This study outlines a new method of automatically estimating weed species and growth stages (from cotyledon until eight leaves are visible) of in situ images covering 18 weed species or families. Images of weeds growing within a variety of crops were gathered across variable environmental conditions with regards to soil types, resolution and light settings. Then, 9649 of these images were used for training the computer, which automatically divided the weeds into nine growth classes. The performance of this proposed convolutional neural network approach was evaluated on a further set of 2516 images, which also varied in term of crop, soil type, image resolution and light conditions. The overall performance of this approach achieved a maximum accuracy of 78% for identifying Polygonum spp. and a minimum accuracy of 46% for blackgrass. In addition, it achieved an average 70% accuracy rate in estimating the number of leaves and 96% accuracy when accepting a deviation of two leaves. These results show that this new method of using deep convolutional neural networks has a relatively high ability to estimate early growth stages across a wide variety of weed species.

For decades, significant effort has been put into the development of plant detection and classification algorithms. However, it has been difficult to compare the performance of the different algorithms, due to the lack of a common testbed, such as a public available annotated reference dataset. In this paper, we present the Open Plant Phenotype Database (OPPD), a public dataset for plant detection and plant classification. The dataset contains 7590 RGB images of 47 plant species. Each species is cultivated under three different growth conditions, to provide a high degree of diversity in terms of visual appearance. The images are collected at the semifield area at Aarhus University, Research Centre Flakkebjerg, Denmark, using a customized data acquisition platform that provides well-illuminated images with a ground resolution of ∼6.6 px mm ${}^{-1}$ . All images are annotated with plant species using the EPPO encoding system, bounding box annotations for detection and extraction of individual plants, applied growth conditions and time passed since seeding. Additionally, the individual plants have been tracked temporally and given unique IDs. The dataset is accompanied by two experiments for: (1) plant instance detection and (2) plant species classification. The experiments introduce evaluation metrics and methods for the two tasks and provide baselines for future work on the data.

Concerns around herbicide resistance, human risk, and the environmental impacts of current weed control strategies have led to an increasing demand for alternative weed management methods. Many new weed management strategies are under development; however, the poor availability of accurate weed maps, and a lack of confidence in the outcomes of alternative weed management strategies, has hindered their adoption. Developments in field sampling and processing, combined with spatial modelling, can support the implementation and assessment of new and more integrated weed management strategies. Our review focuses on the biological and mathematical aspects of assembling within-field weed models. We describe both static and spatio-temporal models of within-field weed distributions (including both cellular automata (CA) and non-CA models), discussing issues surrounding the spatial processes of weed dispersal and competition and the environmental and anthropogenic processes that affect weed spatial and spatio-temporal distributions. We also examine issues surrounding model uncertainty. By reviewing the current state-of-the-art in both static and temporally dynamic weed spatial modelling we highlight some of the strengths and weaknesses of current techniques, together with current and emerging areas of interest for the application of spatial models, including targeted weed treatments, economic analysis, herbicide resistance and integrated weed management, the dispersal of biocontrol agents, and invasive weed species.

Annual ryegrass species (Lolium spp.) infest cereal crops worldwide. Ryegrass populations with multiple resistance to the acetyl coenzyme A carboxylase (ACCase) and acetolactate synthase (ALS) inhibitors are an increasing problem in several European countries. We investigated the resistance pattern and level of resistance in ryegrass populations collected in Denmark, Greece and Italy and studied the diversity of mechanisms endowing resistance, both target-site and metabolism based. All populations showed high resistance indexes (RI) to the ALS inhibitors, iodosufuron-methyl-sodium + mesosulfuron-methyl (RI from 8 to 70), whereas the responses to the two ACCase inhibitors, clodinafop-propargyl and pinoxaden, differed. The Greek and Italian populations were moderately to highly resistant to clodinafop (RI > 8) and showed low to moderate resistance to pinoxaden (RI ranged from 3 to 13) except for one Italian population. In contrast, the Danish Lolium populations showed low to moderate resistance to clodinafop (RI ranged from 2 to 7) and only one population was resistant to pinoxaden. Different mutant ACCase alleles (Leu₁₇₈₁, Cys₂₀₂₇, Asn₂₀₄₁, Val₂₀₄₁, Gly₂₀₇₈, Arg₂₀₈₈, Ala₂₀₉₆) and ALS alleles (Gly₁₂₂, Ala₁₉₇, Gln₁₉₇, Leu₁₉₇, Ser₁₉₇, Thr₁₉₇, Val₂₀₅, Asn₃₇₆, Glu₃₇₆, Leu₅₇₄) endowing resistance were detected in the Greek and Italian populations. In several plants, no mutated ALS and ACCase alleles were found showing a great heterogeneity within and among the Greek and Italian populations. Conversely, no mutant ACCase alleles were identified in the four Danish populations and only one mutant ALS allele (Leu₅₇₄) was detected in two Danish populations. The expression level of nitronate monooxygenase (NMO), glutathione S-transferase (GST) and cytochrome P450s (CYP72A1 and CYP72A2) varied broadly among populations and individual plants within the populations. Constitutive up-regulation of GST, CYP72A1 and CYP72A2 was detected in resistant plants respect to susceptible plants in one Danish and one Italian population. It appears that the mechanisms underlying resistance are rather complex and diversified among Lolium spp. populations from the three countries, coevolution of both target-site resistance and metabolic based herbicide resistance appears to be a common feature in Denmark and Italy. This must be considered and carefully evaluated in adopting resistance management strategies to control Lolium spp. in cereal crops.

Loose silky bentgrass (Apera spica-venti) is an important weed in Europe with a recent increase in herbicide resistance cases. The lack of genetic information about this noxious weed limits its biological understanding such as growth, reproduction, genetic variation, molecular ecology and metabolic herbicide resistance. This study produced a reference transcriptome for A. spica-venti from different tissues (leaf, root, stem) and various growth stages (seed at phenological stages 05, 07, 08, 09). The de novo assembly was performed on individual and combined dataset followed by functional annotations. Individual transcripts and gene families involved in metabolic based herbicide resistance were identified. Eight separate transcriptome assemblies were performed and compared. The combined transcriptome assembly consists of 83,349 contigs with an N50 and average contig length of 762 and 658 bp, respectively. This dataset contains 74,724 transcripts consisting of total 54,846,111 bp. Among them 94% had a homologue to UniProtKB, 73% retrieved a GO mapping, and 50% were functionally annotated. Compared with other grass species, A. spica-venti has 26% proteins in common to Brachypodium distachyon, and 41% to Lolium spp. Glycosyltransferases had the highest number of transcripts in each tissue followed by the cytochrome P450s. The GSTF1 and CYP89A2 transcripts were recovered from the majority of tissues and aligned at a maximum of 66 and 30% to proven herbicide resistant allele from Alopecurus myosuroides and Lolium rigidum, respectively. De novo transcriptome assembly enabled the generation of the first reference transcriptome of A. spica-venti. This can serve as stepping stone for understanding the metabolic herbicide resistance as well as the general biology of this problematic weed. Furthermore, this large-scale sequence data is a valuable scientific resource for comparative transcriptome analysis for Poaceae grasses.