Res-U2Net: untrained deep learning for phase retrieval and image reconstruction
Carlos Osorio Quero, Daniel Leykam, Irving Rondón
Abstract
Conventional deep learning-based image reconstruction methods require a large amount of training data, which can be hard to obtain in practice. Untrained deep learning methods overcome this limitation by training a network to invert a physical model of the image formation process. Here we present a novel, to our knowledge, untrained Res-U2Net model for phase retrieval. We use the extracted phase information to determine changes in an object's surface and generate a mesh representation of its 3D structure. We compare the performance of Res-U2Net phase retrieval against UNet and U2Net using images from the GDXRAY dataset.
Topics & Concepts
Deep learningComputer scienceArtificial intelligenceImage (mathematics)Representation (politics)Image retrievalProcess (computing)Pattern recognition (psychology)Phase (matter)Phase retrievalObject (grammar)Machine learningComputer visionMathematicsFourier transformLawMathematical analysisOrganic chemistryPolitical sciencePoliticsChemistryOperating systemAdvanced X-ray Imaging TechniquesOptical measurement and interference techniquesImage and Object Detection Techniques