Banana is the most famous fruit in the world

Banana is the most famous fruit in the world: every year we eat 100 billion of it. Yet their future is threatened by a multitude of illnesses that devastate plants around the world.

 Researchers have now created a tool to tackle these silent murderers: an artificially intelligent smartphone app that can scan banana plants for early signs of infection and alert farmers to their crops before it takes hold.

Researchers discovered in field research in India, China, the Democratic Republic of Congo, Benin, Colombia and Uganda that the new tool — which uses the phone's camera to screen crops — is at least 90% precise in defining the six most severe illnesses and pests that plague banana crops.

These include two globally destructive diseases known as Black Sigatoka and Fusarium Wilt that together decimated vast stretches of banana plantations around the world.

This invention — produced by scientists from worldwide organizations including Biodiversity International and the International Center for Tropical Agriculture — is motivated by the harsh truth of the risks of wiping out the bananas we know and enjoy worldwide, by the increasing threat of illness.

This particularly impacts the Cavendish, the most commonly consumed banana variety in the world, and the sort that most individuals in the west typically purchase in grocery stores.

Our love for this specific variety has led to the establishment of huge monoculture plantations in banana-growing nations, which has caused the demise of the fruit: without genetic variety on these farms, they are left exposed to disease ravages, which can quickly determine vast cropland tracts in one go.

One of the most voracious of these threats, Fusarium Wilt, has already impacted plants in Asia and Africa. Even Latin America, home to Ecuador, the world's biggest exporter of Cavendish bananas, has recently reached the disease.

The smartphone app is based on a computer model that makes its predictions using algorithmic deep learning. The scientists collected 18,000 pictures of banana crops from farms around the globe to generate this, many showing different indications of bad health.

They then used the gallery to train multiple computer models in these pictures to define the hallmarks of specific pests or disease. According to the scientists, the app's main advantage is its flexibility.

It can recognize illness indications on any portion of the plant, and can correctly recognize illness even in photographs of low quality, or in pictures where there is a lot of background noise — such as leaf debris covering the floor. In some instances, the illness detection rate was as high as 100%.

In addition to defending the world's favourite fruit, the app is intended to protect smallholders ' livelihoods in particular. These farmers typically depend on tiny parcels of land to sustain themselves and their families, places where a disease outbreak could ruin a lifetime. The app is specifically intended to prevent this catastrophe by equipping farmers with a DIY, handheld instrument to check plants frequently and quickly recognize the risk of disease.

For the next phase of growth, this is precisely what the scientists have in mind. Now they hope to use the app to create an interconnected, worldwide system, fuelled by farmers who share their crop data. This would allow specialists to identify where the outbreaks of disease start and to figure out how to contain them before spreading out of control.

The scientists decided to call the ' Tumaini ' app, which in Swahili implies ' hope. ' They believe this is what the instrument provides to the most susceptible banana farmers on the planet. "Not just an app, this is a instrument that adds to an early warning system that directly promotes farmers," they claim.

DxMinds : Mobile App Development Company in UK provides AI Solution for farmer in detecting the crop. For free quote: mail us at : info@dxminds.com or call at 7483546629  .