Segmentace vršků palivových proutků
| Thesis title in Czech: | Segmentace vršků palivových proutků |
|---|---|
| Thesis title in English: | Segmentation of fuel rod top nozzles |
| Key words: | image segmentation|nuclear fuel inspection|top nozzle|neural network |
| Academic year of topic announcement: | 2023/2024 |
| Thesis type: | diploma thesis |
| Thesis language: | |
| Department: | Department of Software and Computer Science Education (32-KSVI) |
| Supervisor: | RNDr. Jan Blažek, Ph.D. |
| Author: | hidden - assigned and confirmed by the Study Dept. |
| Date of registration: | 02.04.2024 |
| Date of assignment: | 02.04.2024 |
| Confirmed by Study dept. on: | 02.04.2024 |
| Advisors: | Mgr. Tomáš Karella |
| Guidelines |
| Nuclear fuel undergoes visual inspection during shutdowns of nuclear power plants, monitoring various parameters including fuel rod growth. Rod growth is closely associated with fuel burnout and is thus utilized as a proxy for measuring burnout. Currently, a manual procedure exists for this purpose, involving the synchronization and alignment of vertical axis metadata with video frames containing images of the top and bottom nozzles of the fuel rods within a fuel assembly. The goal is to automate this procedure.
The thesis aims to address the challenge of precisely segmenting rod nozzles in video frames. Classical image processing algorithms as well as neural networks can be employed for this purpose. The proposed method generates masks that accurately identify individual fuel rods with well-defined boundaries. It's important to note that the dataset may contain multiple design types of fuel rods. A dataset comprising real fuel assemblies for both training and testing will be provided by Research Centre Řež, s.r.o. However, a separate synthetic dataset will be used for publication and demonstration purposes due to data ownership considerations. |
| References |
| Alexander Kirillov and Eric Mintun and Nikhila Ravi and Hanzi Mao and Chloe Rolland and Laura Gustafson and Tete Xiao and Spencer Whitehead and Alexander C. Berg and Wan-Yen Lo and Piotr Dollár and Ross Girshick, Segment Anything, 2023, eprint 2304.02643 arXiv
Cao, H. et al. (2023). Swin-Unet: Unet-Like Pure Transformer for Medical Image Segmentation. In: Karlinsky, L., Michaeli, T., Nishino, K. (eds) Computer Vision – ECCV 2022 Workshops. ECCV 2022. Lecture Notes in Computer Science, vol 13803. Springer, Cham. https://doi.org/10.1007/978-3-031-25066-8_9 Rafael C. Gonzalez and Richard E. Woods. 2006. Digital Image Processing (3rd Edition). Prentice-Hall, Inc., USA. |