DIGITAL FORENSICS
Technology
We built image and video recognition technology modules that can be used to process large amounts of image and video files and quickly find matches between them. Results of processing these image and video files are stored locally and can be reused in the future to run matches with the newly processed images and videos. Unlike hashes such as MD5, the technology will detect the following non identical images as matching:
different resolutions
blurred
monochrome flipped
The technology will also find matching segments within video files even in different resolution files that are encoded with different encoders, where content is flipped or blurred and audio missing or muted.
Use case
When hardware devices are taken as part of an investigation, and their images are taken and sent for processing to gather evidence, our technology can be used to quickly find matching images and video files between these devices. Previously processed cases and their devices can also quickly be scanned for matches. Being able to automatically make a connection between a new device with a device processed in another case in the past can help uncover leads not easily identifiable without this technology. If an image or a video was taken with a mobile device camera and shared through chat applications, a lower resolution file is usually shared between devices making a simple hash unsuitable for matching these files while Beyond Seen Screen's technology will identify a positive match.
Feedback
We would like to get more feedback about the possible application of this technology. Do you find the ability to match non identical image and video files useful in your line of work?
If you click any buttons above, your email client will be started and an email message will be prefilled allowing you to give us quick feedback.
Email us to find out more
