Learning-based image quality enhancement on CubeSat

The DISCO-2 project is driven by students and aims to develop and deploy a 3-unit CubeSat into low Earth orbit. Its mission focuses on conducting Earth observations over Greenland and supporting various research objectives. The satellite has three cameras onboard: infrared, wide-angle, and standard (main camera). Due to the limitations of the imaging hardware and the challenging conditions on the ground, the captured image might suffer from quality degradation, such as low light, noise, blurriness, shadow, etc. This proposal explores the full potential of the three cameras and runs learning-based image processing applications on power-constrained edge devices. Several topics include:

• Reference-based image low-light enhancement and denoising – enhancing the image quality at night in terms of illumination and contrast while suppressing noise by fusing the information collected from the three cameras.
• Reference-based image super-resolution – improving the spatial resolution of the captured image to produce richer details by fusing the information collected from the three cameras.
• Image de-fogging, de-hazing, or shadow removal – enhancing the image visibility under fog, haze, or shadows.

Multiple projects are possible, groups of 2+ are preferred. You must have experience with deep learning frameworks (PyTorch or Tensorflow) and the HPC at ITU.

References

Li, C., Li, Z., Liu, X., & Li, S. (2022). The Influence of Image Degradation on Hyperspectral Image Classification. Remote Sensing, 14(20), 5199.

Gharbi, M., Chen, J., Barron, J. T., Hasinoff, S. W., & Durand, F. (2017). Deep bilateral learning for real-time image enhancement. ACM Transactions on Graphics (TOG), 36(4), 1-12.

Shim, G., Park, J., & Kweon, I. S. (2020). Robust reference-based super-resolution with similarity-aware deformable convolution. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (pp. 8425-8434).

Galdran, A., Alvarez-Gila, A., Bria, A., Vazquez-Corral, J., & Bertalmío, M. (2018). On the duality between retinex and image dehazing. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (pp. 8212-8221).

Fu, L., Zhou, C., Guo, Q., Juefei-Xu, F., Yu, H., Feng, W., … & Wang, S. (2021). Auto-exposure fusion for single-image shadow removal. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (pp. 10571-10580).