Search

Robust plant image segmentation under natural illumination condition is still a challenging process for vision-based agricultural applications. One of the challenging aspects of natural condition is the large variation of illumination intensity. Illumination condition in the field continually changes, depending on the sunlight intensity, position, and moving clouds. This change affects RGB pixel values of acquired image and leads to inconsistent colour appearance of plant. Within this condition, plant segmentation based on RGB indices mostly produces poor threshold result. Besides, when shadows are presented in the scene, which is not uncommon in the field, plant segmentation becomes even more challenging. Excessive green (ExG) and other RGB indices have been widely used for plant image segmentation. Although ExG based segmentation is generally accepted as one of the most common and effective methods, it often provides poor segmentation results especially when the image scene contains an extreme illumination difference caused by dark shadows. To build an automated mobile weed control system, within the framework of the SmartBot project with the focus on the detection and control of volunteer potatoes in sugar beet, the vision-based system should first be able to detect plants out from the soil background even under dark shadow region. The objective of this research was to evaluate the segmentation robustness of illuminationinvariant transformation in comparison with ExG method under natural illumination conditions. Using illumination-invariant transformation, global and local thresholds (Otsu with reconstruction) were assessed to segment plant images. The ground shadow detection process was implemented to remove ground shadow region and background. Global threshold outperformed ExG, and local threshold could effectively remove the soil background region. Even under extreme illumination difference in a scene including sharp dark shadows due to bright sunshine, the illumination-invariant transformation produced robust segmentation results.

Verslag van het zevende Fieldrobotevenement, dat op 6 en 7 juli in Wageningen werd gehouden. Ieder team wordt afzonderlijk besproken, met daarbij de resultaten, sterke en zwakke punten.

De glastuinbouw heeft te maken met grote veranderingen die het noodzakelijk maken om na te denken over aanpassingen aan de gebruikte strategieën voor beheersing van fysiologische afwijkingen en ziekten en plagen. Het overschakelen op marktgedreven grootschalige productie vergt een strakke sturing, gebaseerd op een continu inzicht in de gewasontwikkeling, afgezet tegen een strak teeltplan. Het terugdringen van het gebruik van fossiele brandstoffen en het zoveel mogelijk binnen de kas houden van gedoseerde CO2 dwingen tot een beter inzicht in fysiologische (deel)processen onder invloed van het microklimaat rondom de plantenonderdelen

Bijdrage aan de KNPV-voorjaarsvergadering

This review focuses on the monitoring of crop health status at greenhouse scale, based on the measurement of volatile organic compounds (VOCs) emitted from the plants. The review includes the most important factors that affect the emission of these VOCs from greenhouse crops. Since both, stress factors as well as non-stress factors have an effect on the emission, they are covered separately. The review provides an overview of processes that affect the gas balance of plant VOCs in the greenhouse including the loss processes. These processes are considered as important since they contribute to the time-dynamic concentration profiles of plant-emitted VOCs. In addition, we review the most popular techniques currently in use to measure volatiles emitted from plants with emphasis on greenhouse application. Dynamic sampling in combination with gas chromatography coupled to mass spectrometry is considered as most appropriate method for application at greenhouse scale. It is recommended to evaluate the state-of-the-art in the fields concerned with this method and to explore the development of a new instrument based on the specific needs for application in greenhouse practice. However, to apply such an instrument at greenhouse-scale remains a challenge, mainly due to the high costs associated with it.

In this paper a small agricultural robot named Cropscout II is described. Besides the objective to participate in the annual Field Robot Event competition Cropscout II operates as a modular test bed for autonomous robot control using sensor fusion techniques and artificial intelligence. The main challenge in this aspect is to cope with the poorly structured environment and the variation in shape, size and color of biological objects encountered in the open field. The very flexible and modular design of the system in both the electrical and mechanical way proofed to have many advantages. Unless some of the tasks to complete were solved very well the final conclusion is that it is still a big challenge to build a robot for the wide variety of different and unpredictable outdoor conditions. Future research on all aspects is essential

Autonomous robots for agricultural practices will become reality soon. These mobile robots could take over regular tasks such as scouting for weeds and diseases, plant specific applications, yield and field mapping and for instance the release of info-chemicals for attracting predators of pests. This paper presents WURking, a small sized sub-canopy autonomous robot that can be used for a wide range of tasks on the farm of the future. WURking was designed for navigating within row crops like corn. It consists of a mobile platform with three independently driven steerable wheels. The robot carries ultrasound sensors and a gyroscope used for navigation between the crop rows, end of row detection and headland turning. A camera is mounted to detect objects like weeds. High level control of this robot was implemented using the visual programming language LabView. A data fusion technique is used to extract the position and orientation of the robot relative to the crop rows, from the redundant set of sensor data. Feedback linearization of the non-linear system dynamics yielded a simple linear controller structure which was fed by state estimates generated by a Kalman filter using the raw ultrasound sensor data. The headland turning was based on a proportional controller using data from the gyroscope. This paper contains a description of the robot as well as results of a performance test performed in 2008. This test revealed that the robot was able to navigate through the corn field with a maximum offset of ±10 cm from the centre of the rows and an maximum orientation error of ±0.15 rad.

In 2003 beleefde het Field Robot Event (FRE) in Wageningen zijn primeur. De FRE is een internationale wedstrijd, waarin field robots met elkaar de competitie aangaan om zo goed mogelijk autonoom door een perceel te navigeren en een landbouwkundige taak uit te voeren. Dit jaar wordt het Field Robot Event voor de 7e keer gehouden in een bijzondere editie: de samenwerkende Europese onderzoeksgroepen binnen het EU FutureFarm project demonstreren hun ‘prototype robotic systems’ en parallel aan het FRE wordt de Joint International Agricultural Congress gehouden. Nu de 7e editie in aantocht is, stellen we ons in dit artikel de vraag in hoeverre de afgelopen 6 edities hebben bijgedragen aan het realiseren van deze doelstellingen

Zeer uitgebreid verslag van het 4e Fieldrobotevent, dat gehouden werd op 23 en 24 juni 2006 in Stuttgart/Hohenheim

Precision Agriculture is becoming ever more relevant as the agricultural industry struggles to come to terms with the environment, economics, traceability, vehicle guidance and crop management. Whilst some benefits have proved elusive, others contribute positively to today's agriculture. Research continues to be necessary and needs to be reported and disseminated to a wide audience.These proceedings contain the reviewed papers from the 7th European Conference on Precision Agriculture. The papers reflect the wide range of disciplines that impinge upon precision agriculture including remote sensing, plant disease and weed detection, yield monitoring, soil sensing, geo statistics and path planning, regional and crop modelling, cooperation and guidance of robots, precision application, ICT in precision agriculture, future farming and European relevance for precision agriculture.The broad range of research topics reported is a valuable resource for researchers, advisors, teachers and professionals in agriculture.Also note that the reviewed papers from the 4th European Conference on Precision Livestock Farming are presented in a companion publication

Botrytis blight caused by the fungus Botrytis cinerea is probably the most common disease of greenhouse-grown crops like tomato. Botrytis blight in tomato plants is mainly detected by visual inspection or destructive biochemical and molecular determinations. These methods are time consuming and not suitable for large sample sizes. In contrast we propose a non-destructive detection method for plant diagnosis using volatiles as early indicator of plant diseases. This paper presents the changes in volatile production during an infection of tomato plants by B. cinerea. Volatile emission from tomato plants before and after inoculation with B. cinerea were analyzed using on-line gas chromatography coupled to mass spectrometry. The emission was monitored from 2 to 72 hours after inoculation with a time resolution of 1 hour. The data was subjected to principal component analysis for fast interpretation of the variation between samples. Results show that infected plants produce different number and amounts of volatile metabolites compared to control plants. The emission did not only depend on the stage of infection, but also on the diurnal rhythm of the plants. Further research is required to determine the uniqueness of the volatile pattern of B. cinerea infected tomato plants. If so, then, disease specific compounds could be used as early indicator for B. cinerea detection in greenhouse crop production

Sweet pepper production is characterized by large fluctuations in fruit yield per week. Synchronization of yield patterns between nurseries in response to weather conditions leads to variations in market supply and affects price formation. In order to improve supply chain efficiency, auctions and wholesalers require nurseries to supply production forecasts several weeks in advance. At the nursery level, yield prediction is helpful for efficient labour planning. Because the dynamics of sweet pepper production are complex, growers find it difficult to understand and effectively control the production process. A model can be useful to assist growers in making efficient decisions with regard to crop and climate management. It is hypothesized that a phase-shift in the feedback-regulation of fruit set by growing fruits could be the cause of the periodic yield pattern observed in practice. This paper describes a dynamic model of sweet pepper production based on this principle. Model output was compared with measurements of sweet pepper production obtained from a commercial nursery. The calibrated model was able to simulate the measured data fairly well, in particular the yield fluctuations. These results are consistent with the hypothesis and serve as a promising starting point for further research aiming at a better understanding of the underlying processes and, consequently, model improvement.

Using an unknown input observer, with an output linearising feedback, an on-line estimator for the ventilation flow or rate is developed for natural ventilated greenhouses. The paper shows how to design, implement and tune such an estimator in practice and shows that it performs well. Furthermore the practical value for the grower is discussed.