SummaryNon‐destructive assessment of herbicide effects may be able to support integrated weed management. To test whether effects of herbicides on canopy variables could be detected by sensors, two crops were used as models and treated with herbicides at BBCH 20 using a logarithmic sprayer. Twelve days after spraying at BBCH 25 and 42 days after sowing, nine sensor systems scanned a spring barley and an oilseed rape field experiment sown at different densities and sprayed with increasing field rates of glyphosate and tribenuron‐methyl. The objective was to compare ED50s for crops and weeds derived by the different sensors in relation to crop density and herbicides. Although sensors were not directly developed to detect herbicide symptoms, they all detected changes in canopy colours or height and crop density. Generally ED50s showed the same pattern in response to crop density within herbicide, but there were marked differences between barley and oilseed rape. We suggest that the results of comparing the various sensor outputs could become a stepping stone to future standardisation for the benefit of the research and development of sensors that will detect herbicide effect on crops and weeds, particularly at the most vulnerable stages of development of the canopy.

Many different sensors have been proposed to estimate plant status parameters like nutrition status, coverage or plant size. Such parameters are key factors for precise management. This study combined four different sensors in a field trial with spring barley and oil seed rape. The following commercial sensors were used: LiDAR, spectrometer, ultrasonic device, and a commercial opto-electronic device. Spectral indices were calculated from the spectrometer and opto-electronic devices, and plant height from the ultrasonic and LiDAR sensors. A robotic software framework was used for simultaneous measurements with multiple sensors. The fusion of features from multiple sensors permitted the estimation of health status parameters for sensitive plants in a herbicide stress trial.

Gerhards R, Gutjahr C, Weis M, Keller M, Sökefeld M, Möhring J & Piepho HP (2011). Using precision farming technology to quantify yield effects attributed to weed competition and herbicide application. Weed Research 52, 6–15.SummaryField experiments using precision farming technology and Geographic Information Systems, following a so‐called Precision Experimental Design, were conducted in maize, winter barley and winter wheat and compared with two randomised plot experiments in maize to quantify yield effects attributed to weed competition and weed control. Fields were divided into cells, and weed densities for all weed species, soil conductivity and grain yield were measured in each cell. Untreated plots and herbicide treatments against grass weeds or broad‐leaved weeds were included in all three experiments. Chenopodium album, Polygonum spp. and Echinochloa crus‐galli were the dominating weed species in maize. Stellaria media, Veronica hederifolia, Matricaria chamomilla, Alopecurus myosuroides and Galium aparine were the most abundant weed species in the winter barley and winter wheat fields. All species were distributed heterogeneously within the fields with densities ranging from 0 to more than 200 plants m−2. In the Precision Experimental Design, it was found that grass‐weed competition and herbicide application had a significant effect on grain yield, using a linear mixed model with spatial correlation structure to determine the effects of groups of weed species, soil variability and herbicide application on grain yield separately. When a conventional plot experiment was set up in the same field, no statistically significant grain yield difference between the treatments was found. The results highlight the benefits of Precision Experimental Design for studying weed–crop competition. Data can be used to calculate yield loss functions for groups of weed species and to create a decision support system for site‐specific weed control.

Variability of weed infestation needs to be assessed for site-specific weed management. Since manual weed sampling is too time consuming for prctical applications, a system for automatic weed sampling was developed. The system uses bispectral images, which are processed to derive shape features of the plants. The shape features are used for the discrimination of weed and crop species by using a classification step. In this paper we evaluate different classification algorithms with main focus on k-nearest neighbours, decision tree learning and Support Vector Machine classifiers. Data mining techniques were applied to select an optimal subset of the shape features, which then were used for the classification. Since the classification is a crucial step for the weed detection, three different classification algorithms are tested and their influence on the results is assessed. The plant shape varies between different species and also within one species at different growth stages. The training of the classifiers is run by using prototype information which is selected manually from the images. Performance measures for classification accuracy are evaluated by using cross validation techniques and by comparing the results with manually assessed weed infestation.

Grain yield often varies within agricultural fields as a result of the variation in soil characteristics, competition from weeds, management practices and their causal interactions. To implement appropriate management decisions, yield variability needs to be explained and quantified. A new experimental design was established and tested in a field experiment to detect yield variation in relation to the variation in soil quality, the heterogeneity of weed distribution and weed control within a field. Weed seedling distribution and density, apparent soil electrical conductivity (EC a ) and grain yield were recorded and mapped in a 3.5 ha winter wheat field during 2005 and 2006. A linear mixed model with an anisotropic spatial correlation structure was used to estimate the effect of soil characteristics, weed competition and herbicide treatment on crop yield. The results showed that all properties had a strong effect on grain yield. By adding herbicide costs and current grain price into the model, thresholds of weed density were derived for site-specific weed control. This experimental approach enables the variation of yield within agricultural fields to be explained, and an understanding of the effects on yield of the factors that affect it and their causal interactions to be gained. The approach can be applied to improve decision algorithms for the patch spraying of weeds.