Improving the image quality in sparse-angle X-ray computed microtomography using deep learning

X-ray computed microtomography represents a non-invasive method, which allows us to visualize the internal structure of objects, and therefore it is used in both industry and research. However, the measurement time required for data acquisition can be in the range of tens of hours. One way to shorten the measurement time is to reduce the number of acquired data, but this negatively affects the quality of the resulting image reconstruction. To improve the quality of the resulting image reconstruction, various interpolation techniques can be applied. In this work, selected interpolation methods that use deep learning will be applied to data from the laboratory of X-ray micro and nano computed tomography at CEITEC BUT and to data from a public source, and their effectiveness will be evaluated