Detail Enhancement Evaluation
Yukyung Choi (KAIST), Namil Kim (NAVER LAB), Soonmin Hwang (KAIST), Jongchan Park (KAIST), In So Kweon (KAIST)
TEN: Detail Enhancer
TEN: Detail Enhancer
The DDE benchmark for multispectral thermal image enhancement tasks that contains 267 pairs of 16 bits format. This benchmark is consist of outdoor-specific and indoor-specific dataset, named RGBT-67 and T-138 respectively. They are all of good quality with clear edges but various scenarios, thus are very suitable for the enhancement training.
- download RGBT-67: FLIR A655sc, Outdoor
- download T-138 (set1, set2): FLIR T-640, Support MSX, Outdoor & Indoor
- download T-i3System
- download baseline code
Evaluation
Evaluation
Related Works
Related Works
- [BF-DRP] New technique for the visualization of high dynamic range infrared images, Optical Engineering (OE) 2009.
- [BF-DDE] Display and detail enhancement for high-dynamic-range infrared images, Optical Engineering (OE) 2011.
- [PF-DDE] Detail enhancement for infrared images based on propagated image filter, Mathematical Problems in Engineering (MPE) 2016.
Citation
Citation
To be appear.
Patent
Patent
- Apparatus and Method for Thermal Image Enhancement Application date: Aug. 5, 2016 (KR 1020160100058)