Browsing by Subject "Sensor"

Now showing 1 - 10 of 10

A study of integrated weed control strategies for establishing soybean (Glycine max L. MERR.) in the German production system
(2017) Weber, Jonas Felix; Gerhards, Roland
Soybean (Glycine max L. MERR.) has expanded to become one of the most traded agriculture products worldwide in recent decades. Europe is one of the primary importing regions; however, the dependence on soybean imports has been critically assessed by the public. To reduce the dependency on soybean imports, increased local soybean production should be favoured. In addition to environmental conditions, weeds are a major limiting factor for soybean yield under German climate conditions. Weeds can be successfully controlled with herbicides, although crop injury frequently occurs after application. Sensor-based screening would be helpful for a rapid evaluation of cultivar tolerance to herbicide application. Alternatively, mechanical weed control strategies can be applied. Since soybean production is currently introduced to the regional crop production, weed control efficiency of conventional mechanical tools (e.g., hoeing and harrowing) have to be evaluated. By using automatic guiding systems intra-row elements could be utilised to increase the weed control efficiency of mechanical hoeing. Other than that, agronomical practices such as the tillage system or cover crops influences the occurrence of weeds. The most common practise worldwide for soybean cultivation is the no-tillage system, which has not yet been investigated under local conditions. Therefore, different weed control strategies in soybean production were investigated according to the following major objectives of this thesis: - Detection of crop injury by herbicides using a chlorophyll fluorescence imaging sensor for different soybean cultivars. - Evaluation of the conventional mechanical strategies of hoeing and harrowing in soybean. - Examination of the weed control efficiency in inter- and intra-row areas using RTK-GNSS precision steering and an optical camera guiding system for mechanical weed control in soybean. - Evaluation of the efficiency of ‘tillage’, ‘reduced tillage’ and ‘no- tillage’ cultivation systems and the influence of cover crops on weed suppression in local soybean production. The Imaging-PAM-sensor based on chlorophyll fluorescence imaging was utilised to investigate the response of different soybean cultivars to the application of herbicides. The measurements indicated significant differences with respect to injury to the cultivars after herbicide application. Herbicides containing the active ingredient ‘metribuzin’ resulted in significant differences in the level of crop injury depending on the cultivar. The active ingredients ‘dimethenamid’ and ‘clomazone’ resulted in less injury, independent of the cultivar. The PAM-sensor was able to detect stress symptoms 3 to 7 days before visual symptoms appeared. An investigation of hoeing and harrowing, which are conventional mechanical techniques for weed control, showed 78% and 72% weed control efficiency, respectively. In further experiments, the results of precision steering systems using RTK-GNSS and an optical camera guiding system additionally equipped with intra-row elements (e.g., finger weeders) were compared with the results of conventional hoeing. Mechanical weed control using automatic steering technology and an intra-row element (finger weeder) reduced the weed density by 89% compared with 68% in the conventional hoeing system. With respect to crop yields, statistical benefits of precision steering were not observed. However, the driving speed could be increased from 4 km h−1 in the conventional hoeing system to 10 km h−1 using the automatic steering systems. In an additional experiment, two cover crops species, rye (Secale cereale L.) and barley (Hordeum vulgare L.), were grown for preventive weed control in soybean production. The cover crops were transformed into a mulch layer using a roller-crimper immediately before soybean was sown using a no-tillage technique. Conventional tillage was performed to compare the systems with respect to their weed control efficiency, crop development and soybean yield. The results showed that the no-tillage system had a greater effect on suppressing summer annual weed species (Chenopodium album (L.), Echinochloa crus-galli (L.) P. Beauv. and Amaranthus retroflexus (L.)) than the tillage systems. Conventional tillage and reduced tillage showed increased suppression of the weed species Matricaria inodora (L.), Stellaria media (L.) Vill. and Sonchus arvensis (L.), which were present in the no-tillage system. However, in the conventional tillage and reduced tillage systems, an additional weed control treatment was necessary to suppress the summer annual weeds and ensure high yields. The cover crop rye resulted in weed control similar to that of barley in the no-tillage system. Despite the low weed density, the no-tillage system with a rolled cover crop showed a yield reduced of 47%, whereas the yield of the reduced tillage system was decrease by 23% compared with the conventional tillage system.
Development of a sensor-based harrowing system using digital image analysis to achieve a uniform weed control selectivity in cereals
(2021) Spaeth, Michael; Gerhards, Roland
Using intelligent sensor technology for site-specific weed control can increase the efficacy of traditional weed control implements. Several scientific studies successfully used intelligent sensors for automatic harrow control by taking many different parameters into account such as weed density, soil resistance factor, and plant growth. However, none of the systems was practically feasible because these factors made the control system too complex and unattractive for farmers. Defining only one parameter (crop soil cover) instead of many provides a new and simple approach which was investigated in this work. The first scientific publication focuses on the development, practical implementation and testing of the automatic harrow control system. Two RGB-cameras were mounted before and after the harrow and constantly monitored crop cover. The CSC was then computed out of these resulting images. The image analysis, decision support system and automatic control of harrowing intensity by hydraulic adjustment of the tine angle were installed on a controller which was mounted on the harrow. Eight field experiments were carried out in spring cereals. Mode of harrowing intensity was changed in four experiments by speed, number of passes and tine angle. Each mode was varied in five intensities. In four experiments, only the intensity of harrowing was changed. Modes of intensity were not significantly different among each other. However, intensity had significant effects on WCE and CSC. Cereal plants recovered well from 10% CSC, and selectivity was in the constant range at 10% CSC. Therefore, 10% CSC was the threshold for the decision algorithm. If the actual CSC was below 10% CSC, intensity was increased. If the actual CSC was higher than 10%, intensity was decreased. The new system was tested in an additional field study. Threshold values for CSC were set at 10%, 30% and 60%. Automatic tine angle adjustment precisely realised the three different CSC values with variations of 1.5% to 3%. The next publication discussed and assessed the site-specific field adaptation of the development in cereals. In 2020, three field experiments were conducted in winter wheat and spring oats to investigate the response of the weed control efficacy and the crop to different harrowing intensities, in southwest Germany. In all experiments, six levels of CSC were tested. Each experiment contained an untreated control and an herbicide treatment as a comparison to the harrowing treatments. The results showed an increase in the WCE with an increasing CSC threshold. Difficult-to-control weed species such as Cirsium arvense (L.) and Galium aparine (L.) were best controlled with a CSC threshold of 70%. With a CSC threshold of 20% it was possible to control up to 98% of Thlaspi arvense (L.) The highest crop biomass, grain yield, and selectivity were achieved with an CSC threshold of 20–25% at all trial locations. With this harrowing intensity, grain yields were higher than in the herbicide control plots and a WCE of 68–98% was achieved. The last scientific article compares pairwise a conventional harrow intensity with automatic sensor-based harrowing intensity. Five field experiments in cereals were conducted at three locations in southwestern Germany in 2019 and 2020 to investigate if camera-based harrowing resulted in a more homogenous CSC and higher WCE, biomass, and crop grain yield than a conventional harrow with a constant intensity across the whole plot. For this purpose, pairwise comparisons of three fixed harrowing intensities (10 °, 40 °, and 70 ° tine angle) and three predefined CSC thresholds (CSC of 10%, 20%, and 60%) were realized in randomized complete block designs. Camera-based adjustment of the intensity resulted in 6-16% less standard deviation variation of CSC compared to fixed settings of tine angle. Crop density, WCE, crop biomass and grain yield were significantly higher for camera-based harrowing than for conventional harrowing. WCE and yields of all automatic adjusted harrowing treatments were equal to the herbicide control plots. In this PhD-thesis, a sensor-based harrow was developed and successfully investigated as an alternative to conventional herbicide application in cereals. A permanent, equal replacement of chemical weed control in arable farming systems can only be achieved using modern, sensor-based mechanical weed control approaches. Therefore, the efficacy of the mechanical weed control method can be improved and increased continuously. It has been shown that the precise adjustment of mechanical weed control methods to site-specific weed conditions allows similar WCE results as an herbicide application without causing yield losses. These findings contribute towards modern plant protection strategies to reduce the herbicide use and to establish the acceptance of technical progress in society.
Effects of weeds on yield and determination of economic thresholds for site-specific weed control using sensor technology
(2014) Keller, Martina; Gerhards, Roland
Weeds can cause high yield losses. Knowledge about the weeds occurring, their distribution within fields and their effects on the crop yield is important to achieve effective weed control. The critical period for weed control (CPWC) and the economic threshold (ET) are important key concepts and management tools in weed control. While the former helps to time weed control in crops of low competitiveness, the latter provides a decision aid to determine whether weed control is necessary. This decision is generally taken at the field level. Weeds have been found to be distributed heterogeneously within fields. Site-specific weed control (SSWC) addresses this sub-field variation by determining weed distribution as input, by taking control decisions in the decision component and by providing control measures as output at high spatial resolution. Sensor systems for automated weed recognition were identified as prerequisite for SSWC since costs for scouting are too high. While experiences with SSWC using sensor data as input are still scarce, studies showed that considerable herbicide savings could be achieved with SSWC. ETs can serve as thresholds for the decision component in SSWC systems. However, the commonly used ETs were suggested decades ago and have not been updated to changing conditions since. The same is the case for the CPWC in maize in Germany. In addition, the approaches to determine the CPWC are usually not based on economic considerations, which are highly relevant to farmers. Thus, the objectives of this thesis are: 1. To test different models and to provide a straightforward approach to integrate economical aspects in the concept of the CPWC for two weed control strategies: Herbicide based (Germany) and hoeing based (Benin); 2. To determine the effect of weeds on yield and to calculate ETs under current conditions which can be used for SSWC; 3. To evaluate the use of bi-spectral cameras and shape-based classification algorithms for weed detection in SSWC; and 4. To determine changes in weed frequencies, herbicide use and yield over the last 20 years in southwestern Germany. Datasets in maize from Germany and Benin served as input for the CPWC analyses. The log-logistic model was found to provide a similar fit as the commonly used models but its parameters are biologically meaningful. For Germany, analyses using a full cost model revealed that farmers should aim at applying herbicides early before the 4-leaf stage of maize. In Benin, where weed control is mainly done by hoeing, analyses showed that one well- timed weeding operation around the 10-leaf stage could already be cost-effective. A second weeding operation at a later stage would assure profit. The precision experimental design (PED) was employed to determine the effect of weeds, soil properties and herbicides on crop yield in three winter wheat trials. In this design, large field trials’ geodata of weed distribution, herbicide application, soil properties and yield are used to model the effects of the former three on yield. Galium aparine, other broadleaved weeds and Alopecurus myosuroides reduced yield by 17.5, 1.2 and 12.4 kg ha-1 plant-1 m2 determined by weed counts. The determined thresholds for SSWC with independently applied herbicides were 4, 48 and 12 plants m-2, respectively. Bi-spectral camera based weed–yield estimates were difficult to interpret showing that this technology still needs to be improved. However, large weed patches were correctly identified. ETs derived of field trials’ data carried out at several sites over 13 years in the framework of the ’Gemeinschaftsversuche Baden-Württemberg’ were 9.2-9.8 and 4.5-8.9 % absolute weed coverage for winter wheat and winter barley and 3.7% to 5.5% relative weed coverage for maize. Overall, the weed frequencies in winter cereals were found to be more stable than the weed frequencies in maize during the observation period. In maize, a frequency increase of thermophilic species was found. Trends of considerable yield increases of 0.16, 0.08 and 0.2 t ha-1 for winter wheat, winter barely and maize, respectively, were estimated if weeds were successfully controlled. In order to evaluate the use of bi-spectral cameras and shapebased classification algorithms for weed detection in SSWC, herbicides were applied site-specifically using weed densities determined by bi-spectral camera technology in a winter wheat and maize field. Threshold values were employed for decision taking. Using this approach herbicide savings between 58 and 83 % could be achieved. Such reductions in herbicide use would meet the demand of society to minimize the release of plant protection products in the environment. Misclassification occurred if weeds overlapped with crop plants and crop leaf tips were frequently misclassified as grass weeds. Improvements in equipment, especially between the interfaces of camera, classification algorithms, decision component and sprayer are advisable for further trials. In conclusion, the derived ETs can be easily implemented in a straightforward SSWC system or can serve as decision aid for farmers in winter wheat and winter barley. Further model testing and adjusting would be necessary. For maize, the use of ETs at the field level is not suggested by this study, however the need for early weed control is clearly demonstrated. Bi-spectral camera technology combined with classification algorithms to detect weeds is promising for research use and for SSWC, but still requires some technical improvements.
Entwicklung, Implementierung und Bewertung eines IT-Systems zur Prozessdokumentation und -unterstützung in der landwirtschaftlichen Nutztierhaltung
(2010) Kuhlmann, Arne; Jungbluth, Thomas
In livestock farming, the use of automation technology is common. Automation technology is able to perform sub-processes, whereby the farmer is supported in his daily work. The data produced by this technology is usually monitored manually. The same applies to the collection of process parameters such as resource consumption and climate data. Therefore overall process monitoring and process documentation require high workload. Caused by structural change and the demand for food safety and traceability, livestock farming needs to introduce information technology. This document is dealing with the topics collection, storage, usage and exchange of data on farms and in their environment using the example of pig fattening. A stable for fattening pigs was used to analyse the conditions, requirements and implementation options for achieving the objectives process documentation and process support. Based on the conclusions drawn, a prototype was developed, that focuses on the full integration of all technical components in the stable using communication and data standards. Besides the presentation and evaluation of the system, concrete benefits for science and practice are presented using selected examples. Furthermore possibilities for improvements regarding the used technologies and standards are pointed out.
Laser backscattering imaging in agriculture
(2023) Wu, Zhangkai; Müller, Joachim
Non-destructive optical sensor technology (NDOST) is an essential part of agriculture. The unique capabilities of laser notably enhanced NDOST. Laser backscattering imaging (LBI) is a technology that captures light patterns scattered by a material to analyze its properties. It is particularly suitable for agriculture due to its affordability and the optical scattering nature of agricultural products. The images generated by LBI are related to the optical parameters of the examined objects. Crucial tasks in LBI include the selection of an appropriate laser, the extraction of image features, and the utilization of a prediction model for analysis. LBI has been employed in numerous scenarios, such as maturity detection and drying monitoring. The main challenges for LBI involve establishing a precise theoretical framework and uncovering new applications within agriculture. This study aims to enhance the foundational knowledge about LBI and explore additional application scenarios. The first study focused on basic research about LBI. Currently, researchers rarely document the cell size of their samples and treat the optical coefficient as a constant within agricultural products, which is questionable. This studys purpose was to use glass filter matrices as controlled models and to evaluate the effects of pore size, different solutes, concentrations, and wavelength. The used porous glass discs had pore diameters ranging from 1 to 160 µm. We applied aqueous solutions of NaCl (0, 1, 2, 3, and 4 mol/L) and NaH₂PO₄ (0, 0.8, 1.6, 2.4, and 3.2 mol/L) to fill the pores. The LBI system incorporated laser modules at three different wavelengths (405, 635, 780 nm). The results illustrated that three of the four examined experimental factors (the pore diameter, the solutes, and their concentrations) have a substantial impact on LBI. However, no clear differences in LBI patterns were observed among the three utilized wavelengths. Consequently, when deploying LBI on fruits, for instance, a thorough consideration of cell sizes at various depths from the fruit surface is required. The second study focused on a possible application scenario of LBI: the sedimentation process of crude sesame oil. Oil sedimentation is a process where gravity is used to remove solid impurities, resulting in a clearer oil. This study examined the sedimentation process in crude sesame oil using LBI. In situ and laboratory experiments were conducted over 30 days, involving an LBI system directly attached to a transparent sedimentation tank with 120L of crude oil. Both the oil properties and sedimentation curve were analyzed along with the LBI images. There was a dramatic drop in oil particle-related properties (at least 87%), a 90% decrease in water content, and minor changes in oxidation-related properties. The sedimentation speed was about −7 mm/h, then became less than −2 mm/h, revealing two stages: diluted and hindered sedimentation. The crude oils surface displayed a distinctive scattering spot and a Tyndall effect within the oil, showing an increasing path length as sedimentation proceeded. The findings offer practical insights for enhancing sedimentation tank and LBI system design. The third study focused on another possible application scenario of LBI: leaf wetness measurement. Leaf wetness plays a pivotal role in managing plant fungi diseases. Existing optical techniques categorize leaf wetness as a binary problem – either wet or dry. In contrast, this research developed a platform capable of semi-automatically measuring droplet deposition on grape leaves using an LBI system. The leaf area, mean intensity per pixel in the red channel, and droplet count using information from the green channel were extracted from the scattering images. The study employed a generalized additive model (GAM) to predict leaf wetness with the extracted features. The prediction of the test dataset achieved an R-squared value of 0.78. The extraction of image features was found to be influenced by factors such as image resolution and leaf orientation. The method introduced in this study offers the potential for precise quantification of leaf wetness with an LBI system. In conclusion, our study highlights the importance of considering cell size in agricultural applications of LBI. Besides, LBI was found useful in monitoring plant oil sedimentation and quantifying leaf wetness. This suggests its potential for scenarios involving state changes in suspensions or colloids and differentiating materials with distinct optical properties. However, using a porous matrix as a model introduces an inherent error. Additionally, advancements are necessary to transition the application studies into practical use. Future LBI development could be facilitated by building a comprehensive database on light interactions with diverse cells and tissues.
Measuring grazing behaviour of dairy cows : validation of sensor technologies and assessing application potential in intensive pasture-based milk production systems
(2019) Werner, Jessica; Schick, Matthias
Grazing is the natural feed intake behaviour of a cow. However, in the last century, intensive confinement systems with silage feeding and concentrate supplementation have replaced many extensive pasture-based milk production systems. Grazed grass is now acknowledged as the cheapest feed available as a consequence of rising machinery, labour and feeding costs. Thus there is a renewed interest in intensive pasture-based milking systems. In addition, policy objectives, societal expectations and environmental concerns have all supported reconsiderations for pasture-based milk production. Novel technology to aid measuring and managing grassland and cow grazing behaviour have the potential to facilitate improved performance. Until recently, sensor technologies for dairy farms were mainly developed for measuring feeding behaviour of housed cows. Adapting and calibrating these technologies to grazing context would therefore further support improved pasture-based dairying. In this thesis, two sensor technologies were validated against visual observation. The RumiWatch noseband sensor (Itin+Hoch, Switzerland) is a high precision technology designed for research applications. It can measure detailed grazing behaviour such as grazing bites, rumination chews, time spent grazing and time spent ruminating. The MooMonitor+ (Dairymaster, Ireland) is the second technology assessed in this thesis. It is a collar based accelerometer and is primarily designed for use on commercial farms. The initial development was for oestrus detection. It can now monitor grazing and rumination times. The results of the studies reported in this thesis revealed that both sensors were highly accurate compared to visual observation. The implementation of sensor technology on commercial dairy farms is still slow. This is especially true on pasture-based dairy systems. The management of grazing cows is thus largely not supported by technology. With increasing herd sizes and skilled labour shortages, sensor technology to support grazing management will likely improve some major dairy farm management challenges. A key factor in pasture-based milk production is the correct grass allocation to maximize the grass utilization per cow. Cow behaviour is indicative of the quantity and quality of feed available as well as animal performance, health and welfare. Thus, the measurement of cow grazing behaviour is an important management indicator. A further study of detailed individual grazing behaviour aimed to identify behavioural indicators of restricted versus sufficient availability of grass. Such objective measurement has potential since currently grass allocation is based on subjective eye measurements and calculations per herd. To identify behavioural indicators, a group of 30 cows in total were allocated a restricted pasture allowance of 60 % of their intake capacity. Their behavioural characteristics were compared to those of 10 cows with pasture allowance of 100 % of their intake capacity. The grazing behaviour and activity of cows was measured using the RumiWatchSystem, consisting of the noseband sensor and pedometer. The results showed that bite frequency was continuously higher for cows with a restricted grass allocation, but also rumination behaviour was affected by the restriction. This study contributes vital information towards developing a decision support tool for automated allocation of grass based on feedback from individual cows rather than herd based measurements. Further research activities should focus on identification of significant changes in grazing behaviour of cows at individual animal and herd level. This would allow implementation of specific thresholds to be used in decision support tools. After developing and validating the decision support tools, the application of automated solutions for grazing management can improve efficiency and productivity of pasture-based milk production systems.
Perception for context awareness of agricultural robots
(2018) Reiser, David; Griepentrog, Hans
Context awareness is one key point for the realisation of robust autonomous systems in unstructured environments like agriculture. Robots need a precise description of their environment so that tasks could be planned and executed correctly. When using a robot system in a controlled, not changing environment, the programmer maybe could model all possible circumstances to get the system reliable. However, the situation gets more complex when the environment and the objects are changing their shape, position or behaviour. Perception for context awareness in agriculture means to detect and classify objects of interest in the environment correctly and react to them. The aim of this cumulative dissertation was to apply different strategies to increase context awareness with perception in mobile robots in agriculture. The objectives of this thesis were to address five aspects of environment perception: (I) test static local sensor communication with a mobile vehicle, (II) detect unstructured objects in a controlled environment, (III) describe the influence of growth stage to algorithm outcomes, (IV) use the gained sensor information to detect single plants and (V) improve the robustness of algorithms under noisy conditions. First, the communication between a static Wireless Sensor Network and a mobile robot was investigated. The wireless sensor nodes were able to send local data from sensors attached to the systems. The sensors were placed in a vineyard and the robot followed automatically the row structure to receive the data. It was possible to localize the single nodes just with the exact robot position and the attenuation model of the received signal strength with triangulation. The precision was 0.6 m and more precise than a provided differential global navigation satellite system signal. The second research area focused on the detection of unstructured objects in point clouds. Therefore, a low-cost sonar sensor was attached to a 3D-frame with millimetre level accuracy to exactly localize the sensor position. With the sensor position and the sensor reading, a 3D point cloud was created. In the workspace, 10 individual plant species were placed. They could be detected automatically with an accuracy of 2.7 cm. An attached valve was able to spray these specific plant positions, which resulted in a liquid saving of 72%, compared to a conventional spraying method, covering the whole crop row area. As plants are dynamic objects, the third objective of describing the plant growth with adequate sensor data, was important to characterise the unstructured agriculture domain. For revering and testing algorithms to the same data, maize rows were planted in a greenhouse. The exact positions of all plants were measured with a total station. Then a robot vehicle was guided through the crop rows and the data of attached sensors were recorded. With the help of the total station, it was possible to track down the vehicle position and to refer all data to the same coordinate frame. The data recording was performed over 7 times over a period of 6 weeks. This created datasets could afterwards be used to assess different algorithms and to test them against different growth changes of the plants. It could be shown that a basic RANSAC line following algorithm could not perform correctly under all growth stages without additional filtering. The fourth paper used this created datasets to search for single plants with a sensor normally used for obstacle avoidance. One tilted laser scanner was used with the exact robot position to create 3D point clouds, where two different methods for single plant detection were applied. Both methods used the spacing to detect single plants. The second method used the fixed plant spacing and row beginning, to resolve the plant positions iteratively. The first method reached detection rates of 73.7% and a root mean square error of 3.6 cm. The iterative second method reached a detection rate of 100% with an accuracy of 2.6 - 3.0 cm. For assessing the robustness of the plant detection, an algorithm was used to detect the plant positions in six different growth stages of the given datasets. A graph-cut based algorithm was used, what improved the results for single plant detection. As the algorithm was not sensitive against overlaying and noisy point clouds, a detection rate of 100% was realised, with an accuracy for the estimated height of the plants with 1.55 cm. The stem position was resolved with an accuracy of 2.05 cm. This thesis showed up different methods of perception for context awareness, which could help to improve the robustness of robots in agriculture. When the objects in the environment are known, it could be possible to react and interact smarter with the environment as it is the case in agricultural robotics. Especially the detection of single plants before the robot reaches them could help to improve the navigation and interaction of agricultural robots.
RumiWatch - Development and assessment of a sensor-based behavior monitoring system for ruminants
(2018) Zehner, Nils; Schick, Matthias
Sustainable and competitive milk production is highly dependent on securing the performance potential, health and fertility of dairy cows. Therefore, farmers can benefit from sensor data of animal monitoring systems to improve health management and work processes in dairy farming. The research during this PhD thesis aimed to contribute to the development and evaluation of a scientifically validated, sensor-based animal monitoring system that comprises a device for measurement of ingestive behavior and a device for measurement of movement behavior in cattle that interact as a system with system-specific software. Further aim of this thesis was to evaluate application potentials for this animal monitoring system by means of calving prediction in dairy cows and measurement of chewing activity in horses. The underlying experimental work was structured into four separate studies. The aim of the first study was to develop and validate a novel scientific monitoring device for automated measurement of rumination and eating behavior in dairy cows. Research works for this study aimed to provide a complete and detailed technical specification of the functionality of this device and to perform a validation under field conditions in stable-fed cows. The objective of the second study was to develop and validate a novel algorithm to monitor lying, standing, and walking behavior based on the output of a triaxial accelerometer collected from loose-housed dairy cows. The third study aimed to use automated measurements of ingestive behavior obtained from the developed sensor device to develop and validate a predictive model for calving in dairy cows. The aim of the fourth study was to investigate the suitability and validity of the developed sensor system for automated measurement of chewing activity in horses. In conclusion, the RumiWatch noseband sensor and pedometer that were developed and validated in the current project represent a suitable measuring instrument for automated recording of ingestive and locomotor behavior in dairy cows. The system-specific software is suitable for research purposes and shows a high performance for classification of extended parameters of rumination, eating, lying, standing, and walking behavior. The achieved validation results indicate that the measuring performance satisfies scientific requirements. Further application potentials were demonstrated by means of automated calving prediction in dairy cows and automated measurement of chewing activity in horses. The development and validation of a predictive model for calving time using measurements of the RumiWatch noseband sensor revealed a high amount of false positive alerts that was prohibitive for application of the model in farming practice. However, the analyses showed that particularly parameters of ruminating behavior have predictive value and should be taken into consideration for future research on calving prediction models. Furthermore, it was successfully demonstrated that it is feasible to apply the RumiWatch noseband sensor to horses. The results of direct observation compared with the automatic measurement showed a very high overall agreement of the observed and automatically measured data and, after minor refinements, this measuring device has the potential to become a valuable and easy-to-use tool for equine research and management.
Spatial combination of sensor data deriving from mobile platforms for precision farming applications
(2019) Zecha, Christoph Walter; Gerhards, Roland
This thesis combines optical sensors on a ground and on an aerial platform for field measurements in wheat, to identify nitrogen (N) levels, estimating biomass (BM) and predicting yield. The Multiplex Research (MP) fluorescence sensor was used for the first time in wheat. The individual objectives were: (i) Evaluation of different available sensors and sensor platforms used in Precision Farming (PF) to quantify the crop nutrition status, (ii) Acquisition of ground and aerial sensor data with two ground spectrometers, an aerial spectrometer and a ground fluorescence sensor, (iii) Development of effective post-processing methods for correction of the sensor data, (iv) Analysis and evaluation of the sensors with regard to the mapping of biomass, yield and nitrogen content in the plant, and (v) Yield simulation as a function of different sensor signals. This thesis contains three papers, published in international peer-reviewed journals. The first publication is a literature review on sensor platforms used in agricultural research. A subdivision of sensors and their applications was done, based on a detailed categorization model. It evaluates strengths and weaknesses, and discusses research results gathered with aerial and ground platforms with different sensors. Also, autonomous robots and swarm technologies suitable for PF tasks were reviewed. The second publication focuses on spectral and fluorescence sensors for BM, yield and N detection. The ground sensors were mounted on the Hohenheim research sensor platform “Sensicle”. A further spectrometer was installed in a fixed-wing Unmanned Aerial Vehicle (UAV). In this study, the sensors of the Sensicle and the UAV were used to determine plant characteristics and yield of three-year field trials at the research station Ihinger Hof, Renningen (Germany), an institution of the University of Hohenheim, Stuttgart (Germany). Winter wheat (Triticum aestivum L.) was sown on three research fields, with different N levels applied to each field. The measurements in the field were geo-referenced and logged with an absolute GPS accuracy of ±2.5 cm. The GPS data of the UAV was corrected based on the pitch and roll position of the UAV at each measurement. In the first step of the data analysis, raw data obtained from the sensors was post-processed and was converted into indices and ratios relating to plant characteristics. The converted ground sensor data were analysed, and the results of the correlations were interpreted related to the dependent variables (DV) BM weight, wheat yield and available N. The results showed significant positive correlations between the DV’s and the Sensicle sensor data. For the third paper, the UAV sensor data was included into the evaluations. The UAV data analysis revealed low significant results for only one field in the year 2011. A multirotor UAV was considered as a more viable aerial platform, that allows for more precision and higher payload. Thereby, the ground sensors showed their strength at a close measuring distance to the plant and a smaller measurement footprint. The results of the two ground spectrometers showed significant positive correlations between yield and the indices from CropSpec, NDVI (Normalised Difference Vegetation Index) and REIP (Red-Edge Inflection Point). Also, FERARI and SFR (Simple Fluorescence Ratio) of the MP fluorescence sensor were chosen for the yield prediction model analysis. With the available N, CropSpec and REIP correlated significantly. The BM weight correlated with REIP even at a very early growing stage (Z 31), and with SAVI (Soil-Adjusted Vegetation Index) at ripening stage (Z 85). REIP, FERARI and SFR showed high correlations to the available N, especially in June and July. The ratios and signals of the MP sensor were highly significant compared to the BM weight above Z 85. Both ground spectrometers are suitable for data comparison and data combination with the active MP fluorescence sensor. Through a combination of fluorescence ratios and spectrometer indices, linear models for the prediction of wheat yield were generated, correlating significantly over the course of the vegetative period for research field Lammwirt (LW) in 2012. The best model for field LW in 2012 was selected for cross-validation with the measurements of the fields Inneres Täle (IT) and Riech (RI) in 2011 and 2012. However, it was not significant. By exchanging only one spectral index with a fluorescence ratio in a similar linear model, it showed significant correlations. This work successfully proves the combination of different sensor ratios and indices for the detection of plant characteristics, offering better and more robust predictions and quantifications of field parameters without employing destructive methods. The MP sensor proved to be universally applicable, showing significant correlations to the investigated characteristics such as BM weight, wheat yield and available N.
Untersuchungen zum Herbizidstress in Zuckerrüben mit drei feldtauglichen optischen Sensoren und Methoden der Bildanalyse
(2014) Roeb, Johannes
Weed management in sugar beets is based on the repeated use of herbicide mixtures after crop emergence. Due to the limited selectivity of active ingredients, herbicide treatments not only control the weeds but will reduce the growth of sugar beets also. Yield losses due to herbicide stress are expected to range between 5-15%. Using optical sensors is a nondestructive method to assess changes in reflection, leaf fluorescence or chlorophyll fluorescence kinetics induced by herbicides. To evaluate the applicability of three optical sensors for assessing herbicide stress and to measure the influence of herbicides and herbicide mixtures on sugar beets, a pot experiment was performed at the University of Hohenheim, Germany. Sugar beets were grown under natural light and temperature conditions and treated with the active ingredients metamitron, phenmedipham, desmedipham, ethofumesate, triflusulfuron-methyl and dimethenamid-P in their commercially available formulation and practical dosage. In total five single herbicides as well as five different herbicide mixtures were applied in the cotyledon stage (EC 10), the two-leaf stage (EC 12) or the four- to six-leaf stage (EC 14/16) of sugar beets. Stress reactions were monitored with three optical sensors: Images of a digital camera (Canon EOS 1000D) were used to determine leaf coverage area, plant shape and leaf color. Measurements were performed about every second day and a growth index had been calculated. A multispectral fluorometer (FORCE-A MULTIPLEX®) was used to detect the blue-green fluorescence, red fluorescence and far-red fluorescence and to calculate fluorescence indices. A portable imaging sensor of chlorophyll fluorescence kinetics (WALZ IMAGING PAM) was used on a daily basis to determine changes in the maximum quantum efficacy (Fv/Fm) induced by herbicide treatments. Four respectively two weeks after the treatment sugar beets were harvested for dry matter analysis. For each of the herbicide treatments and for each of the three application dates five Mitscherlich pots were used for replication. Each pot had about four sugar beet plants. Based on digital imaging it was possible to measure leaf coverage area and determine growth depressions induced by herbicide treatments containing mixtures of the active ingredients phenmedipham, desmedipham and ethofumesate. Herbicide mixtures with more active ingredients increased the stress reaction of sugar beets. Differentiation between untreated plants and sugar beets treated with different herbicides or mixtures was possible a few days after application. Results were correlated with dry matter. Changes in plant shape parameters indicated a delayed development of herbicide treated plants. Higher red content of leaf color was attributed to a relative loss of chlorophyll. Measurements with the FORCE-A MULTIPLEX® fluorometer detected an increase in red and far-red fluorescence but not blue-green fluorescence within 1-2 days after treatments with the aforementioned mixture of active ingredients. About the same trends were found at all application dates. Most fluorescence values were affected by growth stage and leaf area of sugar beets. Thus, although differences between treated and untreated plants were strong, it was not possible to discriminate between stress reactions on different herbicide treatments. Based on the maximum quantum efficacy (Fv/Fm) measured with the WALZ IMAGING-PAM chlorophyll fluorescence sensor previous studies, describing the time course of stress reaction on application of PSII-inhibitors in sugar beets were confirmed. After a strong decrease of Fv/Fm within one day, recovery to the initial value was observed within ten days. Quantification of herbicide stress induced by PSII-inhibitors was possible due to different intensities and durations of the stress reaction. Photochemical stress response to treatments with metamitron or chloridazon was lower than with products containing phenmedipham or desmedipham. Stress reaction on herbicide mixtures not only depended on content of PSII-inhibitors but also on formulation. Weather conditions were more important than the sugar beet development stage considering the stress reaction. Observations from previous studies, indicating an increase in herbicide stress after precipitation and at low temperatures, were also confirmed in this study. Differences in stress reactions of cotyledons and first true leaves can be explained by a higher uptake of herbicides in young tissues. The influence of other herbicides, mixtures, dosages and formulations on herbicide stress in sugar beets has to be further investigated. Moreover the complex interrelations between sugar beet development stage, weather conditions and stress reaction could only be investigated in systematic field trials. For the measurement of stress reactions on herbicides the described optical sensors and methods can be used, each having different advantages and disadvantages.