Zimpapers Logo

Sifelani Tsiko Agric, Environment & Innovations Editor
Ugandan developers have come up with an agriculture app that can easily help millions of smallholder farmers in Zimbabwe and across the continent spot the fall armyworm — a voracious pest that entered the continent from the Americas in 2016.

Fall armyworm has been destroying maize crops and posing a major threat to food security and agricultural trade.

According to a Ventures Beat report, Hansu Mobile and Intelligent Innovations app developers first collected data samples from nearby fields using their smartphones to capture images.

“They categorised these before tapping Google’s TensorFlow to retrain an AI model with transfer learning, a technique where a model suited to one task is re-purposed on a second, related task using TensorFlow Converter, a tool that takes a TensorFlow model and generates a lightweight version.

“They integrated a trained image classifier into the aforementioned FlatButter app,” the publication reported.

At present, the app developers have taken nearly 4 000 data samples as the exercise continues to snap images of affected maize.

“In any case, the interface is locked in tight: Users focus their smartphone camera on a crop and capture an image, after which the app’s model analyses the plants in view of damage and suggests possible solutions,” the publication said.

Ugandan app developers were now aiming to cover coffee and cassava diseases as they shift to cloud services and Google’s Firebase development platform.

“Our plan is to collect more data and to scale the solution to handle more pests and diseases,” said Hansu Mobile and Intelligent Innovations CEO Nsubuga Hassan.

“With improved hardware and greater localised understanding, there’s huge scope for machine learning to make a difference in the fight against hunger.”

Worldwide, there is a growing trend towards the adoption of new technological apps to detect crop diseases and other agricultural problems.

In recent months, Microsoft highlighted a Santa Cruz-based start-up — Conservation Metrics — that’s leveraging machine learning to track African savanna elephants.

A team of researchers are said to have developed a machine learning algorithm trained on Snapshot Serengeti that can identify, describe, and count wildlife with 96,6 percent accuracy.

Zimbabwean farmers now have access to a number of apps that are now helping them right from accessing inputs, early warning alerts, crop mapping and planning to market access. Agricultural apps have also helped to link farmers to genuine suppliers of pesticides and other vital inputs on the market.

Other apps have also helped to connect farmers to suppliers, buyers and traders.

Spotting the fall armyworm has been problematic for the majority of smallholder farmers in Zimbabwe and the entire African continent since it is a recent invader, unlike the true armyworm introduced from the Americas decades ago.

The latest Ugandan-developed app can help farmers easily know whether they have fall armyworm in their fields or whether they should look for other potential pests.

“Farmers only need to take pictures of the pests in their gardens and send to the app.

“They immediately get feedback on whether it is fall armyworm or not. If it is fall armyworm, you are guided on the pesticides and genuine suppliers who can also be accessed through the same application,” a Ugandan developer was quoted saying.

“The app can be downloaded on your smartphone from the Google Play Store for free.”

The United Nation’s Food and Agriculture Organisation says outbreaks of the fall armyworm in Zimbabwe and other SADC countries has affected food security in a region ravaged by a severe drought last season.

The fall armyworm (Spodoptera frugiperda) pest was first reported in Africa in 2016. Native to the Americas, the fall armyworm can feed on 80 different crop species, including maize, Africa’s major staple food consumed by more than 300 million on the continent. The voracious pest has since spread to more than 30 African countries, including Zimbabwe and most other countries in the SADC region, posing a significant threat to food security, income, and livelihoods.

“Within the SADC region, only Mauritius and Lesotho have not been affected.

The first report of the fall armyworm in Zimbabwe was in 2016. First reports were from Bubi District in Matabeleland North.

In the 2018-2019 cropping season, the pest affected maize in all the country’s 10 provinces, affecting hundreds of hectares of maize and pastures.

Plant experts say in order to combat the spread of the fall armyworm, the Government requires to strengthen, monitoring and surveillance, including eliciting the support of Community-Based Armyworm Forecasting (CBAF) units in various parts of the country.

They say early detection is key to sound management of the pest.

“The sudden appearance of fall armyworm makes it overwhelming for crop protection authorities that need to react urgently during the short time that is available for control operations,” an FAO crop expert said.

“The fall armyworm needs to be controlled during the early days of its life to reduce the rate of recurrence.

“The fall armyworm in particular, represents major concern as it is reported that it can have 10 to 12 cycles and hence could continue recurring following the first spray.”