Dr. Trilochan Mohapatra
Secretary (DARE) & Director General (ICAR)
Department of Agricultural Research & Education (DARE)
Indian Council for Agricultural Research (ICAR)
Ministry of Agriculture and Farmers Welfare, New Delhi
Sustainability and profitability of Indian agriculture : Challenges and opportunities
Prof. HONDA Kiyoshi
Dept. of Astronautics and Aeronautics,
Innovative Energy Science and Engineering
Chubu University, Japan
IoTs, Crop modeling, Data standards and interoperability solutions in Precision Agriculture
Progressing IoT is acquiring and accumulating massive amount of agro-environmental data. However, raw observation data is often not enough for decision making. Integrating data from heterogeneous sensor systems is still a challenge. We need to form a data pipeline in order to translate the observed data into actionable information. Data standard secures the interoperability for data integration. Analytics that are provided by API makes it possible to develop decision support systems (DSS) with much less effort than ever.
Prof. Dr. Michael Clasen
Department for Business and Computer Science,
University of Applied Sciences and Arts Hanover, Germany
The future role of digital marketplaces in precision agriculture
- Start with a historical overview of digital marketplaces in agriculture
- Analyse, how the digital marketplaces have reduced transaction costs in the past, and how they could reduce them in future
- Show new developments in agriculture like web-based farm management systems and self-steering agriculture (farming 4.0). Thinking about the consequences for digital marketplaces.
Prof. Seishi NINOMIYA
International Field Phenomics Research Laboratory,
Institute for Sustainable Agro-ecosystem Services
University of Tokyo, Japan
How should ICT contribute to agriculture?
Information and communication technology (ICT) such as the Internet and smart phones has been dynamically changing our life during the last decade. It has been also revolutionizing industries, sometimes creating totally new business model such as share economy. However, agriculture is not really sharing its benefit yet. Compared with other industries, agriculture has two specific features; uncertainty and site-specificity. Uncertainty is mainly caused by unpredictable environmental conditions while site-specificity is caused by varying local conditions of farming such as weather, soil, crops, cultivars and cultivation methods. These features require agriculture to be dynamically adjusted and customized, depending on time and locations.
We used to believe that efficiency of production can be achievable only by large scale mass production and that customization is costly. But, we know several examples that such customization was realized efficiently by using ICT. The concept of Industry 4.0 proposed by Germany, maximally utilizes IoT, bigdata and AI for very efficient customized productions. Considering above-mentioned features of agriculture, I expect that ICT should contribute to improving efficiency and sustainability of agricultural productions in many ways.
This presentation discusses about the current and future contributions of ICT to agriculture, showing practical examples regarding decision support, farm automation and knowledge transfer in agriculture.
Soumik Sarkar, Ph.D.
Department of Mechanical Engineering
Iowa State University
Deep learning for agricultural analytics
With the advent of efficient and cost-effective sensors as well as high performance computing technologies, the traditional plant-based agriculture is turning into an efficient cyber-physical system. To enable such precision, farmers, researchers and the agriculture industry are collecting vast amount of data at varying spatial and temporal scales for both experimental and production agriculture. To make sense of this ‘Big Data’, advanced machine learning tools are being widely deployed to solve various problems in this domain. Over the past few years, deep learning-based models have outperformed all other state-of-the-art machine-learning techniques for many detection, classification and prediction problems. With the capability of handling very large data sets, these very large models extract hierarchical features at different scales of data without the need of explicit handcrafting. However, the vast potential of deep learning techniques in precision agriculture is far from being realized. This talk will discuss some recent success stories of deep learning for agricultural analytics with a focus on plant stress phenotyping.
Prof. Takaharu KAMEOKA
Laboratory of Agricultural and Food Systems,
Division of Environmental Science and Technology,
Graduate School of Bioresources
Mie University, Japan
Multiband Spectroscopic Sensing for Digital Agriculture in Food System
We have been developing the integrated investigations on multiband optical sensing of metabolites, biological systems, and foodstuffs. For measuring the crop physiology required for digital agriculture, we also applied such sensing techniques to the measurements of plants and agricultural materials at the field. This is because optical sensing is simple to use, non-destructive, simultaneous, and non-chemical. Another reason is that results can be gotten easily and quickly.