Combining computer vision and deep learning to enable ultra-scale aerial phenotyping and precision agriculture:A case study of lettuce production

Aerial imagery is regularly used by crop researchers,growers and farmers to monitor crops during the growing season.To extract meaningful information from large-scale aerial images collected from the field,high-throughput phenotypic analysis solutions are required,which not only produce high-quality measures of key crop traits,but also support professionals to make prompt and reliable crop management decisions.Here,we report AirSurf,an automated and open-source analytic platform that combines modern computer vision,up-to-date machine learning,and modular software engineering in order to measure yield-related phenotypes from ultra-large aerial imagery.To quantify millions of in-field lettuces acquired by fixed-wing light aircrafts equipped with normalised difference vegetation index(NDVI)sensors,we customised AirSurf by combining computer vision algorithms and a deep-learning classifier trained with over 100,000 labelled lettuce signals.The tailored platform,AirSurf-Lettuce,is capable of scoring and categorising iceberg lettuces with high accuracy(>98%).Furthermore,novel analysis functions have been developed to map lettuce size distribution across the field,based on which associated global positioning system(GPS)tagged harvest regions have been identified to enable growers and farmers to conduct precision agricultural practises in order to improve the actual yield as well as crop marketability before the harvest.

Chlorophyllide-a Oxygenase 1(OsCAO1) Over-Expression Affects Rice Photosynthetic Rate and Grain Yield

Leaf color and photosynthesis are important factors for rice growth and development.Hence,improving the photosynthetic rate is an effective approach for increasing rice yield.We isolated a gene,chlorophyllide-a oxygenase 1(OsCAO1),which characterized a rice near-isogenic line named fgl(faded green leaf).

A Strigolactone Biosynthesis Gene Contributed to the Green Revolution in Rice

Plant architecture is a complex agronomic trait and a major factor of crop yield,which is affected by several important hormones.Strigolactones(SLs)are identified as a new class hormoneinhibiting branching in many plant species and have been shown to be involved in various developmental processes.Genetical and chemical modulation of the SL pathway is recognized as a promising approach to modify plant architecture.However,whether and how the genes involved in the SL pathway could be utilized in breeding still remain elusive.Here,we demonstrate that a partial loss-of-function allele of the SL biosynthesis gene,HIGH TILLERING AND DWARF 1/DWARF17(HTD1/D17),which encodes CAROTENOID CLEAVAGE DIOXYGENASE 7(CCD7),increases tiller number and improves grain yield in rice.We found that the HTD1 gene had been widely utilized and co-selected with Semidwarf 1(SD1),both contributing to the improvement of plant architecture in modern rice varieties since the Green Revolution in the 1960s.Understanding how phytohormone pathway genes regulate plant architecture and how they have been utilized and selected in breeding will lay the foundation for developing the rational approaches toward improving crop yield.

High-quality genome assembly of Huazhan and Tianfeng,the parents of an elite rice hybrid Tian-you-hua-zhan

High-quality rice reference genomes have accelerated the comprehensive identification of genome-wide variations and research on functional genomics and breeding.Tian-you-hua-zhan has been a leading hybrid in China over the past decade.Here,de novo genome assembly strategy optimization for the rice indica lines Huazhan(HZ)and Tianfeng(TF),including sequencing platforms,assembly pipelines and sequence depth,was carried out.The PacBio and Nanopore platforms for long-read se-quencing were utilized,with the Canu,wtdbg2,SMARTdenovo,Flye,Canu-wtdbg2,Canu-SMARTdenovo and Canu-Flye assemblers.The combination of PacBio and Canu was optimal,considering the contig N50 length,contig number,assembled genome size and polishing process.The assembled contigs were scaffolded with Hi-C data,resulting in two“golden quality”rice reference genomes,and evaluated using the scaffold N50,BUSCO,and LTR assembly index.Furthermore,42,625 and 41,815 non-transposable element genes were annotated for HZ and TF,respectively.Based on our assembly of HZ and TF,as well as Zhenshan97,Minghui63,Shuhui498 and 9311,comprehensive variations were identified using Nipponbare as a reference.The de novo assembly strategy for rice we optimized and the“golden quality”rice genomes we produced for HZ and TF will benefit rice genomics and breeding research,especially with respect to uncovering the genomic basis of the elite traits of HZ and TF.

An Exploration of Deep-Learning Based Phenotypic Analysis to Detect Spike Regions in Field Conditions for UK Bread Wheat

Wheat is one of the major crops in the world,with a global demand expected to reach 850 million tons by 2050 that is clearly outpacing current supply.The continual pressure to sustain wheat yield due to the world’s growing population under fluctuating climate conditions requires breeders to increase yield and yield stability across environments.We are working to integrate deep learning into field-based phenotypic analysis to assist breeders in this endeavour.We have utilised wheat images collected by distributed CropQuant phenotyping workstations deployed for multiyear field experiments of UK bread wheat varieties.Based on these image series,we have developed a deep-learning based analysis pipeline to segment spike regions from complicated backgrounds.As a first step towards robust measurement of key yield traits in the field,we present a promising approach that employ Fully Convolutional Network(FCN)to performsemantic segmentation of images to segment wheat spike regions.We also demonstrate the benefits of transfer learning through the use of parameters obtained from other image datasets.We found that the FCN architecture had achieved a Mean classification Accuracy(MA)>82%on validation data and>76%on test data and Mean Intersection over Union value(MIoU)>73%on validation data and and>64%on test datasets.Through this phenomics research,we trust our attempt is likely to form a sound foundation for extracting key yield-related traits such as spikes per unit area and spikelet number per spike,which can be used to assist yield-focused wheat breeding objectives in near future.

Easy MPE:Extraction of Quality Microplot Images for UAV-Based High-Throughput Field Phenotyping

Microplot extraction(PE)is a necessary image processing step in unmanned aerial vehicle-(UAV-)based research on breeding fields.At present,it is manually using ArcGIS,QGIS,or other GIS-based software,but achieving the desired accuracy is timeconsuming.We therefore developed an intuitive,easy-to-use semiautomatic program for MPE called Easy MPE to enable researchers and others to access reliable plot data UAV images of whole fields under variable field conditions.The program uses four major steps:(1)binary segmentation,(2)microplot extraction,(3)production of∗.shp files to enable further file manipulation,and(4)projection of individual microplots generated from the orthomosaic back onto the raw aerial UAV images to preserve the image quality.Crop rows were successfully identified in all trial fields.The performance of the proposed method was evaluated by calculating the intersection-over-union(IOU)ratio between microplots determined manually and by Easy MPE:the average IOU(±SD)of all trials was 91%(±3).

OsWR2 recruits HDA704 to regulate the deacetylation of H4K8ac in the promoter of OsABI5 in response to drought stress

Drought stress is a major environmental factor that limits the growth, development, and yield of rice(Oryza sativa L.). Histone deacetylases(HDACs) are involved in the regulation of drought stress responses. HDA704 is an RPD3/HDA1 class HDAC that mediates the deacetylation of H4K8(lysine 8of histone H4) for drought tolerance in rice. In this study, we show that plants overexpressing HDA704(HDA704-OE) are resistant to drought stress and sensitive to abscisic acid(ABA), whereas HDA704 knockout mutant(hda704) plants displayed decreased drought tolerance and ABA sensitivity.Transcriptome analysis revealed that HDA704 regulates the expression of ABA-related genes in response to drought stress. Moreover, HDA704 was recruited by a drought-resistant transcription factor,WAX SYNTHESIS REGULATORY 2(Os WR2), and co-regulated the expression of the ABA biosynthesis genes NINE-CIS-EPOXYCAROTENOID DIOXYGENASE 3(NCED3), NCED4, and NCED5 under drought stress. HDA704 also repressed the expression of ABA-INSENSITIVE 5(Os ABI5) and DWARF AND SMALL SEED 1(Os DSS1) by regulating H4K8ac levels in the promoter regions in response to polyethylene glycol 6000 treatment. In agreement, the loss of Os ABI5 function increased resistance to dehydration stress in rice. Our results demonstrate that HDA704 is a positive regulator of the drought stress response and offers avenues for improving drought resistance in rice.

Global Wheat Head Detection(GWHD)Dataset:A Large and Diverse Dataset of High-Resolution RGB-Labelled Images to Develop and Benchmark Wheat Head Detection Methods

The detection of wheat heads in plant images is an important task for estimating pertinent wheat traits including head population density and head characteristics such as health,size,maturity stage,and the presence of awns.Several studies have developed methods for wheat head detection from high-resolution RGB imagery based on machine learning algorithms.However,these methods have generally been calibrated and validated on limited datasets.High variability in observational conditions,genotypic differences,development stages,and head orientation makes wheat head detection a challenge for computer vision.Further,possible blurring due to motion or wind and overlap between heads for dense populations make this task even more complex.Through a joint international collaborative effort,we have built a large,diverse,and well-labelled dataset of wheat images,called the Global Wheat Head Detection(GWHD)dataset.It contains 4700 high-resolution RGB images and 190000 labelled wheat heads collected from several countries around the world at different growth stages with a wide range of genotypes.Guidelines for image acquisition,associating minimum metadata to respect FAIR principles,and consistent head labelling methods are proposed when developing new head detection datasets.The GWHD dataset is publicly available at http://www.global-wheat.com/and aimed at developing and benchmarking methods for wheat head detection.

Global Wheat Head Detection 2021:An Improved Dataset for Benchmarking Wheat Head Detection Methods

The Global Wheat Head Detection(GWHD)dataset was created in 2020 and has assembled 193,634 labelled wheat heads from 4700 RGB images acquired from various acquisition platforms and 7 countries/institutions.With an associated competition hosted in Kaggle,GWHD_2020 has successfully attracted attention from both the computer vision and agricultural science communities.From this first experience,a few avenues for improvements have been identified regarding data size,head diversity,and label reliability.To address these issues,the 2020 dataset has been reexamined,relabeled,and complemented by adding 1722 images from 5 additional countries,allowing for 81,553 additional wheat heads.We now release in 2021 a new version of the Global Wheat Head Detection dataset,which is bigger,more diverse,and less noisy than the GWHD_2020 version.

Global Co-transcriptional Splicing in Arabidopsis and the Correlation with Splicing Regulation in Mature RNAs

RNA splicing and spliceosome assembly in eukaryotes occur mainly during transcription.However,co-transcriptional splicing has not yet been explored in plants.Here,we built transcriptomes of nascent chromatin RNAs in Arabidopsis thaliana and showed that nearly all introns undergo co-transcriptional splicing,which occurs with higher efficiency for introns in protein-coding genes than for those in noncoding RNAs.Total intron number and intron position are two predominant features that correlate with co-transcriptional splicing efficiency,and introns with alternative 5′or 3′splice sites are less efficiently spliced.Furthermore,we found that mutations in genes encoding trans-acting proteins lead to more introns with increased splicing defects in nascent RNAs than in mature RNAs,and that introns with increased splicing defects in mature RNAs are inefficiently spliced at the co-transcriptional level.Collectively,our results not only uncovered widespread co-transcriptional splicing in Arabidopsis but also identified features that may affect or be affected by co-transcriptional splicing efficiency.