Researchers have developed a way to search cloud storage sites like DropBox and Google Drive for digital evidence of illegal activities.

Evidence of crimes may be uploaded to the cloud, which makes it hard for investigators to identify the owners of the data. Purdue researchers built a cloud forensic model that scans images and video files as they are uploaded to cloud storage sites and flags potential evidence of cybercrimes, according to Fahad Salamh, a doctoral researcher at the Purdue Polytechnic Institute who helped create the system.

The StegnoCloud system taps deep learning models to classify child exploitation, drug trafficking and illegal firearms transactions and reports criminal activities via a forensic evidence collection system.

The solution allows cloud service providers to collect logs that have been flagged, block the associated accounts and report the activity to law enforcement based on a cloud search warrant request. It also reduces evidence storage size and the amount of time required to filter out false positives, and it makes it easy for CSPs to transmit digital evidence to evidence collection and analysis platforms, according to university officials.

The Purdue team tested more than 4,500 images, and the model accurately classified an image roughly 96% of the time.

“It is important to automate the process of digital forensic and incident response in order to cope with advanced technology and sophisticated hiding techniques and to reduce the mass storage of digital evidence on cases involving cloud storage applications,” Salamh said.

 

To learn how Granite Discovery‘s certified digital forensics team can help you, contact us today.

 

To view original post click here.

by GCN Staff / December 10, 2019