Current artificially distorted image quality assessment (IQA) databases are small in size and limited in content. Larger IQA databases that are diverse in content could benefit the development of deep learning based IQA methods.
For this purpose, we created two datasets, the Konstanz artificially distorted image quality database (KADID-10k) and the Konstanz artificially distorted image quality set (KADIS-700k). The former contains 81 pristine images, each degraded by 25 distortions in 5 levels. The latter has 140,000 pristine images, with 5 degraded versions each, where the distortion was chosen randomly. Through the use of crowdsourcing, we conducted a subjective IQA study on KADID-10k and obtained 30 degradation category ratings (DCRs) per image.
With the given data, we proposed a NR-IQA model based on deep feature learning, the source code is available from
https://github.com/LinHanhe/DeepFL-IQA
**The pristine images in KADID-10k**
![enter image description here][1]
**Interface for subjective IQA study**
![enter image description here][2]
**Histogram of DMOS**
![enter image description here][3]
[1]: https://files.osf.io/v1/resources/xkqjh/providers/osfstorage/5eafe3f262d4ab00ce6c4b3c?mode=render
[2]: https://files.osf.io/v1/resources/xkqjh/providers/osfstorage/5eafe4b962d4ab00ca6c5274?mode=render
[3]: https://files.osf.io/v1/resources/xkqjh/providers/osfstorage/5eafe4f562d4ab00ce6c4da6?mode=render