Abstract: Various methods and systems are provided for training a denoising system for a digital imaging system. The denoising system can be a deep learning denoising system formed as a blind or non-blind denoising system in which the training dataset provided to the denoising system includes a noisy image formed with simulated noise added to a clean digital image, and a reference image formed of the clean image having residual noise added thereto, where the residual noise is a fraction of the simulated noise used to form the noisy image. The use of the residual noise within the reference image of the training dataset teaches the DL network in the training process to remove less than all the noise during subsequent inferencing of digital images from the digital imaging system. By leaving selected amounts of noise in the digital images, the denoiser can be tuned to improve image attributes and texture.

Abstract: A method for performing fault-tolerant reconstruction of an image of an object is provided. The method includes acquiring a dataset corresponding to the object via a plurality of sensors, and detecting anomalous data within the dataset based at least in part on a statistical difference between the anomalous data and reference data within the dataset. The anomalous data is acquired by at least a first sensor of the plurality, and the reference data is acquired by at least a second sensor of the plurality that neighbors the first sensor.

Abstract: According to one embodiment, a method of image analysis is provided. The method includes binning image data into a plurality of sinogram frames, identifying a plurality of initial stationary frames by applying a first analysis technique on the plurality of binned sinogram frames, extracting a plurality of first statistical parameters applying a second analysis technique on the plurality of binned sinogram frames, combining the plurality of first statistical parameters with boundaries of plurality of initial stationary frames to generate a presentation of a joint analysis combining at least some of the plurality of the first statistical parameters and at least some of the plurality of the second statistical parameter, identifying a plurality of final stationary frames from the presentation of the joint analysis, independently reconstructing each of the plurality of final stationary frames, and registering each of the plurality of final stationary frames to a first state.

Abstract: A method for performing fault-tolerant reconstruction of an image of an object is provided. The method includes acquiring a dataset corresponding to the object via a plurality of sensors, and detecting anomalous data within the dataset based at least in part on a statistical difference between the anomalous data and reference data within the dataset. The anomalous data is acquired by at least a first sensor of the plurality, and the reference data is acquired by at least a second sensor of the plurality that neighbors the first sensor.

Abstract: According to one embodiment, a method of image analysis is provided. The method includes binning image data into a plurality of sinogram frames, identifying a plurality of initial stationary frames by applying a first analysis technique on the plurality of binned sinogram frames, extracting a plurality of first statistical parameters applying a second analysis technique on the plurality of binned sinogram frames, combining the plurality of first statistical parameters with boundaries of plurality of initial stationary frames to generate a presentation of a joint analysis combining at least some of the plurality of the first statistical parameters and at least some of the plurality of the second statistical parameter, identifying a plurality of final stationary frames from the presentation of the joint analysis, independently reconstructing each of the plurality of final stationary frames, and registering each of the plurality of final stationary frames to a first state.