CASIA v1.0 focuses on splicing detection evaluation. Image splicing is defined as a simple cut-and-paste operation of image regions from one image onto the same or another image without performing post-processing. It is a fundamental operation of tampering. CASIA v1.0 dataset contains 800 authentic and 921 spliced color images of size 384×256 pixels with JPEG format. The authentic images were mostly collected from the Corel image dataset and others are taken by our own digital cameras. We divide the authentic images into several categories (scene, animal, architecture, character, plant, article, nature and texture) according to their contents and also consider some pre-processing when making spliced images.
For more details, please click here. Here are some examples of CASIA v1.0.
Compared to CASIA V1.0, CASIA V2.0 is with larger size and with more realistic and challenging fake images by using post-processing of tampered regions. It contains 7491 authentic and 5123 tampered color images. The images in CASIA v2.0 are of difference sizes, varying from 240×160 to 900×600 pixels. Unlike CASIA v1.0, CASIA v2.0 includes uncompressed images and also JPEG images with different Q factors. The authentic images are collected from the Corel image dataset, websites with authorization of the image usage and our own images captured from digital cameras. We similarly divided all the authentic images into several categories as we did in V1.0, but we added an extra "indoor" category to consider the impact of image illumination. Moreover, we consider post-processing of (boundary of) spliced regions.
For more details, please click here. Here are some examples of CASIA v2.0.