This paper analyzes the possibility of applying data fusion combined with artificial neural networks (ANN) on a dataset combining hard and soft data for prediction of one of the most devastating crop diseases of winter wheat, i.e., Septoria Tritici (Zymoseptoria tritici). In advanced decision support systems for crop protection choices, disease models form a major component. They reproduce the biophysical processes of disease development and temporal spread as a set of rules or processes to predict disease risk value. However, the adaptation of these rules or processes to incorporate the effects of climate change is complex and requires extensive rework. To remedy this issue, statistical machine learning techniques have been introduced to model disease severity percentage for some diseases. However, the use of artificial neural networks has been limited (mainly to image data) and is unexplored for Septoria Tritici. This paper explores the use of Feed Forward neural networks on fused tabular data for the task of disease severity modelling. First, ten years of trial data ranging from 2008 to 2018 across Europe is used for the creation of the new tabular dataset with a fusion of all important data sources baring impact on disease development: Field-specific data, weather data, crop growth stages, and disease severity observation made by human trial operators (response variable). Next, two implementation architectures of Feed Forward neural networks on tabular data are employed: a) standard architecture with backpropagation, drop out regularization, and batch normalization and b) advanced architecture with improvements such as cyclic learning rate and cosine annealing. The advanced architecture is able to better model the data and make estimations of disease severity with a difference of +-10% giving a better quantifiable estimate of disease stress. For better outreach to farmers, a technique to incorporate such modelling techniques into the well established Decision Support Systems is also presented.

This paper introduces digital farming solutions offered by xarvio TM and how these solutions contribute towards achieving the United Nations Sustainable Development Goals. By leveraging recent advancements in Artificial Intelligence, farmers can apply crop protection more efficiently by targeted usage. Respective modules presented in this paper, namely Spray Timer, Zone Spray, Buffer Zones and Product Recommendation ensure crop protection products are applied at the right time and only where they are needed while also ensuring the right product at the optimal rate. This not only reduces the impact on the environment, but moreover increases the productivity and profitability of the farmer. The impact of our digital solutions is exemplified by real world case studies in two major food production regions: Europe and Brazil. In Europe the use of Artificial Intelligence driven spray timing, variable rate application maps and product recommendation have led to a 30% decrease in fungicide usage on field trial cereal crops and a 72% decrease in tank leftovers reducing environmental pollution. In Brazil the Zone Spray weed maps solution created using Computer Vision techniques resulted in a 61% average savings, cutting back on almost two thirds of herbicide and water consumption. As a result the solutions presented in this paper cater to the UN Sustainable Development Goals of zero hunger and responsible consumption and production.

Early warning systems help combat crop diseases and enable sustainable plant protection by optimizing the use of resources. The application of remote sensing to detect plant diseases like wheat stripe rust, commonly known as yellow rust, is based on the presumption that the presence of a disease has a direct link with the photosynthesis capability and physical structure of a plant at both canopy and tissue level. This causes changes to the solar radiation absorption capability and thus alters the reflectance spectrum. In comparison to existing methods and technologies, remote sensing offers access to near real-time information at both the field and the regional scale to build robust disease models. This study shows the capability of multispectral images along with weather, in situ and phenology data to detect the onset of yellow rust disease. Crop details and disease observation data from field trials across the globe spanning four years (2015–2018) are combined with weather data to model disease severity over time as a value between 0 and 1 with 0 being no disease and 1 being the highest infestation level. Various tree-based ensemble algorithms like CatBoost, Random Forest and XGBoost were experimented with. The XGBoost model performs best with a mean absolute error of 0.1568 and a root mean square error of 0.2081 between the measured disease severity and the predicted disease severity. Being a fast-spreading disease and having caused epidemics in the past, it is important to detect yellow rust disease early so farmers can be warned in advance and favorable management practices can be implemented. Vegetation indices like NDVI, NDRE and NDWI from remote-sensing images were used as auxiliary features along with disease severity predictions over time derived by combining weather, in situ and phenology data. A rule-based approach is presented that uses a combination of both model output and changes in vegetation indices to predict an early disease progression window. Analysis on test trials shows that in 80% of the cases, the predicted progression window was ahead of the first disease observation on the field, offering an opportunity to take timely action that could save yield.

This paper introduces the idea of using social media streams like Twitter to identify occurrences of crop diseases. Climate change and changes in agriculture practices have contributed to a change in crop disease dynamics leading to an increase in crop damages. Monitoring crop disease occurrences across regions is helpful for farmers to prepare for such adverse situations and make effective use of crop protection products thus ensuring enough produce for the growing population and protection of the environment. We investigate Machine Learning and Natural Language Processing techniques in order to spot agricultural discussions on Twitter; then analyze, categorize, and group them; so they can be used by a stakeholder to identify crop disease incidences, patterns, and trends at the regional scale. Current systems using keyword based search of agricultural diseases do not always yield agriculturally relevant tweets and those that do could talk on a range of sub-topics. Therefore, text classification forms the core component of this work. A two fold classification process is employed, classifying agriculturally relevant tweets from the rest and then performing fine-grained categorization on them. The resulting model for agricultural tweets classification performs with 93% accuracy and the fine grained categorization model that categorizes tweets into 6 categories gives 75% accuracy. A prototype of an interactive web based disease monitoring application is also presented. The location estimation is not always accurate but nonetheless, this work acts as a proof of concept for the introduction of social media as a novel data source in precision farming.

Government has taken up Integrated Watershed Development Program (IWMP now PMKSY Watershed) to restore the ecological balance by harnessing, conserving and developing degraded natural resources such as soil, vegetative cover and water in degraded rain fed areas of the country. In Tarlakota Project of Palasa Mandal, Srikakulam District, Andhra Pradesh, the watershed program was started during 2012 and completed in 2018). Several NRM and PSI activities were taken up over the five years period (2012 to 2018) in the project area. The present study was taken up to assess the impact of watershed activities taken up in Tarlakota Project on LULC and NDVI using remote sensing and GIS techniques (2011-2012 to 2018-2019). IRS Resourcessat-2 LISS-IV satellite imageries data of 2012 (Pre 11th October, 2012) and 2018 (Post 4th October, 2018) covering the watershed were used to assess the changes in land use/land cover and NDVI over a period of five years. The images were classified into different land use/land cover categories using supervised classification by maximum likelihood algorithm. They were also classified into different vegetation levels using Normalized Difference Vegetation Index (NDVI) approach. Significant changes were observed in LULC over the five years period due to project implementation. The area under crop land and plantation were increased by 322.93 ha (13.80%) and 173.32 ha (19%) respectively during the project period. This was due to conversion of waste lands and fallow lands into crop lands which may be attributed to better utilization of surface and ground water resources created, adoption of soil and water conservation practices and capacity building of the watershed community. The area under current fallow decreased by 202.20 ha (56.18%) and waste lands by 368.59 ha (65.43%). Substantial increase in the area under dense vegetation 270.95 ha (21.57%) and open vegetation 153.10 ha (8.70%) was observed. The water body area also increased from 128.83 ha to 157.72 ha which might be due to rain water conservation activities taken up in the project area.