Lenta Group tests artificial intelligence in remote acceptance of vegetables and fruits

2025: Artificial Intelligence Testing in Remote Vegetable and Fruit Acceptance

Lenta Group tests machine vision in combination with neural networks and hyperspectral cameras. Solutions allow you to speed up the acceptance of vegetables and fruits, as well as remotely determine their freshness. The company announced this on December 2, 2025.

𝓲

Фото: Tape

In 2021, Lenta Group introduced a project for remote acceptance of vegetables and fruits in distribution centers: the premises were equipped with special tables, additional lighting and high-resolution cameras. The company has created remote decision-making centers, which include agronomists and expert commodity scientists. They accept the product using headsets, large monitors and joysticks to control camcorders. Per month, each worker tracks the quality of more than six tons of fruit and vegetables entering the distribution center.

To speed up remote acceptance, Lenta Group began testing technologies for machine vision combined with neural networks. One of the solutions allows you to achieve more than 90% of the system assessments with the opinion of the inspectors of the monitoring center. This reduces the burden on employees and allows you to multiply the volume of controlled products. The company plans to scale the solution in mid-2026.

The Lenta Group Innovation Center is also studying the possibilities of hyperspectral cameras that recognize the degree of maturity of fruits. They record the spectrum of reflected light for each pixel, which allows you to analyze the parameters of the skin and adjacent tissues: pigments, water, lipids, protein and other chemical and physical characteristics of the surface. Based on the combination of indices, the spectral profile of the fetus is formed, which is then compared with the actual organoleptic estimate. This establishes a relationship between the spectral fingerprint and the real internal state of the fetus. Setting up a hyperspectral camera is required for each species and variety. As of December 2025, the solution is being tested to determine the quality of citrus fruits.

In addition, the company is testing video analytics at production sites. It allows you to identify incidents in real time and monitor compliance with hygiene requirements at supplier sites.

Шаблон:Quote 'author=said Lenta Group Quality Director Yulia Batenina.

2022: Testing of "smart" scales in the hypermarket of St. Petersburg

On March 22, 2022, Lenta announced the testing of smart scales with product recognition technology by appearance. The pilot takes place in the hypermarket of St. Petersburg.

Earlier, the Lenta Innovation Center conducted a technical pilot of a model from another manufacturer in a hypermarket along Savushkina Street.

Russian The production scale is installed in the fresh vegetables and fruits section. After the item is on the scale, variants of the assortment names appear on the screen, from which the buyer selects the correct one and prints the price tag. Plus, the scales that participate in the pilot are high accuracy, the percentage of product recognition is 95-98 percent. The accuracy is affected by the packaging of the product. In addition, the scales display the result on the screen faster in a split second, and also require less training time.

The development allows you to increase the quality and speed of customer service, as well as reduce the time spent assembling orders in the store.

Шаблон:Quote 'author=said Lenta Innovation Director Olga Karaeva.

The company plans to complete the pilot in the next few months.