scholarly journals High-Throughput Phenotyping of Soybean Maturity Using Time Series UAV Imagery and Convolutional Neural Networks

2020 ◽  
Author(s):  
Rodrigo Trevisan ◽  
Osvaldo Pérez ◽  
Nathan Schmitz ◽  
Brian Diers ◽  
Nicolas Martin
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
High Throughput ◽  
Mean Squared Error ◽  
Ground Truth ◽  
Breeding Programs ◽  
Growing Seasons ◽  
Maturity Date ◽  
High Throughput Phenotyping ◽  
Commercial Breeding

Soybean maturity is a trait of critical importance for the development of new soybean cultivars, nevertheless, its characterization based on visual ratings has many challenges. Unmanned aerial vehicles (UAVs) imagery-based high-throughput phenotyping methodologies have been proposed as an alternative to the traditional visual ratings of pod senescence. However, the lack of scalable and accurate methods to extract the desired information from the images remains a significant bottleneck in breeding programs. The objective of this study was to develop an image-based high-throughput phenotyping system for evaluating soybean maturity in breeding programs. Images were acquired twice a week, starting when the earlier lines began maturation until the latest ones were mature. Two complementary convolutional neural networks (CNN) were developed to predict the maturity date. The first using a single date and the second using the five best image dates identified by the first model. The proposed CNN architecture was validated using more than 15,000 ground truth observations from five trials, including data from three growing seasons and two countries. The trained model showed good generalization capability with a root mean squared error lower than two days in four out of five trials. Four methods of estimating prediction uncertainty showed potential at identifying different sources of errors in the maturity date predictions. The architecture used solves limitations of previous research and can be used at scale in commercial breeding programs.

scholarly journals High-Throughput Phenotyping of Soybean Maturity Using Time Series UAV Imagery and Convolutional Neural Networks

2020 ◽  
Vol 12 (21) ◽  
pp. 3617
Author(s):  
Rodrigo Trevisan ◽  
Osvaldo Pérez ◽  
Nathan Schmitz ◽  
Brian Diers ◽  
Nicolas Martin
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
High Throughput ◽  
Mean Squared Error ◽  
Ground Truth ◽  
Breeding Programs ◽  
Growing Seasons ◽  
Maturity Date ◽  
High Throughput Phenotyping ◽  
Commercial Breeding

Soybean maturity is a trait of critical importance for the development of new soybean cultivars, nevertheless, its characterization based on visual ratings has many challenges. Unmanned aerial vehicles (UAVs) imagery-based high-throughput phenotyping methodologies have been proposed as an alternative to the traditional visual ratings of pod senescence. However, the lack of scalable and accurate methods to extract the desired information from the images remains a significant bottleneck in breeding programs. The objective of this study was to develop an image-based high-throughput phenotyping system for evaluating soybean maturity in breeding programs. Images were acquired twice a week, starting when the earlier lines began maturation until the latest ones were mature. Two complementary convolutional neural networks (CNN) were developed to predict the maturity date. The first using a single date and the second using the five best image dates identified by the first model. The proposed CNN architecture was validated using more than 15,000 ground truth observations from five trials, including data from three growing seasons and two countries. The trained model showed good generalization capability with a root mean squared error lower than two days in four out of five trials. Four methods of estimating prediction uncertainty showed potential at identifying different sources of errors in the maturity date predictions. The architecture developed solves limitations of previous research and can be used at scale in commercial breeding programs.

scholarly journals High-throughput phenotyping of two plant-size traits of Eucalyptus species using neural networks

2021 ◽  
Author(s):  
Marcus Vinicius Vieira Borges ◽  
Janielle de Oliveira Garcia ◽  
Tays Silva Batista ◽  
Alexsandra Nogueira Martins Silva ◽  
Fabio Henrique Rojo Baio ◽  
...  
Keyword(s):  
Neural Networks ◽  
High Throughput ◽  
Mean Squared Error ◽  
Vegetation Indices ◽  
Plant Size ◽  
Eucalyptus Species ◽  
Forest Inventories ◽  
Spectral Bands ◽  
High Throughput Phenotyping ◽  
Input Variables

AbstractIn forest modeling to estimate the volume of wood, artificial intelligence has been shown to be quite efficient, especially using artificial neural networks (ANNs). Here we tested whether diameter at breast height (DBH) and the total plant height (Ht) of eucalyptus can be predicted at the stand level using spectral bands measured by an unmanned aerial vehicle (UAV) multispectral sensor and vegetation indices. To do so, using the data obtained by the UAV as input variables, we tested different configurations (number of hidden layers and number of neurons in each layer) of ANNs for predicting DBH and Ht at stand level for different Eucalyptus species. The experimental design was randomized blocks with four replicates, with 20 trees in each experimental plot. The treatments comprised five Eucalyptus species (E. camaldulensis, E. uroplylla, E. saligna, E. grandis, and E. urograndis) and Corymbria citriodora. DBH and Ht for each plot at the stand level were measured seven times in separate overflights by the UAV, so that the multispectral sensor could obtain spectral bands to calculate vegetation indices (VIs). ANNs were then constructed using spectral bands and VIs as input layers, in addition to the categorical variable (species), to predict DBH and Ht at the stand level simultaneously. This report represents one of the first applications of high-throughput phenotyping for plant size traits in Eucalyptus species. In general, ANNs containing three hidden layers gave better statistical performance (higher estimated r, lower estimated root mean squared error–RMSE) due to their greater capacity for self-learning. Among these ANNs, the best contained eight neurons in the first layer, seven in the second, and five in the third (8 − 7 − 5). The results reported here reveal the potential of using the generated models to perform accurate forest inventories based on spectral bands and VIs obtained with a UAV multispectral sensor and ANNs, reducing labor and time.

scholarly journals Validation of UAV-based alfalfa biomass predictability using photogrammetry with fully automatic plot segmentation

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Zhou Tang ◽  
Atit Parajuli ◽  
Chunpeng James Chen ◽  
Yang Hu ◽  
Samuel Revolinski ◽  
...  
Keyword(s):  
Prediction Model ◽  
High Throughput ◽  
Vegetation Index ◽  
Normalized Difference Vegetation Index ◽  
Selection Process ◽  
Forage Legume ◽  
Breeding Programs ◽  
High Throughput Phenotyping ◽  
Fully Automatic ◽  
Uav Images

AbstractAlfalfa is the most widely cultivated forage legume, with approximately 30 million hectares planted worldwide. Genetic improvements in alfalfa have been highly successful in developing cultivars with exceptional winter hardiness and disease resistance traits. However, genetic improvements have been limited for complex economically important traits such as biomass. One of the major bottlenecks is the labor-intensive phenotyping burden for biomass selection. In this study, we employed two alfalfa fields to pave a path to overcome the challenge by using UAV images with fully automatic field plot segmentation for high-throughput phenotyping. The first field was used to develop the prediction model and the second field to validate the predictions. The first and second fields had 808 and 1025 plots, respectively. The first field had three harvests with biomass measured in May, July, and September of 2019. The second had one harvest with biomass measured in September of 2019. These two fields were imaged one day before harvesting with a DJI Phantom 4 pro UAV carrying an additional Sentera multispectral camera. Alfalfa plot images were extracted by GRID software to quantify vegetative area based on the Normalized Difference Vegetation Index. The prediction model developed from the first field explained 50–70% (R Square) of biomass variation in the second field by incorporating four features from UAV images: vegetative area, plant height, Normalized Green–Red Difference Index, and Normalized Difference Red Edge Index. This result suggests that UAV-based, high-throughput phenotyping could be used to improve the efficiency of the biomass selection process in alfalfa breeding programs.

scholarly journals Review: Cost-Effective Unmanned Aerial Vehicle (UAV) Platform for Field Plant Breeding Application

2020 ◽  
Vol 12 (6) ◽  
pp. 998 ◽  
Author(s):  
GyuJin Jang ◽  
Jaeyoung Kim ◽  
Ju-Kyung Yu ◽  
Hak-Jin Kim ◽  
Yoonha Kim ◽  
...  
Keyword(s):  
Plant Breeding ◽  
Unmanned Aerial Vehicle ◽  
High Throughput ◽  
New Technologies ◽  
Cost Effective ◽  
New Wave ◽  
Breeding Programs ◽  
Modern Agriculture ◽  
Aerial Vehicle ◽  
High Throughput Phenotyping

Utilization of remote sensing is a new wave of modern agriculture that accelerates plant breeding and research, and the performance of farming practices and farm management. High-throughput phenotyping is a key advanced agricultural technology and has been rapidly adopted in plant research. However, technology adoption is not easy due to cost limitations in academia. This article reviews various commercial unmanned aerial vehicle (UAV) platforms as a high-throughput phenotyping technology for plant breeding. It compares known commercial UAV platforms that are cost-effective and manageable in field settings and demonstrates a general workflow for high-throughput phenotyping, including data analysis. The authors expect this article to create opportunities for academics to access new technologies and utilize the information for their research and breeding programs in more workable ways.

scholarly journals Tropical Cyclone Intensity Estimation Using Multi-Dimensional Convolutional Neural Networks from Geostationary Satellite Data

2019 ◽  
Vol 12 (1) ◽  
pp. 108 ◽  
Author(s):  
Juhyun Lee ◽  
Jungho Im ◽  
Dong-Hyun Cha ◽  
Haemi Park ◽  
Seongmun Sim
Keyword(s):  
Neural Networks ◽  
Tropical Cyclone ◽  
Convolutional Neural Networks ◽  
Mean Squared Error ◽  
Single Channel ◽  
Three Dimensional ◽  
Geostationary Satellite ◽  
Lower Atmosphere ◽  
Long Time ◽  
3D Cnn

For a long time, researchers have tried to find a way to analyze tropical cyclone (TC) intensity in real-time. Since there is no standardized method for estimating TC intensity and the most widely used method is a manual algorithm using satellite-based cloud images, there is a bias that varies depending on the TC center and shape. In this study, we adopted convolutional neural networks (CNNs) which are part of a state-of-art approach that analyzes image patterns to estimate TC intensity by mimicking human cloud pattern recognition. Both two dimensional-CNN (2D-CNN) and three-dimensional-CNN (3D-CNN) were used to analyze the relationship between multi-spectral geostationary satellite images and TC intensity. Our best-optimized model produced a root mean squared error (RMSE) of 8.32 kts, resulting in better performance (~35%) than the existing model using the CNN-based approach with a single channel image. Moreover, we analyzed the characteristics of multi-spectral satellite-based TC images according to intensity using a heat map, which is one of the visualization means of CNNs. It shows that the stronger the intensity of the TC, the greater the influence of the TC center in the lower atmosphere. This is consistent with the results from the existing TC initialization method with numerical simulations based on dynamical TC models. Our study suggests the possibility that a deep learning approach can be used to interpret the behavior characteristics of TCs.

scholarly journals Comparison of Classical Computer Vision vs. Convolutional Neural Networks for Weed Mapping in Aerial Images

2020 ◽  
Vol 27 (4) ◽  
pp. 20-33
Author(s):  
Paulo César Pereira Júnior ◽  
Alexandre Monteiro ◽  
Rafael Da Luz Ribeiro ◽  
Antonio Carlos Sobieranski ◽  
Aldo Von Wangenheim
Keyword(s):  
Neural Networks ◽  
Computer Vision ◽  
Convolutional Neural Networks ◽  
Precision Agriculture ◽  
Ground Truth ◽  
Aerial Images ◽  
Weed Mapping ◽  
Classical Models ◽  
Classical Computer ◽  
Better Than

In this paper, we present a comparison between convolutional neural networks and classicalcomputer vision approaches, for the specific precision agriculture problem of weed mapping on sugarcane fields aerial images. A systematic literature review was conducted to find which computer vision methods are being used on this specific problem. The most cited methods were implemented, as well as four models of convolutional neural networks. All implemented approaches were tested using the same dataset, and their results were quantitatively and qualitatively analyzed. The obtained results were compared to a human expert made ground truth, for validation. The results indicate that the convolutional neural networks present better precision and generalize better than the classical models

scholarly journals Automatic onset detection using convolutional neural networks

2019 ◽  
Author(s):  
Willy Cornelissen ◽  
Maurício Loureiro
Keyword(s):  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Ground Truth ◽  
Onset Detection ◽  
Ground Truth Data ◽  
Music Transcription ◽  
Automatic Music Transcription ◽  
Information Research ◽  
Music Research ◽  
Music Information

A very significant task for music research is to estimate instants when meaningful events begin (onset) and when they end (offset). Onset detection is widely applied in many fields: electrocardiograms, seismographic data, stock market results and many Music Information Research(MIR) tasks, such as Automatic Music Transcription, Rhythm Detection, Speech Recognition, etc. Automatic Onset Detection(AOD) received, recently, a huge contribution coming from Artificial Intelligence (AI) methods, mainly Machine Learning and Deep Learning. In this work, the use of Convolutional Neural Networks (CNN) is explored by adapting its original architecture in order to apply the approach to automatic onset detection on audio musical signals. We used a CNN network for onset detection on a very general dataset, well acknowledged by the MIR community, and examined the accuracy of the method by comparison to ground truth data published by the dataset. The results are promising and outperform another methods of musical onset detection.

scholarly journals Aerial High-Throughput Phenotyping Enabling Indirect Selection for Grain Yield at the Early-generation Seed-limited Stages in Breeding Programs

2020 ◽  
Author(s):  
Margaret R. Krause ◽  
Suchismita Mondal ◽  
José Crossa ◽  
Ravi P. Singh ◽  
Francisco Pinto ◽  
...  
Keyword(s):  
Grain Yield ◽  
High Throughput ◽  
Vegetation Indices ◽  
Early Generation ◽  
Breeding Programs ◽  
Breeding Lines ◽  
High Throughput Phenotyping ◽  
Seed Increase ◽  
Yield Trials ◽  
Selection Of

ABSTRACTBreeding programs for wheat and many other crops require one or more generations of seed increase before replicated yield trials can be sown. Extensive phenotyping at this stage of the breeding cycle is challenging due to the small plot size and large number of lines under evaluation. Therefore, breeders typically rely on visual selection of small, unreplicated seed increase plots for the promotion of breeding lines to replicated yield trials. With the development of aerial high-throughput phenotyping technologies, breeders now have the ability to rapidly phenotype thousands of breeding lines for traits that may be useful for indirect selection of grain yield. We evaluated early generation material in the irrigated bread wheat (Triticum aestivum L.) breeding program at the International Maize and Wheat Improvement Center to determine if aerial measurements of vegetation indices assessed on small, unreplicated plots were predictive of grain yield. To test this approach, two sets of 1,008 breeding lines were sown both as replicated yield trials and as small, unreplicated plots during two breeding cycles. Vegetation indices collected with an unmanned aerial vehicle in the small plots were observed to be heritable and moderately correlated with grain yield assessed in replicated yield trials. Furthermore, vegetation indices were more predictive of grain yield than univariate genomic selection, while multi-trait genomic selection approaches that combined genomic information with the aerial phenotypes were found to have the highest predictive abilities overall. A related experiment showed that selection approaches for grain yield based on vegetation indices could be more effective than visual selection; however, selection on the vegetation indices alone would have also driven a directional response in phenology due to confounding between those traits. A restricted selection index was proposed for improving grain yield without affecting the distribution of phenology in the breeding population. The results of these experiments provide a promising outlook for the use of aerial high-throughput phenotyping traits to improve selection at the early-generation seed-limited stage of wheat breeding programs.

Sign in / Sign up

Export Citation Format

Share Document