Computer Vision News - August 2018

histogram_ref_file -- this parameter specifies a file containing a reference histogram to be used for normalization; this parameter is based on the paper “ New variants of a method of MRI scale standardization ” IEEE transactions on medical imaging 19.2 (2000): 143-150 [TRAINING] defines the training parameters, such as loss function, number of iterations, number of patches to sample from each volume, number of checkpoints (points at which the network parameters, values and hidden state are backed up), image augmentation to be used during training (the amount of scaling, rotation, etc.). 19 Tool Computer Vision News NiftyNet name = unet activation_function = prelu batch_size = 1 decay = 0 reg_type = L2 volume_padding_size = 44 # histogram normalisation histogram_ref_file = ./example_volumes/monomodal_parcellation/standardisation_models.txt norm_type = percentile cutoff = (0.01, 0.99) normalisation = True whitening = True normalise_foreground_only=True foreground_type = otsu_plus multimod_foreground_type = and queue_length = 128 window_sampling = uniform