By Sonia Semedo, University of Cape Verde
The emergence of pests and diseases has destroyed agricultural crops in Cape Verde and often farmers do not realise this until late and eventually lose their production. Our research proposal is to monitor the growth of crops and with the help of artificial intelligence detect and identify potential pests that are affecting the crops and alert farmers in time to save the production.
At this time, we are not aware of a solution that would identify Cape Verdean pests. So, we believe that with the creation of a database of pests that affect crops and their evolution in Cape Verde we can use artificial intelligence to identify early the potential pest that is affecting the production. This early detection may enable farmers to implement containment and extermination measures.
Agriculture in Cape Verde is strongly dominated by maize and beans in rain-fed agriculture and tomatoes, tubers, and cabbages in irrigated agriculture. In this work, we will begin with detecting two pests, nezara viridula and spodoptera frugiperda. We have chosen those two species because they affect both irrigated and non-irrigated crops. Based on preliminary studies, we were able to identify the most affected crops: tomatoes, maize, potatoes and cabbages.
The start of this project was only possible thanks to an AI4D Africa scholarship. This funding allowed us to involve two research assistants, an undergraduate student and a PhD student, thus creating a small team to study pests in the University of Cabo Verde and we are now starting to attract colleagues from different background to this theme.
Use AI for agricultural pest detection in Cape Verde has the potential to greatly benefit the country’s agriculture sector. Cape Verde is an archipelago located off the coast of West Africa with a largely arid climate. Agriculture is an important sector of the economy, accounting for about 12% of GDP, but is vulnerable to pests and diseases that can devastate crops and threaten food security
AI-based pest detection systems can help farmers identify and respond to pest infestations quickly and effectively. These systems can use various technologies such as computer vision, machine learning, and remote sensing to monitor crops and detect signs of pest damage. For example, computer vision algorithms can analyze images of crops to identify patterns of discoloration or other signs of stress that may indicate pest infestations. Machine learning algorithms can then use these patterns to develop predictive models that can alert farmers to potential pest outbreaks before they occur
Remote sensing technologies such as satellite imagery can also be used to monitor crop health and identify areas of stress or damage caused by pests. These technologies can be particularly useful in Cape Verde, where the scattered nature of agricultural lands across multiple islands can make it difficult for farmers to monitor their crops regularly.
Implementing AI-based pest detection systems in Cape Verde would require investment in infrastructure, technology, and training. However, the potential benefits in terms of increased crop yields and reduced losses from pest damage could be significant, and could help improve food security and boost economic growth in the country.