This paper presents a method for automating weed detection in colour images despite heavy leaf occlusion. A fully convolutional neural network is used to detect the weeds. The network is trained and validated on a total of more than 17,000 annotations of weeds in images from winter wheat fields, which have been collected using a camera mounted on an all-terrain vehicle. Hereby, the network is able to automatically detect single weed instances in cereal fields despite heavy leaf occlusion.

Many plant diseases have distinct visual symptoms which can be used to identify and classify them correctly. This paper presents a potato disease classification algorithm which leverages these distinct appearances and the recent advances in computer vision made possible by deep learning. The algorithm uses a deep convolutional neural network training it to classify the tubers into five classes, four diseases classes and a healthy potato class. The database of images used in this study, containing potatoes of different shapes, sizes and diseases, was acquired, classified, and labelled manually by experts. The models were trained over different train-test splits to better understand the amount of image data needed to apply deep learning for such classification tasks.

A variable rate drip irrigation (VRDI) system was implemented in early 2013 in a 4.05-ha area inside a drip-irrigated Cabernet Sauvignon vineyard measuring 12.5 total ha. The VRDI area was split into 140 15×15-meter irrigation zones which were watered independently during three seasons with weekly schedules based on estimated actual ET. Irrigation was scheduled with the objective of decreasing spatial variability while maintaining high yields. Compared to an adjacent, 4.05-ha, conventionally drip irrigated section of the vineyard (CDI); VRDI increased yield and water use efficiency in all three years and decreased spatial dependency and structure in 2013 and 2015.

Precision agriculture (PA) may improve the sustainability of Chinese agriculture. Ten experts were interviewed and 34 farm workers surveyed regarding their understanding, attitudes and perceptions towards PA. PA technologies were considered inaccessible, unsuitable and unnecessary for smaller farms. High cost, lack of perceived benefits, and skills and capability required to adopt PA represented barriers to adoption. Financial incentives/subsidies, the need for tangible benefits and tailored solutions to be demonstrated to farmers, and agronomic and peer support were desired. Future research should further explore PA with Chinese stakeholders and end-users in China, to inform future socio-technological developments.

The detection and identification of plant diseases is a fundamental task in sustainable crop production. An accurate estimate of disease incidence, disease severity and negative effects on yield quality and quantity is important for precision crop production, horticulture, plant breeding or fungicide screening as well as in basic and applied plant research. Particularly hyperspectral imaging of diseased plants offers insight into processes during pathogenesis. By hyperspectral imaging and subsequent data analysis routines, it was possible to realize an early detection, identification and quantification of different relevant plant diseases. Depending on the measuring scale, even subtle processes of defence and resistance mechanism of plants could be evaluated. Within this scope, recent results from studies in barley, wheat and sugar beet and their relevant foliar diseases will be presented.

The livestock sector has a key and growing role in the agricultural economy. It is a major provider of livelihood support for a large part of the world's poor. It is also an important determinant of human health and diet. Over the last three decades, the global livestock sector has rapidly evolved in response to human population growth, income growth and urbanization. And until recently, much of the focus of the sector has been geared towards satiating this demand. However, the rapid growth in demand for animal protein, has resulted in complex interactions among bio-physical resources, economic and social objectives with implications for the natural resource-base and the environment. Livestock production has a large impact on the world's natural resources and contributes significantly to environmental problems such as ecosystem pollution and degradation, global warming and climate change by emission of greenhouse gases and biodiversity loss. This paper provides an overview of the growth within the livestock sector, explores the inter-linkages between the rapidly increasing demand for animal protein and environmental consequences, as well as advances possible technical and policy interventions that are appropriate for the enhancement of the sector's role in food security, poverty reduction, economic development while contributing to environmental sustainability.

In Spain, the use of annual cover crops is a crop management practice for irrigated vineyards that allows controlling vineyard vigor and yield, which also leads to improve the crop quality. Recently, Cynodon dactylon (bermudagrass) has been reported to infest those cover crops and colonize the grapevine rows, resulting in significant yield and economic losses due to the competition for water and nutrients. From timely unmanned aerial vehicle (UAV) imagery, the objective of this research was to map C. dactylon patches in order to provide an optimized site-specific weed management. A quadrocopter UAV equipped with a point-and-shoot camera was used to collect a set of aerial red-green-blue (RGB) images over a commercial vineyard plot, coinciding with the dormant period of C. dactylon (February 2016). Object-based image analysis (OBIA) techniques were used to develop an innovative algorithm for early discrimination and mapping of C. dactylon, which had the ability to solve the limitation of spectral similarity of this weed with cover crops or bare soil. As a general result, the classified maps of the studied vineyard showed four main classes, i.e. vine, cover crop, C. dactylon and bare soil, with 85% overall accuracy. These weed maps allow developing new strategies for site-specific control of C. dactylon populations in the context of precision viticulture.

This paper investigates the effectiveness of using a UAV with dual commercial off-the-shelf (COTS) cameras, one un-modified and one modified to sense near infra-red (NIR) wavelengths to identify the onset of disease within a trial crop of potatoes. The trial was composed of 2 plots of 16 drills containing 12 tubers exposed to the blackleg disease-causing bacterial pathogen (Pectobacterium atrosepticum) in order to demonstrate best practise tuber storage and haulm destruction methods. Eleven sets of aerial data were gathered between 27/5/2016~29/7/2016 and compared with ground truth data collected on 14/7/2016. Visual analysis of the data could only detect the onset of disease and not the specific infection and resulted in a user accuracy (UA) of 83% and producer accuracy (PA) of 78%, with a total accuracy (TA) of 91% and Kappa coefficient (K) of 0.75. The building blocks of an automated classification routine have been constructed using pixel and object based image analysis (OBIA) methods, which have shown promising first results (UA 65%, PA 73%, TA 87%, K 0.61) but requires further refinement to achieve an equivalent level of accuracy as that of the visual analysis.

LiDAR (Light detection and ranging) technology is an alternative to current manual methods of canopy geometry estimations in orange trees. The objective of this work was to compare different types of canopy volume estimations of orange trees, some inspired on manual methods and others based on a LiDAR sensor. A point cloud was generated for 25 individual trees using a laser scanning system. The convex-hull and the alpha-shape surface reconstruction algorithms were tested. LiDAR derived models are able to represent orange trees more accurately than traditional methods. However, results differ significantly from the current manual method. In addition, different 3D modeling algorithms resulted in different canopy volume estimations. Therefore, a new standard method should be developed and established.

Precision agriculture technologies have been adopted individually and in bundles. A sample of 348 Kansas Farm Management Association farm-level observations provides insight into technology adoption patterns of precision agriculture technologies. Estimated transition probabilities shed light on how adoption paths lead to bundling of technologies. Three information intensive technologies were assigned to one of eight possible bundles, and the sequence of adoption was examined using Markov transition processes. The probability that farms remain with the same bundle or transition to a different bundle by the next time period are reported. Farms with the complete bundle of all three technologies were likely to persist with their current technology.

This study aimed to examine the effects of replacing rolled barley (high in starch) with citrus pulp (high in digestible fibre) in a supplement on intake and performance of young growing cattle offered grass silage ad libitum for 101 days. Weaned, early- and late-maturing breed, male suckled beef calves (n=120) were blocked by sire breed, gender and weight and from within block randomly assigned to one of two concentrate supplements based mainly on rolled barley (BAR) or citrus pulp (CIT) and formulated to have similar concentrations of true protein digestible in the small intestine. On day 87, blood samples were taken before and 2 h after feeding, and rumen fluid samples were collected 2 h post-feeding. Supplement type did not affect (P>0.05) grass silage intake, live weight gain, final live weight, ultrasonically assessed body composition or measurements of skeletal size. Rumen pH (6.64 v. 6.79), ammonia (51 v. 81 mg/l) and acetate-to-propionate ratio (2.7 v. 3.2) were lower (P<0.001) for CIT than BAR. In conclusion, citrus pulp can replace barley in concentrate supplements for growing cattle without negatively affecting performance.

We investigated relationship between pasture biomass and measures of height and NDVI (normalised difference vegetation index). The pastures were tall fescue (Festuca arundinacea), perennial ryegrass (Lolium perenne), and phalaris (Phalaris aquatica) located in Tasmania, Victoria and in the Northern Tablelands of NSW, Australia. Using the Trimble® GreenSeeker® Handheld active optical sensor (AOS) to measure NDVI, and a rising plate meter, the optimal model to estimate green dry biomass (GDM) during two years was a combination of NDVI and falling plate height index. The combined index was significantly correlated with GDM in each region during winter and spring (r2=0.62–0.77, P<0.001). Regional calibrations provided a smaller error in estimates of green biomass, required for potential application in the field, compared to a single overall calibration. Data collected in a third year will be used to test the accuracy of the models.

This work assesses the potential of Sentinel-2A images in precision agriculture for Barley production in a case study. Two workflows are proposed: 1) images were acquired with a relatively simple methodology to follow the crop development; 2) two images around harvest time were downloaded and processed using a more complex and accurate methodology to calculate four vegetation indices (NDVI, WDRVI, GRVI and GNDVI) to be correlated to yield with linear regression models. Yield data were acquired with a yield monitor installed in a combine harvester. Green-based vegetation indices performed slightly better. However, the highest correlation coefficient was 0.48. Better results may be achieved with earlier imagery and other vegetation indices. Sentinel-2 is a promising tool for precision agriculture in large arable crop fields.

Genomic selection is rapidly becoming the state-of-the-art genetic selection methodology in dairy cattle breeding schemes around the world. The objective of this paper was to explore possibilities to apply genomic selection for traits related to dairy cow robustness. Deterministic simulations indicate that replacing progeny testing with genomic selection may favour genetic response for production traits at the expense of robustness traits, owing to a disproportional change in accuracies obtained across trait groups. Nevertheless, several options are available to improve the accuracy of genomic selection for robustness traits. Moreover, genomic selection opens up the opportunity to begin selection for new traits using specialised reference populations of limited size where phenotyping of large populations of animals is currently prohibitive. Reference populations for such traits may be nucleus-type herds, research herds or pooled data from (international) research experiments or research herds. The RobustMilk project has set an example for the latter approach, by collating international data for progesterone-based traits, feed intake and energy balance-related traits. Reference population design, both in terms of relatedness of the animals and variability in phenotypic performance, is important to optimise the accuracy of genomic selection. Use of indicator traits, combined with multi-trait genomic prediction models, can further contribute to improved accuracy of genomic prediction for robustness traits. Experience to date indicates that for newly recorded robustness traits that are negatively correlated with the main breeding goal, cow reference populations of ⩾10 000 are required when genotyping is based on medium- or high-density single-nucleotide polymorphism arrays. Further genotyping advances (e.g. sequencing) combined with post-genomics technologies will enhance the opportunities for (genomic) selection to improve cow robustness.

New remote sensing technologies have provided unprecedented results in vineyard monitoring. The aim of this work was to evaluate different sources of images and processing methodologies to describe spatial variability of spectral-based and canopy-based vegetation indices within a vineyard, and their relationship with productive and qualitative vine parameters. Comparison between image-derived indices from Sentinel 2 NDVI, unfiltered and filtered UAV NDVI and with agronomic features have been performed. UAV images allow calculating new non-spectral indices based on canopy architecture that provide additional and useful information to the growers with regards to within-vineyard management zone delineation.

The most important factors known to influence the eating quality of beef are well established and include both pre- and post-slaughter events with many of the determinants interacting with each other. A substantial programme of work has been conducted by the Agri-Food and Biosciences Institute in Northern Ireland aimed at quantifying those factors of most importance to the local beef industry. Post-slaughter effects such as carcase chilling and electrical stimulation, ageing, carcase hanging and cooking method have been shown to have a significant impact on eating quality when compared with pre-slaughter activities such as animal handling and lairage time in the Northern Ireland studies. However, the effect of animal breed, particularly the use of dairy breed animals, was shown to significantly improve eating quality. Many of these factors were found to interact with each other.

This work was to evaluate the differences of soil and plant analysis development (SPAD) and normalized difference vegetation index (NDVI) readings and their relationship with leaf nitrogen accumulation (LNA). The study explored new indices to diagnose nitrogen (N) status. These indices were obtained by multiplying SPAD readings and leaf area index (LAI). Linear regression relationships between Chlorophyll values and N indicators showed the SPAD readings (Chl: LNA=0.0546×Chl-0.479, R2=0.94***, P<0.001). The projected results suggested that Chl values could play an important role for improving N status diagnosis from stem elongation to heading stages in paddy rice.

This paper evaluates the potential of very high resolution multispectral (Worldview-3) satellite imagery for mapping yield parameters in avocado and macadamia orchards. An evaluation of 18 structural and pigment based vegetation indices (VIs) derived from Worldview-3 imagery identified a positive relationship to nut/ fruit weight (kg/tree) R2>0.69 for macadamia and R2>0.68 for avocado; and nut/ fruit number (per tree) R2>0.6 for macadamia and R2>0.61 for avocado. Using the algorithms derived between the optimal VI and the measured parameter, yield and nut/ fruit number maps were derived for each block. In the absence of a commercial yield monitor, the resulting yield maps offer significant benefit to growers for improving orchard management, harvest scheduling, and forward selling decisions.