American foodservice distributor anticipates restaurants' needs with machine vision and optical character recognition solution

Synaptiq’s machine vision and optical character recognition solution for retrieving, analyzing, and learning menu data provides customized foodservice plans for hundreds of thousands of restaurants. The result is more efficient supply chains, less waste, and successful customers.

Problem:

This multi-billion dollar American foodservice distributor wanted to sell proactively to restaurant clients but needed to understand their purchasing needs better. Specifically, the company analyzed restaurant menus, identifying popular dishes, ingredients, and the associated costs for each. However, menus can be extremely challenging to read, with varying fonts, designs, and colors. The challenge included building a model capable of analyzing hundreds of thousands of menus quickly enough to impact the company's sales and increase revenues positively.

What can other industries learn from AI in food services?

Our client's leadership is committed to uncovering new opportunities with their data and better anticipating the needs of restaurants, which improves sales and enhances their customer service. Working on this project also resulted in our Chief Data Scientist, Dr. Tim Oates, further investigating their data. He helped internal teams “think outside the box,” finding new opportunities for their data and leveraging it to drive additional revenues for the company. Synaptiq became a trusted partner through this work, sharing ideas and testing new hypotheses, such as correlations between foot traffic and sales. The initial engagement grew into a partnership, where all players work together to drive sales and optimization.

Solution:

Synaptiq developed a machine vision solution for this American foodservice distributor that analyzed and retrieved menu information from thousands of restaurant menus across the country. The goal was to extract the information, including menu items, descriptions, and prices, so the client could map all relevant data, from descriptions to ingredients. Ultimately our client wanted to sell more food products to their customers based on an in-depth understanding of their most popular menu items.

Synaptiq approached the problem as a standard object-detection and visual to verbal pairing process, which included autonomous data analysis. The AI solution included trained machine vision models for the menu items (name of dish, description of dish, and price) and incorporated OCR (Optical Character Recognition) technology to transform images of words into digital text.

In summary, Synaptiq:

• Built a machine vision model to detect similar objects.

• Trained the model to recognize menu items by descriptors.
  

• Implemented optical character recognition to turn images into digital text.

Outcome:

The result included a series of models that analyzed the contents of thousands of menus, retrieved information accurately, and processed this information significantly faster than manual labor.

The machine vision solution will "grow with" our client, recognizing and classifying menu data with more precision as it receives more data. OCR will allow the system to transcribe menus of all types, includes those with a creative design flair.

The model’s accuracy will improve as it sees more unique fonts, a form of collecting and applying more training data. Our client can now build a database to identify opportunities for sales (new restaurants that look like they could be purchasing even more products from our client) in addition to “wowing” current customers with their proactive, personalized, intuitive service.

HUMANKIND OF IMPACT

10,000 restaurant owners nationwide can maximize profits and rebound faster from pandemic losses.

100,000,000 more lives could be saved using rapid C-IMT results devices.

HUMANKIND OF IMPACT