Developing a Cloud Computing Based Approach for Forensic Analysis Using OCR

One of the major issues in digital forensics is that cyber criminals tend to hide digital evidence before forensic tools are applied to find them. A simple but effective method to evade detection is to embed textual information in a picture such as a screenshot. On the detective´s side, we can use Optical Character Recognition (OCR) algorithms to retrieve the textual information from a picture or from a PDF document. However, the main challenges of the current solutions on standalone computers include scalability and performance. In this paper, we present a cloud-based framework which consists of a set of virtual machines and a job management system. We also show the mathematical derivation of our system configurations.