Improving Food Security & Crop Yield In Malaysia Using Machine Learning
Agriculture is an essential sector in Malaysia’s economy, contributing 12% to the national GDP and providing employment for 16% of the entire population, while industry provides 37.6% of GDP (occupying 36% of the labour force in 2017), and services provide 53.6% (occupying 53% of the labour force in 2017) (CIA 2020). The country has a total land area of 33.03 million ha, of which as of 2015, 23.1% is agricultural land, 63.6% is forest area and 13.3% is for other land uses. Peninsular Malaysia has the largest land area suitable for agriculture accounting for nearly 48% of the total agricultural land area (Olaniyi et al., 2013). The key industrial crops include oil palm, rubber, cocoa and tobacco, which mainly serve the export market. Crops that are referred to as “food crops” primarily serve, though not exclusively, the domestic market, mainly comprise paddy, fisheries, fruits and vegetables. Other miscellaneous crops include sugarcane, cassava, maize and sweet potato, which cater for both export and domestic markets. For a long time, Malaysia’s agricultural policy has mainly revolved around the industrial crops and, to some extent, the food crops (Fatah, 2017).
With Covid19 cases currently rising again in Malaysia, the agriculture industry was classified as critical by the government during the application of the Movement Control Order, allowing businesses to operate as usual. Because around 10% of Malaysia’s labour force is involved in agriculture (Department of Statistics, Malaysia 2019), ensuring food security and protecting the well-being of workers in the sector is critical during the crisis, especially as the average wage of workers in the agriculture, fishing and forestry sector is lowest compared to other sectors, exposing workers to economic hardships.
Labour restrictions risk causing bottlenecks in horticulture, livestock, aquaculture and production systems, as well as for planting and harvesting of crops that are both labour-intensive and seasonally specific. It is of critical importance to designate safe working conditions for the agriculture workforce, to secure future growing seasons and avoid negative impacts on future food security and supply (Abdullah et al. 2020).
In order to maintain and improving food security and crop yield despite labour restrictions in agriculture to prevent pandemic, at a time where the world needs to produce more with fewer resources, AI could help to transform agriculture worldwide and especially in Malaysia. The ability of agricultural equipment to help actors better think, predict, and advise farmers via a variety of AI applications presents Malaysia with the potential to achieve food security in the country. By using satellite imagery which is land cover satellite images from Google Earth Engine to predict crop yield.
The Project Goals
1. Applying Malaysia based open-source satellite imagery dataset to make crop yield prediction.
2. Improving accuracy of crop yield prediction.
3.Create a data visualisation methodology
The Learning Outcomes
1. Collect satellite imagery from Google Earth Engine
2. Apply CV models to identify different crop types
3. Visualise your data using QGSI, pandas & matplotlib
Link to the Original Project: https://omdena.com/projects/foodsecurity-ai/
We will be running an AI project soon…. Stay Tuned!
Malaysia Chapter Lead
Machine Learning Engineer
Manivanan Sehgar, finished his undergraduate in Artificial Intelligence from Multimedia University, Malaysia. He is founder of “thenewbieai” and co-founder “DISH” (startup company in Malaysia) and Machine Learning Engineer at Omdena. He is an AI enthusiasts who believes AI is good to be adapted in daily life because it ease workload sustainability .
“It keep grows as you believe in vision”