The image processing apparatus acquires a plurality of types of candidate images based on an endoscope image, performs control of displaying, on a display, a display image based on at least one type of candidate image, performs a first analysis process on one or the plurality of types of candidate images set in advance, selects at least one type of candidate image from the plurality of types of candidate images as an optimum image based on a first analysis process result obtained through the first analysis process, and obtains a second analysis process result by performing a second analysis process on the optimum image.

An image edge detection method for processing an image including multiple pixels includes performing a convolution operation on the image by a gradient operator in a first direction and a gradient operator in a second direction to obtain a first-direction gradient data and a second-direction gradient data, wherein the first direction is perpendicular to the second direction, performing a gradient statistical calculation within a neighboring area of a target pixel of the image according to the first-direction gradient data and the second-direction gradient data to obtain a gradient statistic, and determining an edge significance corresponding to the target pixel according to the gradient statistic.

The present embodiment provides an image restoration method and apparatus which generate independent different restoration models by performing learning for each of different resolutions, receive a distorted image, and apply a restoration model corresponding to the resolution of the distorted image among the independent different restoration models to restore the distorted image into an improved upscaled image centering on a restoration target object within the distorted image.

Described herein is an image processing apparatus (701) comprising one or more processors (704) configured to: receive (601) a plurality of input images (301, 302, 303); for each input image, form (602) a set of decomposed data by decomposing the input image (301, 302, 303) or a filtered version thereof (307, 308, 309) into a plurality of frequency-specific components (313) each representing the occurrence of features of a respective frequency interval in the input image or the filtered version thereof; process (603) each set of decomposed data using one or more convolutional neural networks to form a combined image dataset (327); and subject (604) the combined image dataset (327) to a construction operation that is adapted for image construction from a plurality of frequency-specific components to thereby form an output image (333) representing a combination of the input images. The resulting HDR output image may have fewer artifacts and provide a better quality result. The apparatus is also computationally efficient, having a good balance between accuracy and efficiency.

An example electronic device may include a memory, an image sensor, and at least one processor operatively connected to the memory and the image sensor, wherein the memory is configured to store instructions which, when executed by the at least one processor, cause the electronic device to acquire, through the image sensor, a plurality of images including first images having a first size and at least one second image having a second size larger than the first size, acquire a first synthesis image based on the first images, acquire a first moving object portion from the first synthesis image, based on synthesis area information including at least one of moving object location information or background location information, identify a second moving object portion in the at least one second image, based on the synthesis area information, and acquire a second synthesis image by replacing the second moving object portion by the first moving object portion, and other embodiments are possible.

An information device including a capturing function uses the capturing function to capture an image of a disc wheel to which a plurality of wheel fasteners are fastened. Then, the information device inspects whether any of the wheel fasteners appearing in the captured image is loose by comparing the captured image with a reference image in consideration of a reference point of the disc wheel appearing in the captured image.

Various methods and systems are provided for real-time image segmentation of medical image data. In one example, the real-time image segmentation of the medical image data may include updating an initial segmentation of the medical image data in real-time. The update may be based on a user input to a regularization brush applied to the medical image data, the user input to the regularization brush allowing modification of a volume of the initial segmentation.

Provided are methods for correcting multi-zone motion blur, which include executing, using at least one processor, an alignment of at least one image capturing device with at least one collimating device in a plurality of collimating devices, causing a rotation of at least one collimating device, receiving at least one image of at least one target object captured by the image capturing device for processing by at least one rotating collimating device, and determining, based on the at least one processed image, a degradation of the received image of the target object.

An image processing method and apparatus, and a non-transitory computer-readable storage medium are provided. The method includes: acquiring an exposure duration and at least one motion component corresponding to at least one direction within the exposure duration of a current video frame; performing a first noise reduction operation on the current video frame in a two-dimensional space domain in response to a first motion component of the at least one motion component being greater than a preset motion component threshold; and performing a second noise reduction operation on the current video frame in a three-dimensional space domain in response to the at least one motion component being less than the preset motion component threshold.

Recurrent blurring may be used to render frames of a virtual environment, where the radius of a filter for a pixel is based on a number of successfully accumulated frames that correspond to that pixel. To account for rejections of accumulated samples for the pixel, ray-traced samples from a lower resolution version of a ray-traced render may be used to increase the effective sample count for the pixel. Parallax may be used to control the accumulation speed along with an angle between a view vector that corresponds to the pixel. A magnitude of one or more dimensions of a filter applied to the pixel may be based on an angle of a view vector that corresponds to the pixel to cause reflections to elongate along an axis under glancing angles. The dimension(s) may be based on a direction of a reflected specular lobe associated with the pixel.