science

Forget about facial recognition, this Japanese bakery has built a bread recognition AI

[ad_1]

Japanese company trains AI to recognize different types of breads and pastries and is selling the technology to bakeries

  • Japanese tech firm Brain Co. has developed a system to recognize baked goods
  • Called BakeryScan, the AI-powered tech is used in more than 400 bakeries
  • The system evaluates bread for size, shape, color, and surface patterns

Faces are far from the only thing artificial intelligence researchers want computers to recognize. 

One Japanese tech company has created an AI-powered camera system, called BakeryScan, that is able to recognize different types of breads and pastries. 

Developed by Japanese tech and media firm Brain Co., the scanning tools are currently installed in more than 400 retail locations across Japan. 

One of the technology’s biggest clients is the Andersen Group, a chain of high-end Danish-inspired bakeries that operate across Japan, according to a report from The Asahi Shimbun.

Scroll down for video 

BakeryScan (pictured above) is an AI-powered checkout system that uses cameras and computers to identify different types of baked goods

BakeryScan (pictured above) is an AI-powered checkout system that uses cameras and computers to identify different types of baked goods

BakeryScan works through a camera that’s mounted above a backlit checkout tray.

Customers place their selections on the tray and then the camera analyzes the bread or pastries, cataloging their size, shape and color to match them with one of up to 100 different types stored in the checkout system.

The cashier confirms the match via a touchscreen display, and then the customer pays.

The recognition process takes place in about second.

BakeryScan analyzes breads for their shape, size, color, and signs of different fillings and makes a determination in about a second

BakeryScan analyzes breads for their shape, size, color, and signs of different fillings and makes a determination in about a second

One of the challenges for the project was learning to distinguish baked goods that appear identical from the outside but have different fillings or ingredients.

The developers had to design a code that would allow the AI to recognize small spots where ingredients poked through the surface, or otherwise scan for faint shadow patterns on the surface that hint at what’s inside.

In instances where the recognition software can’t guarantee the accuracy of its finding, a yellow edge appears around its selection on the touch display. 

BakeryScan is currently used in more than 400 bakeries across Japan

BakeryScan is currently used in more than 400 bakeries across Japan

HOW DOES BREADSCAN WORK?

BreadScan is an AI-powered checkout system used in Japanese bakeries.

Customers place their desired bread or pastry on a tray at the checkout counter.

A camera mounted above the tray photographs the baked goods and analyzes them for color, shape, size, and signs of different fillings.

Within a second, the AI identifies the bread or pastry and sends the information to a touchscreen display for the cashier.

If the AI isn’t 100 percent certain of its selection, the choice is highlighted in yellow and the cashier makes the final determination. 

The cashier then makes the final determination, either overriding the AI’s recommendation or confirming it.

‘There are more than 10,000 bread stores across the nation,’ Brain’s Hisashi Kanbe told The Asahi Shimbun.

‘I wanted to pitch our product to the huge market to move our business forward.’

Bread may just be the beginning of a much bigger and potentially more lucrative business.

Scientists from the Louis Pasteur Center for Medical Research in Kyoto have reached out to Brain Co. to inquire about using the technology to help them identify cancer, which they say resembles bread at the cellular level.

Identifying cancer cells to determine whether tumors are benign or malignant can be labor intensive.

Having an AI assistant could dramatically speed up the process and lead to earlier diagnoses and more effective treatment for patients.

[ad_2]

READ SOURCE

This website uses cookies. By continuing to use this site, you accept our use of cookies.  Learn more