Disclosed herein are systems and methods for adjusting appearance of objects in medical images.

A device comprises processing circuitry to receive an input polygon comprised of a plurality of segments connected by a plurality of points, some of the plurality of segments extending in a first direction and remaining ones of the plurality of segments extending in a second direction perpendicular to the first direction. The processing circuitry performs a first set of operations on the input polygon to generate an intermediate polygon having a plurality of vertices, a number of the plurality of vertices being fewer than a number of the plurality of points. The processing circuitry performs a second set of operations on the intermediate polygon based on traits of the plurality of vertices to generate a final polygon with a contour that represents the input polygon, the contour including at least one section that extends in a third direction different than the first direction and the second direction.

An image processing apparatus, including a memory configured to store one or more instructions; and at least one processor configured to execute the one or more instructions to: based on a first image and a probability model, optimize an estimated pixel value and estimated gradient values of each pixel of an original image corresponding to the first image, obtain an estimated original image based on the optimized estimated pixel value of the each pixel of the original image, obtain a decontour map based on the optimized estimated pixel value and the estimated gradient values of the each pixel of the original image, and generate a second image by combining the first image with the estimated original image based on the decontour map.

A method for building a map includes: acquiring an original grayscale map, preprocessing the original grayscale map to obtain a preprocessed map, binarizing the preprocessed map to obtain a binarized map, performing a boundary filling to the preprocessed map and the binarized map to obtain a boundary-filled preprocessed map and a boundary-filled binarized map, performing a boundary thinning to the boundary-filled binarized map to obtain a thinned binarized map, and performing a boundary thinning to the boundary-filled preprocessed map, according to the thinned binarized map, to obtain a thinned preprocessed map.

An electronic device and method for shape refinement of a 3D mesh reconstructed from images is disclosed. A set of images of an object is acquired and used to estimate a first 3D mesh of a head portion of the object. A first set of operations is executed on the first 3D mesh to generate a second 3D mesh. The first set of operations includes a removal of one or more regions which are unneeded for head-shape estimation and/or a removal of one or more mesh artifacts associated with a 3D shape or a topology of the first 3D mesh. A 3D template mesh is processed to determine a set of filling patches which corresponds to a set of holes in the second 3D mesh. Based on the second 3D mesh and the set of filling patches, a hole filling operation is executed to generate a final 3D mesh.

An image processing apparatus performs first noise reduction processing on a plurality of radiation images corresponding to mutually-different radiation energies, generates a decomposition image by energy subtraction processing using the plurality of radiation images obtained by performing the first noise reduction processing, and performs second noise reduction processing on the decomposition image, wherein the second noise reduction processing uses a filter that differs from a filter used in the first noise reduction processing in at least one of size and type.

A system, method, and computer program product are provided for capturing digital images. In use, an ambient image comprising at least two ambient pixels and a flash image comprising at least two flash pixels is captured, via a camera module. Next, at least one de-noised pixel based on the ambient image is generated. Additionally, a resulting image is generated by combining the at least one de-noised pixel and a corresponding flash pixel. Additional systems, methods, and computer program products are also presented.

An apparatus and method are described for a non-uniform rasterizer. For example, one embodiment of an apparatus comprises: a graphics processor to process graphics data and render images using the graphics data; and a non-uniform rasterizer within the graphics processor to determine different resolutions to be used for different regions of an image, the non-uniform rasterizer to receive a plurality of polygons to be rasterized and to responsively rasterize the polygons in accordance with the different resolutions.

An image processing method, a drive device, a display panel, and a wearable device are disclosed. The image processing method includes: determining an adjacent display pixel adjacent to each grayscale transition region in the row direction or in the column direction in the display image region according to a position of the grayscale transition region; determining a transition pixel in the grayscale transition region; acquiring a first pixel grayscale, in which the first pixel grayscale is a grayscale of the adjacent display pixel; acquiring a second pixel grayscale; adjusting a third pixel grayscale of the transition pixel according to the first pixel grayscale, the second pixel grayscale and the transition pixel, in which the third pixel grayscale is between the first pixel grayscale and the second pixel grayscale; and transmitting the third pixel grayscale to the display panel for display.

A hybrid method of low light image enhancement in an electronic device comprises downsampling a high resolution, low light image to obtain a low resolution, low light image and inputting the low resolution, low light image into a low light enhancement model of a deep neural network to obtain a low resolution, enhanced image where semantic information such as true colors, edges, and brightness level are recovered. The high resolution, low light image, the low resolution, low light image, and the low resolution, enhanced image are each input into a mathematical model that performs upsampling and super-resolution tasks to generate a high resolution, enhanced image.