Investigation and Automating Extraction of Thumbnails Produced by Image Viewers

Wybren van der Meer, Kim-Kwang Raymond Choo, M. Tahar Kechadi, Nhien-An Le-Khac

Abstract

Digital forensic investigations often rely on residual artifacts when primary evidence is deleted or encrypted. Thumbnails generated by image viewing applications serve as a critical "shadow" record of such evidence. However, third-party viewers often utilize proprietary or non-standard database formats that are frequently overlooked by traditional forensic tools. This paper presents a methodology for identifying, analyzing, and automating the extraction of thumbnails from popular image viewers like XnView, ACDSee, and FastStone. By combining real-time disk monitoring with refined data carving and SQLite parsing, the authors developed an automated Python-based tool capable of recovering thousands of thumbnails in under a minute, even from within complex database structures.

Introduction

In the realm of digital forensics, images are frequently the most vital pieces of evidence. Unfortunately, they are also among the easiest to delete or conceal within encrypted volumes. While data carving can sometimes recover deleted files, it is a time-consuming process that fails if the data has been overwritten. Thumbnails—low-resolution versions of images—provide a resilient alternative. These artifacts are often stored independently by the operating system or specialized viewing software. While system-level thumbnails (e.g., Windows thumbcache) are well-understood, the "wild west" of third-party application storage remains a challenge. This research addresses the need for automated identification and extraction of these application-driven forensic artifacts.

Methodology

The researchers employed a multi-stage approach to decode the storage mechanisms of various applications. First, they used FolderChangesView to monitor disk activity in real-time, identifying the specific directories and files where viewers saved their databases. Following identification, the internal structure of these databases was analyzed.

Forensic workflow for thumbnail extraction

For applications using SQLite (e.g., XnView, Adobe Lightroom), the team developed Python scripts to query Binary Large Object (BLOB) fields. For custom binary formats (e.g., ACDSee), they utilized signature-based carving. A critical refinement was the use of a "block size of 1" during carving. Standard tools often scan in 512-byte increments, which can miss thumbnails packed tightly within a database. By examining every byte, the researchers ensured maximum recovery.

Results and Insights

The study categorized storage strategies into four types: SQLite databases, custom binary databases, loose files (without extensions), and in-memory storage. The latter, found in apps like IrfanView, represents a significant forensic challenge as data is lost upon system shutdown.

Extraction of image data from an SQLite BLOB field

The automated script demonstrated high efficiency, processing forensic images and returning thousands of thumbnails in approximately one minute. For instance, it successfully extracted over 1,400 thumbnails from ACDSee’s proprietary format and parsed obfuscated data in XnView that standard industry tools like EnCase typically miss or require manual effort to recover.

Conclusion

Automating the extraction of application-specific thumbnails significantly reduces the time investment for forensic investigators. While the current tool relies on known database filenames and cannot recover RAM-only artifacts without live forensics, it provides a powerful proof-of-concept for deeper automated analysis. Future work involves expanding the tool to generically scan all SQLite databases for image-like BLOB data, regardless of the originating application.

Summary updated

Abstract

Today, in digital forensics, images normally provide important information within an investigation. However, not all images may still be available within a forensic digital investigation as they were all deleted for example. Data carving can be used in this case to retrieve deleted images but the carving time is normally significant and these images can be moreover overwritten by other data. One of the solutions is to look at thumbnails of images that are no longer available. These thumbnails can often be found within databases created by either operating systems or image viewers. In literature, most research and practical focus on the extraction of thumbnails from databases created by the operating system. There is a little research working on the thumbnails created by the image reviewers as these thumbnails are application-driven in terms of pre-defined sizes, adjustments and storage location. Eventually, thumbnail databases from image viewers are significant forensic artefacts for investigators as these programs deal with large amounts of images. However, investigating these databases so far is still manual or semi-automatic task that leads to the huge amount of forensic time. Therefore, in this paper we propose a new approach of automating extraction of thumbnails produced by image viewers. We also test our approach with popular image viewers in different storage structures and locations to show its robustness.
BibTeX
@inproceedings{meer2017investigation,
  author = {Wybren van der Meer and Kim-Kwang Raymond Choo and M. Tahar Kechadi and Nhien-An Le-Khac},
  title = {{Investigation and Automating Extraction of Thumbnails Produced by Image Viewers}},
  booktitle = {TrustCom/BigDataSE/ICESS},
  year = {2017},
  doi = {10.1109/trustcom/bigdatase/icess.2017.355},
}
Figure