Proof of concept of automated mapping of weeds in arable fields

The project investigated whether it would be possible to remove the main technical hindrance to precision application of herbicides to arable crops in the UK, namely creating geo-referenced weed maps for each field. The ultimate goal is an information system so that agronomists and farmers can plan precision weed control and create spraying maps. The project focussed on black-grass in wheat, but research was also carried out on barley and beans and on wild-oats, barren brome, rye-grass, cleavers and thistles which form stable patches in arable fields. Farmers may also make special efforts to control them. Using cameras mounted on farm machinery, the project explored the feasibility of automating the process of mapping black-grass in fields. Geo-referenced images were captured from June to December 2009, using sprayers, a tractor, combine harvesters and on foot. Cameras were mounted on the sprayer boom, on windows or on top of tractor and combine cabs and images were captured with a range of vibration levels and at speeds up to 20 km h-1. For acceptability to farmers, it was important that every image containing black-grass was classified as containing black-grass; false negatives are highly undesirable. The software algorithms recorded no false negatives in sample images analysed to date, although some black-grass heads were unclassified and there were also false positives. The density of black-grass heads per unit area estimated by machine vision increased as a linear function of the actual density with a mean detection rate of 47% of black-grass heads in sample images at T3 within a density range of 13 to 1230 heads m-2. A final part of the project was to create geo-referenced weed maps using software written in previous HGCA-funded projects and two examples show that geo-location by machine vision compares well with manually-mapped weed patches. The consortium therefore demonstrated for the first time the feasibility of using a GPS-linked computer-controlled camera system mounted on farm machinery (tractor, sprayer or combine) to geo-reference black-grass in winter wheat between black-grass head emergence and seed shedding.

Final report on research project PS2010, proof of concept of automated mapping of weeds in arable fields

Many weeds occur in patches but farmers frequently spray whole fields to control the weeds in these patches. Given a geo-referenced weed map, technology exists to confine spraying to these patches. Adoption of patch spraying by arable farmers has, however, been negligible partly due to the difficulty of constructing weed maps. Building on previous DEFRA and HGCA projects, this proposal aims to develop and evaluate a machine vision system to automate the weed mapping process. The project thereby addresses the principal technical stumbling block to widespread adoption of site specific weed management (SSWM). The accuracy of weed identification by machine vision based on a single field survey may be inadequate to create herbicide application maps. We therefore propose to test the hypothesis that sufficiently accurate weed maps can be constructed by integrating information from geo-referenced images captured automatically at different times of the year during normal field activities. Accuracy of identification will also be increased by utilising a priori knowledge of weeds present in fields. To prove this concept, images will be captured from arable fields on two farms and processed offline to identify and map the weeds, focussing especially on black-grass, wild oats, barren brome, couch grass and cleavers. As advocated by Lutman et al. (2002), the approach uncouples the weed mapping and treatment processes and builds on the observation that patches of these weeds are quite stable in arable fields. There are three main aspects to the project. 1) Machine vision hardware. Hardware component parts of the system are one or more cameras connected to a single board computer (Concurrent Solutions LLC) and interfaced with an accurate Global Positioning System (GPS) supplied by Patchwork Technology. The camera(s) will take separate measurements for each of the three primary colours of visible light (red, green and blue) in each pixel. The basic proof of concept can be achieved in principle using a single camera system, but in practice systems with more than one camera may need to be installed so that larger fractions of each field can be photographed. Hardware will be reviewed regularly during the project in response to feedback from other work packages and updated as required. 2) Image capture and weed identification software. The machine vision system will be attached to toolbars of farm machinery so that images can be collected during different field operations. Images will be captured at different ground speeds, in different directions and at different crop growth stages as well as in different crop backgrounds. Having captured geo-referenced images in the field, image analysis software will be developed to identify weed species by Murray State and Reading Universities with advice from The Arable Group. A wide range of pattern recognition and in particular Bayesian Networks will be used to advance the state of the art in machine vision-based weed identification and mapping. Weed identification algorithms used by others are inadequate for this project as we intend to collect and correlate images collected at different growth stages. Plants grown for this purpose by Herbiseed will be used in the first instance. In addition, our image capture and analysis system will include plant characteristics such as leaf shape, size, vein structure, colour and textural pattern, some of which are not detectable by other machine vision systems or are omitted by their algorithms. Using such a list of features observable using our machine vision system, we will determine those that can be used to distinguish weed species of interest. 3) Weed mapping. Geo-referenced maps of weeds in arable fields (Reading University and Syngenta) will be produced with advice from The Arable Group and Patchwork Technology. Natural infestations will be mapped in the fields but we will also introduce specimen plants in pots to facilitate more rigorous system evaluation and testing. Manual weed maps of the same fields will be generated by Reading University, Syngenta and Peter Lutman so that the accuracy of automated mapping can be assessed. The principal hypothesis and concept to be tested is that by combining maps from several surveys, a weed map with acceptable accuracy for endusers can be produced. If the concept is proved and can be commercialised, systems could be retrofitted at low cost onto existing farm machinery. The outputs of the weed mapping software would then link with the precision farming options already built into many commercial sprayers, allowing their use for targeted, site-specific herbicide applications. Immediate economic benefits would, therefore, arise directly from reducing herbicide costs. SSWM will also reduce the overall pesticide load on the crop and so may reduce pesticide residues in food and drinking water, and reduce adverse impacts of pesticides on non-target species and beneficials. Farmers may even choose to leave unsprayed some non-injurious, environmentally-beneficial, low density weed infestations. These benefits fit very well with the anticipated legislation emerging in the new EU Thematic Strategy for Pesticides which will encourage more targeted use of pesticides and greater uptake of Integrated Crop (Pest) Management approaches, and also with the requirements of the Water Framework Directive to reduce levels of pesticides in water bodies. The greater precision of weed management offered by SSWM is therefore a key element in preparing arable farming systems for the future, where policy makers and consumers want to minimise pesticide use and the carbon footprint of farming while maintaining food production and security. The mapping technology could also be used on organic farms to identify areas of fields needing mechanical weed control thereby reducing both carbon footprints and also damage to crops by, for example, spring tines. Objective i. To develop a prototype machine vision system for automated image capture during agricultural field operations; ii. To prove the concept that images captured by the machine vision system over a series of field operations can be processed to identify and geo-reference specific weeds in the field; iii. To generate weed maps from the geo-referenced, weed plants/patches identified in objective (ii).

Cost733cat: a database of weather and circulation type classifications

A new database of weather and circulation type catalogs is presented comprising 17 automated classification methods and five subjective classifications. It was compiled within COST Action 733 "Harmonisation and Applications of Weather Type Classifications for European regions" in order to evaluate different methods for weather and circulation type classification. This paper gives a technical description of the included methods using a new conceptual categorization for classification methods reflecting the strategy for the definition of types. Methods using predefined types include manual and threshold based classifications while methods producing types derived from the input data include those based on eigenvector techniques, leader algorithms and optimization algorithms. In order to allow direct comparisons between the methods, the circulation input data and the methods' configuration were harmonized for producing a subset of standard catalogs of the automated methods. The harmonization includes the data source, the climatic parameters used, the classification period as well as the spatial domain and the number of types. Frequency based characteristics of the resulting catalogs are presented, including variation of class sizes, persistence, seasonal and inter-annual variability as well as trends of the annual frequency time series. The methodological concept of the classifications is partly reflected by these properties of the resulting catalogs. It is shown that the types of subjective classifications compared to automated methods show higher persistence, inter-annual variation and long-term trends. Among the automated classifications optimization methods show a tendency for longer persistence and higher seasonal variation. However, it is also concluded that the distance metric used and the data preprocessing play at least an equally important role for the properties of the resulting classification compared to the algorithm used for type definition and assignment.

Auditory distraction eliminates retrieval induced forgetting: implications for the processing of unattended sound

The Retrieval-Induced Forgetting (RIF) paradigm includes three phases: (a) study/encoding of category exemplars, (b) practicing retrieval of a sub-set of those category exemplars, and (c) recall of all exemplars. At the final recall phase, recall of items that belong to the same categories as those items that undergo retrieval-practice, but that do not undergo retrieval-practice, is impaired. The received view is that this is because retrieval of target category-exemplars (e.g., ‘Tiger’ in the category Four-legged animal) requires inhibition of non-target category-exemplars (e.g., ‘Dog’ and ‘Lion’) that compete for retrieval. Here, we used the RIF paradigm to investigate whether ignoring auditory items during the retrieval-practice phase modulates the inhibitory process. In two experiments, RIF was present when retrieval-practice was conducted in quiet and when conducted in the presence of spoken words that belonged to a category other than that of the items that were targets for retrieval-practice. In contrast, RIF was abolished when words that either were identical to the retrieval-practice words or were only semantically related to the retrieval-practice words were presented as background speech. The results suggest that the act of ignoring speech can reduce inhibition of the non-practiced category-exemplars, thereby eliminating RIF, but only when the spoken words are competitors for retrieval (i.e., belong to the same semantic category as the to-be-retrieved items).

Ever decreasing circles: speech production in semantic dementia

We explored the impact of a degraded semantic system on lexical, morphological and syntactic complexity in language production. We analysed transcripts from connected speech samples from eight patients with semantic dementia (SD) and eight age-matched healthy speakers. The frequency distributions of nouns and verbs were compared for hand-scored data and data extracted using text-analysis software. Lexical measures showed the predicted pattern for nouns and verbs in hand-scored data, and for nouns in software-extracted data, with fewer low frequency items in the speech of the patients relative to controls. The distribution of complex morpho-syntactic forms for the SD group showed a reduced range, with fewer constructions that required multiple auxiliaries and inflections. Finally, the distribution of syntactic constructions also differed between groups, with a pattern that reflects the patients’ characteristic anomia and constraints on morpho-syntactic complexity. The data are in line with previous findings of an absence of gross syntactic errors or violations in SD speech. Alterations in the distributions of morphology and syntax, however, support constraint satisfaction models of speech production in which there is no hard boundary between lexical retrieval and grammatical encoding.

Validity of the ice shelf water plume concept beneath Filchner-Ronne Ice Shelf

In winter, brine rejection from sea ice formation and export in the Weddell Sea, offshore of Filchner-Ronne Ice Shelf (FRIS), leads to the formation of High Salinity Shelf Water (HSSW). This dense water mass enters the cavity beneath FRIS by sinking southward down the sloping continental shelf towards the grounding line. Melting occurs when the HSSW encounters the ice shelf, and the meltwater released cools and freshens the HSSW to form a water mass known as Ice Shelf Water (ISW). If this ISW rises, the ‘ice pump’ is initiated (Lewis and Perkin, 1986), whereby the ascending ISW becomes supercooled and deposits marine ice at shallower locations due to the pressure increase in the in-situ freezing temperature. Sandh¨ager et al. (2004) were able to infer the thickness patterns of marine ice deposits at the base of FRIS (figure 1), so the primary aim of this work is to try to understand the ocean flows that determine these patterns. The plume model we use to investigate ISW flow is described fully by Holland and Feltham (accepted) so only a relatively brief outline is presented here. The plume is simulated by combining a parameterisation of ice shelf basal interaction and a multiplesize- class frazil dynamics model with an unsteady, depth-averaged reduced-gravity plume model. In the model an active region of ISW evolves above and within an expanse of stagnant ambient fluid, which is considered to be ice-free and has fixed profiles of temperature and salinity. The two main assumptions of the model are that there is a well-mixed layer underneath the ice shelf and that the ambient fluid outside the plume is stagnant with fixed properties. The topography of the ice shelf that the plume flows beneath is set to the FRIS ice shelf draft calculated by Sandh¨ager et al. (2004) masked with the grounding line from the Antarctic Digital Database (ADD Consortium, 2002). To initiate the plumes, we assume that the intrusion of dense HSSW initially causes melting at the points on the grounding line where the glaciological tributaries feeding FRIS go afloat.

Sea surface temperature climate change initiative: alternative image classification algorithms for sea-ice affected oceans

We present a Bayesian image classification scheme for discriminating cloud, clear and sea-ice observations at high latitudes to improve identification of areas of clear-sky over ice-free ocean for SST retrieval. We validate the image classification against a manually classified dataset using Advanced Along Track Scanning Radiometer (AATSR) data. A three way classification scheme using a near-infrared textural feature improves classifier accuracy by 9.9 % over the nadir only version of the cloud clearing used in the ATSR Reprocessing for Climate (ARC) project in high latitude regions. The three way classification gives similar numbers of cloud and ice scenes misclassified as clear but significantly more clear-sky cases are correctly identified (89.9 % compared with 65 % for ARC). We also demonstrate the poetential of a Bayesian image classifier including information from the 0.6 micron channel to be used in sea-ice extent and ice surface temperature retrieval with 77.7 % of ice scenes correctly identified and an overall classifier accuracy of 96 %.

Integrated visual vocabulary in latent Dirichlet allocation-based scene classification for IKONOS image

Scene classification based on latent Dirichlet allocation (LDA) is a more general modeling method known as a bag of visual words, in which the construction of a visual vocabulary is a crucial quantization process to ensure success of the classification. A framework is developed using the following new aspects: Gaussian mixture clustering for the quantization process, the use of an integrated visual vocabulary (IVV), which is built as the union of all centroids obtained from the separate quantization process of each class, and the usage of some features, including edge orientation histogram, CIELab color moments, and gray-level co-occurrence matrix (GLCM). The experiments are conducted on IKONOS images with six semantic classes (tree, grassland, residential, commercial/industrial, road, and water). The results show that the use of an IVV increases the overall accuracy (OA) by 11 to 12% and 6% when it is implemented on the selected and all features, respectively. The selected features of CIELab color moments and GLCM provide a better OA than the implementation over CIELab color moment or GLCM as individuals. The latter increases the OA by only ∼2 to 3%. Moreover, the results show that the OA of LDA outperforms the OA of C4.5 and naive Bayes tree by ∼20%. © 2014 Society of Photo-Optical Instrumentation Engineers (SPIE) [DOI: 10.1117/1.JRS.8.083690]

Distinction between semantic and episodic memory: The underlying mechanisms for autonoetic consciousness

In 1972, episodic and semantic memories were considered to reflect different types of knowledge (Tulving, 1972). However, these early definitions encountered many difficulties. Now, Episodic and semantic memories are discussed in terms of awareness associated with retrieval (Wheeler, Stuss, & Tulving, 1997): Autonoetic consciousness (i.e., feeling of remembering) is considered associated with retrieval from the episodic memory system, while noetic consciousness (i.e., feeling of knowing) is considered characterized by retrieval from the semantic memory system. The present article investigated determinants of autonoetic consciousness in order to clarify characteristics of perceptual knowledge that is being recalled, the more strongly the individual feels autonoetic consciousness during retrieval, and that autonoetic consciousness is based on rich sensory-perceptual knowledge. Furthermore, we suggested that the parietal and frontal lobes mediate the process of generating autonoetic consciousness. This suggested that sensory-perceptual knowledge, the parietal lobe and the frontal lobe are important factors for discriminating episodic memory afrom semantic memory.

Instantaneous threat detection based on a semantic representation of activities, zones and trajectories

Threat detection is a challenging problem, because threats appear in many variations and differences to normal behaviour can be very subtle. In this paper, we consider threats on a parking lot, where theft of a truck’s cargo occurs. The threats range from explicit, e.g. a person attacking the truck driver, to implicit, e.g. somebody loitering and then fiddling with the exterior of the truck in order to open it. Our goal is a system that is able to recognize a threat instantaneously as they develop. Typical observables of the threats are a person’s activity, presence in a particular zone and the trajectory. The novelty of this paper is an encoding of these threat observables in a semantic, intermediate-level representation, based on low-level visual features that have no intrinsic semantic meaning themselves. The aim of this representation was to bridge the semantic gap between the low-level tracks and motion and the higher-level notion of threats. In our experiments, we demonstrate that our semantic representation is more descriptive for threat detection than directly using low-level features. We find that a person’s activities are the most important elements of this semantic representation, followed by the person’s trajectory. The proposed threat detection system is very accurate: 96.6 % of the tracks are correctly interpreted, when considering the temporal context.

Datasets related to in-land water for limnology and remote sensing applications: distance-to-land, distance-to-water, water-body identifier and lake-centre co-ordinates

Datasets containing information to locate and identify water bodies have been generated from data locating static-water-bodies with resolution of about 300 m (1/360 deg) recently released by the Land Cover Climate Change Initiative (LC CCI) of the European Space Agency. The LC CCI water-bodies dataset has been obtained from multi-temporal metrics based on time series of the backscattered intensity recorded by ASAR on Envisat between 2005 and 2010. The new derived datasets provide coherently: distance to land, distance to water, water-body identifiers and lake-centre locations. The water-body identifier dataset locates the water bodies assigning the identifiers of the Global Lakes and Wetlands Database (GLWD), and lake centres are defined for in-land waters for which GLWD IDs were determined. The new datasets therefore link recent lake/reservoir/wetlands extent to the GLWD, together with a set of coordinates which locates unambiguously the water bodies in the database. Information on distance-to-land for each water cell and the distance-to-water for each land cell has many potential applications in remote sensing, where the applicability of geophysical retrieval algorithms may be affected by the presence of water or land within a satellite field of view (image pixel). During the generation and validation of the datasets some limitations of the GLWD database and of the LC CCI water-bodies mask have been found. Some examples of the inaccuracies/limitations are presented and discussed. Temporal change in water-body extent is common. Future versions of the LC CCI dataset are planned to represent temporal variation, and this will permit these derived datasets to be updated.