Techniques are provided to adaptively enhance details of images. In one example, a method includes receiving a base image including a plurality of base pixels arranged in rows and columns. The method further includes performing a column-wise filtering process including processing subsets of the base pixels of each column selected by first and second sliding windows to generate a plurality of column-processed pixels. The method further includes performing a row-wise filtering process including processing subsets of the base pixels of each row selected by third and fourth sliding windows to generate a plurality of row-processed pixels. The method further includes combining the column-processed pixels and the row-processed pixels to generate a smoothed image that exhibits reduced detail in relation to the base image, and subtracting the smoothed image from the base image to provide a detail image. Additional methods and systems are also provided.

The present disclosure relates to a class-agnostic object segmentation system that automatically detects, segments, and selects objects within digital images irrespective of object semantic classifications. For example, the object segmentation system utilizes a class-agnostic object segmentation neural network to segment each pixel in a digital image into an object mask. Further, in response to detecting a selection request of a target object, the object segmentation system utilizes a corresponding object mask to automatically select the target object within the digital image. In some implementations, the object segmentation system utilizes a class-agnostic object segmentation neural network to detect and automatically select a partial object in the digital image in response to a target object selection request.

Provided is an image processing apparatus that includes a component obtainer configured to obtain diffusive reflection components and specular reflection components for pixels of an input image, a filter processor configured to perform filter-processing on a diffusive reflection component image, a specular reflection component image, and the input image, a component combiner configured to combine the filter-processed diffusive reflection component image and the filter-processed specular reflection component image to generate a combined image, and an evaluator configured to evaluate a separation accuracy of the diffusive reflection components or the specular reflection components based on the combined image and the filter-processed input image.

A computer-implemented method to background noise from a medical image comprising voxels, each voxel having a voxel intensity, the method comprising: generating a mask based on the medical image, wherein the mask comprises a foreground portion designating foreground voxels, a background portion designating background voxels, and a perimeter separating the foreground portion from the background portion; designating a perimeter portion of the mask, the perimeter portion enclosing the perimeter, a subset of the foreground portion, and a subset of the background portion; designating a threshold for the perimeter portion to separate voxels based on voxel intensity; filtering the medical image with the mask and the threshold to obtain a filtered image; and displaying the filtered image on a display.

There is provided is a method for enhancing a video image capable of achieving high image quality improvement for both a texture area and an edge area. A maximum limit value and a minimum limit value of a video signal value enhanced on the basis of a pixel signal value in an area set around the target pixel are calculated, the maximum limit value and the minimum limit value are corrected according to a texture value calculated as a scale value that indicates a variation of the pixel signal value in the corresponding area, and the corrected maximum limit value and minimum limit value are applied to the enhanced video signal value.

A non-transitory computer-readable storage medium stores executable instructions that, when executed by a processor, cause performance of operations comprising operations to access an image captured by an image sensor, obtain a transfer function for mapping pixel values, determine a faces indication that reflects a proportion of a scene depicted in the image that includes one or more human faces, and modify the transfer function based on the faces indication. Modifying the transfer function based on the faces indication comprises adjusting a gain of the transfer function to move the gain closer to unity. The operations include to apply the transfer function to pixel values of the image to produce a tone mapped image and output the tone mapped image.

A non-transitory computer-readable storage medium stores executable instructions that, when executed by a processor, cause performance of operations comprising operations to access an image captured by an image sensor, obtain a transfer function for mapping pixel values, determine a faces indication that reflects a proportion of a scene depicted in the image that includes one or more human faces, and modify the transfer function based on the faces indication. Modifying the transfer function based on the faces indication comprises adjusting a gain of the transfer function to move the gain closer to unity. The operations include to apply the transfer function to pixel values of the image to produce a tone mapped image and output the tone mapped image.

A device may receive an image and may process the image, with a first model or a second model, to convert the image into a binary image. The device may generate an identifier that identifies the first model, or identifies the second model and a color removed from the image, and may utilize clustering to cluster pixels of the binary image and to generate a segmented image with a quantity of segments. The device may generate a particular number of segments to select, and may select the particular number of segments, as selected segments, from the quantity of segments. The device may mask the selected segments to generate a protected image with masked segments, and may associate the protected image with the identifier and with original pixel data of the masked segments. The device may store the protected image, the identifier, and the original pixel data in a data structure.

Disclosed are an image content extraction method and an image content extraction apparatus, a terminal, and a storage medium. The image content extraction method includes acquiring an image to be processed. Performing high-contrast retention on the image to be processed to obtain a high-contrast image of the image to be processed. Performing image fusion on the image to be processed and the high-contrast image to obtain a fused image. Performing linear light enhancement on the fused image to obtain a linear light-enhanced image. And using first pixel points, having pixel values located within a preset pixel value range, in the linear light-enhanced image as an image content of the image to be processed. The present disclosure can improve the integrity of image content extraction.

A method includes positioning a patient at a first orientation relative to a radiation source. The method further includes using a 3D imaging technique to measure one or more positions of the patient's chest. The method further includes, while using the 3D imaging technique to measure the one or more positions of the patient's chest: generating a model of the patient's chest using the one or more positions of the patient's chest; updating the model of the patient's chest as the patient breathes; and exposing the patient to a dose of radiation using the radiation source, wherein the dose is based on the model of the patient's chest.