The present invention provides a technology that separates a low-contrast-ratio image into sublayer images, classifies each sublayer image into several categories in accordance with the characteristics of each sublayer image, and learns a transformation matrix representing a relationship between the low-contrast-ratio image and a high-contrast-ratio image for each category. In addition, the present invention provides a technology that separates an input low-contrast-ratio image into sublayer images, selects a category corresponding to each sublayer image, and applies a learned transformation matrix to generate a high.

The invention provides a method of processing an image in a diagnostic apparatus 100 of diagnosing a lesion using a captured image of an affected area to be diagnosed, comprising the steps of: (i) classifying the captured image based on a stage of a progression of the lesion (Step S13 and S14); and (ii) performing an image conversion processing, which corresponds to a classification obtained as a result of step (i), on the captured image to generate a converted image (Step S15-17).

Disclosed are systems, methods, and computer-readable storage media to perform an interactive image recolorization process. The method includes receiving user input including a stroke drawn on an image presented on a client device. The stroke comprises a user-specified color. The method further includes determining a region of interest in the image. The method further includes recolorizing the region of interest on the image based on the user-specified color and causing presentation of a result of the recolorization on the client device.

Adaptive focus sweep (AFS) techniques for image processing are described. For one technique, an AFS logic/module can obtain an AFS representing a scene, where the AFS is a sequence of images representing the scene that includes: (i) a first image representing the scene captured at a first focus position; and (ii) a second image representing the scene captured at a second focus position that differs from the first focus position. The first focus position can be associated with a first depth of field (DOField) that is determined based on an autofocus technique. The second focus position can be associated with a second DOField, where the second focus position is at least two DOFields away from the first focus position. The AFS logic/module can detect a foreground of the scene in the first image based on information acquired from the first and second images. Other embodiments are described.

The invention provides a method of processing an image in a diagnostic apparatus 100 of diagnosing a disease using a captured image of an affected area, comprising: a memorizing step of memorizing the captured image (Step S12), and a processing step of processing the captured image memorized (Step S13), wherein in the processing step a region to be diagnosed is subjected to a highlighting process with a specified color thereof maintained.

In an example embodiment, method, apparatus and computer program product are provided. The method includes facilitating receipt of a deconvolved image including a plurality of component images. A guide image is selected from the component images and a cross-filtering is performed of component images other than the guide image to generate filtered component images. The cross-filtering is performed of a component image by iteratively performing, selecting a pixel and a set of neighboring pixels around the pixel in the guide image, computing a set of weights corresponding to the set of neighboring pixels based at least on spatial differences between the pixel and the set of neighboring pixels, and cross-filtering a corresponding pixel of the pixel in the component image based on the set of weights to generate a filtered corresponding pixel in the component image. The filtered component images form a filtered deconvolved image with reduced chromatic aberration.

A method for reducing image disturbances caused by reconstructed defective pixel clusters located in signal-gradient affected diagnostic image regions. An individually adapted central symmetrical pair reconstruction (CSP) kernel is composed for a defective image pixel based on a kernel-pair candidate order encoded in a model thereby using the pixel's validity state. The image impacted by defective pixels is corrected in real-time by statistical filtering or spatial convolution of the kernel-associated image data accessible via a predetermined CSP kernels image-offsets structure.

To extend the working range of depth from defocus (DFD) particularly on small depth of field (DoF) images, DFD is performed on an image pair at multiple spatial resolutions and the depth estimates are then combined. Specific implementations construct a Gaussian pyramid for each image of an image pair, perform DFD on the corresponding pair of images at each level of the two image pyramids, convert DFD depth scores to physical depth values using calibration curves generated for each level, and combine the depth values from all levels in a coarse-to-fine manner to obtain a final depth map that covers the entire depth range of the scene.

Systems and methods are disclosed for calibrating an image sensor using a source image taken from the image sensor and comparing it to a reference image. In one embodiment, the method may involve determining the luminance and chrominance values of portions of the image at successive frequency levels and calculating a standard deviation at each frequency level for both the source image and the reference image. The standard deviation values may be compared and a difference determined. Using unit vector search vectors, noise values may be calculated to determine sensor calibration values.

This application relates to the field of image processing technologies, and discloses a video denoising method performed by a computing device, and a storage medium, which are configured to reduce the computation amount for video denoising and relieve network bandwidth pressure. The method includes: acquiring a current frame of a target video; in accordance with a determination that the current frame is a P frame or a B frame in the target video: determining a reference frame of the current frame from the target video according to a pre-existing time domain reference relationship between the current frame and the reference frame, the time domain reference relationship being established in advance by an encoder; determining a reference block in the reference frame corresponding to a current block in the current frame; and performing denoising on the current block according to the reference block.