1. Adding geodesic information and stochastic patch-wise image prediction for small dataset learning
- Author
-
Adam Hammoumi, Sylvain Desroziers, Maxime Moreaud, Christophe Ducottet, IFP Energies nouvelles (IFPEN), Centre de Morphologie Mathématique (CMM), MINES ParisTech - École nationale supérieure des mines de Paris, Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL), Laboratoire Hubert Curien [Saint Etienne] (LHC), and Institut d'Optique Graduate School (IOGS)-Université Jean Monnet [Saint-Étienne] (UJM)-Centre National de la Recherche Scientifique (CNRS)
- Subjects
0209 industrial biotechnology ,Computer science ,Cognitive Neuroscience ,ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION ,02 engineering and technology ,stratified sampling ,distance transform ,patch-wise segmentation ,[INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI] ,Image (mathematics) ,020901 industrial engineering & automation ,Artificial Intelligence ,0202 electrical engineering, electronic engineering, information engineering ,Segmentation ,business.industry ,Deep learning ,image augmentation ,deep learning ,Pattern recognition ,Image segmentation ,Computer Science Applications ,Range (mathematics) ,Feature (computer vision) ,[INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV] ,020201 artificial intelligence & image processing ,Artificial intelligence ,business ,Distance transform - Abstract
International audience; Most recent methods of image augmentation and prediction are building upon the deep learning paradigm. A careful preparation of the image dataset and the choice of a suitable network architecture are crucial steps to assess the desired image features and, thence, achieve accurate predictions. We first propose to help the learning process by adding structural information with specific distance transform to the input image data. To handle cases with limited number of training samples, we propose a patch-based procedure with a stratified sampling method at inference. We validate our approaches on two image datasets, corresponding to two different tasks. The ability of our method to segment and predict images is investigated through the ISBI 2012 segmentation challenge dataset and generated electric field masks, respectively. The obtained results are evaluated using appropriate metrics: VRand for image segmentation and SSIM, UIQ and PSNR for image prediction. The proposed techniques demonstrate that the established framework is a reliable estimation method that could be used for a wide range of applications.
- Published
- 2021