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.

No abstract

This modelling suggested that the historically observed lower levels of resistance to pre-emergence herbicides are most likely to be due to the smaller number of weed cohorts affected by many pre-emergence herbicides. The lower number of weed cohorts affected by pre-emergence herbicides necessitated the use of additional, effective control measures, thereby reducing resistance evolution. This study highlights the advantages of applying multiple control measures to each weed cohort. © 2016 Society of Chemical Industry.

Harvest weed seed control (HWSC) techniques have been implemented in Australian cropping systems to target and reduce the number of weed seeds entering the seedbank and thereby reduce the number of problematic weeds emerging in subsequent years to infest subsequent crops. However, the influence of HWSC on ameliorating herbicide-resistance (HR) evolution has not been investigated. This research used integrated spatial modeling to examine how the frequency and efficacy of HWSC affected the evolution of resistance to initially effective herbicides. Herbicides were, in all cases, better protected from future resistance evolution when their use was combined with annual HWSC. Outbreaks of multiple HR were very unlikely to occur and were nearly always eliminated by adding annual, efficient HWSC. The efficacy of the HWSC was important, with greater reductions in the number of resistance genes achieved with higher-efficacy HWSC. Annual HWSC was necessary to protect sequences of lower-efficacy herbicides, but HWSC could still protect herbicides if it was used less often than once per year, when the HWSC and the herbicides were highly effective. Our results highlight the potential benefits of combining HWSC with effective herbicides for controlling weed populations and reducing the future evolution of HR.

In weed science and management, models are important and can be used to better understand what has occurred in management scenarios, to predict what will happen and to evaluate the outcomes of control methods. To-date, perspectives on and the understanding of weed models have been disjointed, especially in terms of how they have been applied to advance weed science and management. This paper presents a general overview of the nature and application of a full range of simulation models on the ecology, biology, and management of arable weeds, and how they have been used to provide insights and directions for decision making when long-term weed population trajectories are impractical to be determined using field experimentation. While research on weed biology and ecology has gained momentum over the past four decades, especially for species with high risk for herbicide resistance evolution, knowledge gaps still exist for several life cycle parameters for many agriculturally important weed species. More research efforts should be invested in filling these knowledge gaps, which will lead to better models and ultimately better inform weed management decision making.