proposed to determine the junction position between leaves and pseudo-stems, and the height of the pseudo-stems was measured. Compared with the measured value of artificial point cloud, when counting the number of banana plants, the precision,recall and percentage error of field 1 were 93.51%, 94.02%, and 0.54% respectively; the precision,recall and percentage error of field 2 were 96.34%, 92.00%, and 4.5% respectively; In the measurement of pseudo-stem diameter and height of banana, the root mean square error (RMSE) of pseudo-stem diameter and height of banana plant in field 1 were 0.38 cm and 0.2014 m respectively, and the mean absolute percentage error (MAPE) were 1.30% and 5.11% respectively; the RMSE of pseudo-stem diameter and height of banana plant in field 2 were 0.39 cm and 0.2788 m respectively, and the MAPE were 1.04% and 9.40% respectively.

CONCLUSION

The results show that the method proposed in this paper is suitable for the field measurement of banana count, pseudo-stem diameter, and height and can provide a fast field measurement method for banana plantation management.

Multi-Omics Techniques for Soybean Molecular Breeding

Soybean is a major crop that provides essential protein and oil for food and feed. Since its origin in China over 5000 years ago, soybean has spread throughout the world, becoming the second most important vegetable oil crop and the primary source of plant protein for global consumption. From early domestication and artificial selection through hybridization and ultimately molecular breeding, the history of soybean breeding parallels major advances in plant science throughout the centuries. Now, rapid progress in plant omics is ushering in a new era of precision design breeding, exemplified by the engineering of elite soybean varieties with specific oil compositions to meet various end-use targets. The assembly of soybean reference genomes, made possible by the development of genome sequencing technology and bioinformatics over the past 20 years, was a great step forward in soybean research. It facilitated advances in soybean transcriptomics, proteomics, metabolomics, and phenomics, all of which paved the way for an integrated approach to molecular breeding in soybean. In this review, we summarize the latest progress in omics research, highlight novel findings made possible by omics techniques, note current drawbacks and areas for further research, and suggest that an efficient multi-omics approach may accelerate soybean breeding in the future. This review will be of interest not only to soybean breeders but also to researchers interested in the use of cutting-edge omics technologies for crop research and improvement.

High-Throughput Legume Seed Phenotyping Using a Handheld 3D Laser Scanner

High-throughput phenotyping involves many samples and diverse trait types. For the goal of automatic measurement and batch data processing, a novel method for high-throughput legume seed phenotyping is proposed. A pipeline of automatic data acquisition and processing, including point cloud acquisition, single-seed extraction, pose normalization, three-dimensional (3D) reconstruction, and trait estimation, is proposed. First, a handheld laser scanner is used to obtain the legume seed point clouds in batches. Second, a combined segmentation method using the RANSAC method, the Euclidean segmentation method, and the dimensionality of the features is proposed to conduct single-seed extraction. Third, a coordinate rotation method based on PCA and the table normal is proposed to conduct pose normalization. Fourth, a fast symmetry-based 3D reconstruction method is built to reconstruct a 3D model of the single seed, and the Poisson surface reconstruction method is used for surface reconstruction. Finally, 34 traits, including 11 morphological traits, 11 scale factors, and 12 shape factors, are automatically calculated. A total of 2500 samples of five kinds of legume seeds are measured. Experimental results show that the average accuracies of scanning and segmentation are 99.52% and 100%, respectively. The overall average reconstruction error is 0.014 mm. The average morphological trait measurement accuracy is submillimeter, and the average relative percentage error is within 3%. The proposed method provides a feasible method of batch data acquisition and processing, which will facilitate the automation in high-throughput legume seed phenotyping.

A Deep-Learning-Based Approach for Wheat Yellow Rust Disease Recognition from Unmanned Aerial Vehicle Images

Yellow rust is a disease with a wide range that causes great damage to wheat. The traditional method of manually identifying wheat yellow rust is very inefficient. To improve this situation, this study proposed a deep-learning-based method for identifying wheat yellow rust from unmanned aerial vehicle (UAV) images. The method was based on the pyramid scene parsing network (PSPNet) semantic segmentation model to classify healthy wheat, yellow rust wheat, and bare soil in small-scale UAV images, and to investigate the spatial generalization of the model. In addition, it was proposed to use the high-accuracy classification results of traditional algorithms as weak samples for wheat yellow rust identification. The recognition accuracy of the PSPNet model in this study reached 98%. On this basis, this study used the trained semantic segmentation model to recognize another wheat field. The results showed that the method had certain generalization ability, and its accuracy reached 98%. In addition, the high-accuracy classification result of a support vector machine was used as a weak label by weak supervision, which better solved the labeling problem of large-size images, and the final recognition accuracy reached 94%. Therefore, the present study method facilitated timely control measures to reduce economic losses.

DeepCob: precise and high-throughput analysis of maize cob geometry using deep learning with an application in genebank phenomics

BACKGROUND

Maize cobs are an important component of crop yield that exhibit a high diversity in size, shape and color in native landraces and modern varieties. Various phenotyping approaches were developed to measure maize cob parameters in a high throughput fashion. More recently, deep learning methods like convolutional neural networks (CNNs) became available and were shown to be highly useful for high-throughput plant phenotyping. We aimed at comparing classical image segmentation with deep learning methods for maize cob image segmentation and phenotyping using a large image dataset of native maize landrace diversity from Peru.

RESULTS

Comparison of three image analysis methods showed that a Mask R-CNN trained on a diverse set of maize cob images was highly superior to classical image analysis using the Felzenszwalb-Huttenlocher algorithm and a Window-based CNN due to its robustness to image quality and object segmentation accuracy ([Formula: see text]). We integrated Mask R-CNN into a high-throughput pipeline to segment both maize cobs and rulers in images and perform an automated quantitative analysis of eight phenotypic traits, including diameter, length, ellipticity, asymmetry, aspect ratio and average values of red, green and blue color channels for cob color. Statistical analysis identified key training parameters for efficient iterative model updating. We also show that a small number of 10-20 images is sufficient to update the initial Mask R-CNN model to process new types of cob images. To demonstrate an application of the pipeline we analyzed phenotypic variation in 19,867 maize cobs extracted from 3449 images of 2484 accessions from the maize genebank of Peru to identify phenotypically homogeneous and heterogeneous genebank accessions using multivariate clustering.

CONCLUSIONS

Single Mask R-CNN model and associated analysis pipeline are widely applicable tools for maize cob phenotyping in contexts like genebank phenomics or plant breeding.

Estimation of nitrate nitrogen content in cotton petioles under drip irrigation based on wavelet neural network approach using spectral indices

BACKGROUND

Estimation of nitrate nitrogen (NO3--N) content in petioles is one of the key approaches for monitoring nitrogen (N) nutrition in crops. Rapid, non-destructive, and accurate evaluation of NO3--N contents in cotton petioles under drip irrigation is of great significance.

METHODS

In this study, we discussed the use of hyperspectral data to estimate NO3--N contents in cotton petioles under drip irrigation at different N treatments and growth stages. The correlations among trilateral parameters and six vegetation indices and petiole NO3--N contents were first investigated, after which a traditional regression model for petioles NO3--N content was established. A wavelet neural network (WNN) model for estimating petiole NO3--N content was also established. In addition, the performance of WNN was compared to those of random forest (RF), radial basis function neural network (RBF) and back propagation neural network (BP).

RESULTS

Between the blue edge amplitude (Db) and blue edge area (SDb) of the blue edge parameters was the optimal index for the estimation model of petiole NO3--N content. We found that the prediction results of the blue edge parameters and WNN were 7.3% higher than the coefficient of determination (R2) of the first derivative vegetation index and WNN. Root mean square error (RMSE) and mean absolute error (MAE) were 25.2% and 30.9% lower than first derivative vegetation, respectively, and the performance was better than that of RF, RBF and BP.

CONCLUSIONS

An inexpensive approach consisting of the WNN algorithm and blue edge parameters can be used to enhance the accuracy of NO3--N content estimation in cotton petioles under drip irrigation.