CSIR hosts Machine Learning for Earth Observation in Agriculture programme
The aim of the programme was to build capacity within the country with regard to the combined use of Earth observation technologies (for example, satellite and drone imagery) and machine learning and artificial intelligence methods within the agricultural sector. The programme was funded by the German Agency for International Cooperation (GIZ) and supported by the Department of Science and Innovation (DSI) and the South African National Space Agency (SANSA). The 20 competitively selected trainees who took part in the programme included full and part-time students, as well as full-time staff from government and private agencies.
The Council for Scientific and Industrial Research (CSIR), in collaboration with Move Beyond Consulting, hosted the final presentations and awards ceremony for the inaugural ‘Machine Learning for Earth Observation in Agriculture’ training programme, which was held at the CSIR International Convention Centre on 4 July 2022.
The aim of the programme was to build capacity within the country with regard to the combined use of Earth observation technologies (for example, satellite and drone imagery) and machine learning and artificial intelligence methods within the agricultural sector. The programme was funded by the German Agency for International Cooperation (GIZ) and supported by the Department of Science and Innovation (DSI) and the South African National Space Agency (SANSA). The 20 competitively selected trainees who took part in the programme included full and part-time students, as well as full-time staff from government and private agencies.
“The digitalisation of agriculture has so much potential, but we cannot tap into that potential because of the lack of capacity in the country. Through the collaborative agreement between the GIZ, the DSI and SANSA, we hope to begin to address this through training programmes such as this,” said Prof. Moses Azong Cho, chief researcher in the Advanced Agriculture and Food cluster at the CSIR. Prof. Cho added: “We hope to capacitate participants in the use of these technologies (Earth observation, machine learning and artificial intelligence), such that they may go out and create actionable data that assist farmers to improve their productivity and begin to address broader food security questions in the country and the Southern African Development Community region.”
In addition to staff from the University of the Witwatersrand and the Agricultural Research Council, the CSIR precision agriculture group had four staff members act as expert facilitators who planned, developed and presented training materials over the course of four months. The facilitators also conducted online training, answered participants’ queries, guided participants in their final project assignment, marked participants’ assignments and supervised CSIR equipment for in-field practical exercises. Dr Russell Main, senior researcher with the group, stated: “The training programme aligned well with other (medium to long-term) programmes within the research group, and we hope that it will act as a type of pipeline of talent and skills that could return to the CSIR and/or collaborate with the CSIR in the future.” Dr Main also added: “Through the development of precision agriculture-related information systems, our group also has intensive stakeholder engagement and training exercises planned. So, this training programme provided the appropriate springboard for those.”
Official accreditation for the training programme is being sought, while there is mutual agreement between the organising institutes (GIZ, DSI, SANSA) that the training programme is important enough to become an annual one.