NETRA

An Identity for your image

Comment

Inspiration

With the rise of edited and AI-generated media, it has become difficult to trust photos and videos online. We noticed that when an image is questioned, people often struggle to prove that it is original. Existing solutions either rely on platform trust or complicated tools. We wanted to build a simple system that helps users verify media using structured comparison instead of guesswork.

What it does

NETRA is a media verification system that analyzes a photo or video at the time of capture and stores its structural details in a secure database. Instead of saving the entire image, it records important visual characteristics and object layout information. When someone later wants to verify the media, the system compares the uploaded file with the stored reference data and highlights any differences or inconsistencies.

How we built it

We built NETRA using image processing techniques to extract stable visual features such as object positions, structural layout, and pattern information. These details are stored in a structured JSON format in a database. During verification, the system reprocesses the uploaded media, extracts the same features, and performs a database comparison to detect changes. We also built a web-based interface for uploading and validating media easily.

Challenges we ran into

Making the system robust against compression and screenshots

Deciding what features to store without saving the entire image

Avoiding false positives for minor visual changes

Designing a database structure that is efficient and scalable

Accomplishments that we're proud of

Built a working prototype that performs structured media comparison

Designed a database-driven verification process

Created a system that does not need to store full images

Developed a clear mismatch detection interface

Built With

Share this project:

Updates