SPOT

Inspiration

A receipt contains far more information than just a simple price.

It includes the product name, purchase history, store information, address, and sometimes important information related to food ingredients. However, most receipts are discarded or kept without being used again.

We wanted to turn receipts into useful data.

Our idea was simple: take a photo of a receipt and let AI understand what’s inside it.

SPOT was developed to convert ordinary receipts into structured information, helping users better understand their purchase history, food, and consumption patterns.

Function

SPOT analyzes receipt images using OCR and AI.

Depending on the receipt, the system can provide various types of information.

For example:

Extract text and product information from receipts
Detects location or address information and indicates where the purchase or delivery came from
Analyze food-related purchases
Provide information about ingredients and precautions Organize purchase records into structured data
Analyze repeat purchases and identify buying patterns
Generates AI-based insights based on the user’s purchase history

Instead of treating receipts as disposable paper, SPOT considers them a source of valuable personal information.

How We Made It

This project is centered around an OCR-based processing pipeline.

First, the user uploads a receipt or takes a photo.

The OCR system extracts text from images and converts receipts into data that machines can read.

After that, process the extracted text to identify the following important information:

Product name
Save information
Address
Food-related items
Purchase Information

After the information is structured, AI analyzes the data and derives additional insights.

Location-related information can link the extracted address to a map or a GPS-based API.

In the case of a food receipt, the verified product can be analyzed to provide ingredient information and relevant precautions.

It can also save previous purchase data and be used to identify patterns in multiple receipts.

What we learned

Through this project, we learned that OCR is not simply about converting an image into text.

The more important problem is understanding what that text means.

We learned how unstructured real-world data can be transformed into structured information that AI can analyze.

We also learned how different technologies can work together in one pipeline:

Image → OCR → Structured Data → AI Analysis → User Insight

This project also showed us how everyday objects can become useful data sources when combined with AI.

What's next for SPOT

In the future, we want to improve SPOT by adding:

More accurate receipt classification Better ingredient and food information analysis Purchase history dashboards Personalized purchasing insights Spending and consumption pattern visualization Location-based purchase history Automatic product categorization

Ultimately, our goal is to make SPOT a system that does not just read receipts, but understands what users buy and turns that information into something useful.