Image detection

The project, as reported by Business Standard, is being developed in India and the technology is based on machine learning that produces images to detect defects such as cuts, cracks or pressure damage . As Amazon India representatives explained, the machine is capable of running million evaluations per day , at a much lower cost than any other method.

To privilege quality

"Quality is one of the key elements of fruit and vegetable purchasing decisions and a critical factor in achieving customer satisfaction," Rajeev Rastogi, Amazon India's vice president of machine learning, said at the company's Smbhav event.

And he went on to point out that it is unthinkable for humans to assess the quality of fruits and vegetables by manually examining each individual product. Modern fruit and vegetables, in fact, have their spearhead in selection technologies on the innovation front.

Automation

By taking advantage of the hardware y machine learning , Amazon plans to develop a conveyor-based automatic sorting and packaging machine capable of packaging products in predetermined quality grades. It is estimated that the packaged with this machine will reduce sorting costs by 78% compared to manual sorting which it is still adopted in India.

Flavor and firmness

Amazon also plans to use infrared sensors to detect parameters such as sweetness and ripeness. These cannot be detected in RGB (red, green, blue) images captured by traditional computer vision algorithms and require destructive methods. Amazon thus faces the challenge of the classification of the internal quality of the fruits that, in our sector, has already been assumed by avant-garde companies such as Unitec .