The present invention relates to an image processing system (10), comprising: a processor unit (20) arranged to receive imaging data associated with an imaging system (40) and optical shape sensing data associated with an optical shape sensing system (50) registered with the imaging system (40) such that the optical shape sensing data can be positioned in the imaging system; wherein the processor unit (20) is configured to define in the imaging data a region of interest based on the imaging data and/or the optical shape sensing data and further configured to use the optical shape sensing data as markers within the region of interest such that the processor unit applies image enhancement of imaging data on the region of interest based on received optical shape sensing data.

An image processing method includes a first step of acquiring input data including a captured image and optical system information relating to a state of an optical system used for capturing the captured image and a second step of inputting the input data to a machine learning model and of generating an estimated image acquired by sharpening the captured image or by reshaping blurs included in the captured image.

Methods and apparatus are disclosed for implementing data augmentation within a storage controller of a data storage device based on machine learning data read from a non-volatile memory (NVM) array of a memory die. Some particular aspects relate to configuring the storage controller to generate augmented versions of training images for use in training a Deep Learning Accelerator of an image recognition system by rotating, translating, skewing, cropping, etc., a set of initial training images obtained from a host device and stored in the NVM array. Other aspects relate to controlling components of the memory die to generate noise-augmented images by, for example, storing and then reading training images from worn regions of the NVM array to inject noise into the images. Data augmentation based on data read from multiple memory dies is also described, such as image data spread across multiple NVM arrays or multiple memory dies.

The present invention provides a method, system and device for reflectional glare reduction in endoscope images. The reflective glare refers to clusters of congruent bright and often saturated image pixels caused by heterogenous surfaces of a tissue anatomy reflecting the illumination from a light source, which could be as annoying as distractive. Under automatic exposure control, in a scene of overexposure due to a reflection glare, the brightness of the glare is significantly higher than that of background image of tissue anatomy, and there is usually a gap between the two. Through mapping the pixel depth of the glare into the gap, the contrast between the glare and the background image is reduced, which improves the user experience of the endoscopy.

Techniques are described for using computing devices to perform automated operations to generate mapping information of a defined area via analysis of visual data of images, including by using attribute information exchanged between paired or otherwise grouped images of multiple types to generate enhanced images, and for using the generated mapping information in further automated manners, including to use the generated mapping information for automated navigation and/or to display or otherwise present the generated mapping information. In some situations, the defined area includes an interior of a multi-room building, and the generated information includes at least one or more enhanced images and/or a partial floor plan and/or other modeled representation of the building, with the generating performed in some cases without having measured depth information about distances from the images acquisition locations to walls or other objects in the surrounding building.

Techniques for generating a fused enhanced image. A first image is generated using a first camera of a first modality, and a second image is generated using a second camera of a second modality. Pixels that are common between the two images are identified. Textures for the common pixels are determined. A camera characteristic, which is linked to noise, is identified. A scaling factor is applied to the textures in the first image. A first saliency is determined using the scaled textures. A second saliency is determined using the textures from the second image. An alpha map is generated and reflects edge detection weights that have been computed for each one of the common pixels based on the two saliencies. Based on the alpha map, textures are merged from the common pixels to generate the fused enhanced image.

A system and methods for contrast sensitivity compensation provides for correcting the vision of users whose vision is deficient for discerning high spatial frequencies. The system and methods use measurements of the user's contrast detection as a function of spatial frequency in the image to correct images in real time. The system includes a head-mountable device that includes a camera and a processor that can provide enhanced images at video framing rates.

A system and methods for contrast sensitivity compensation provides for correcting the vision of users whose vision is deficient for discerning high spatial frequencies. The system and methods use measurements of the user's contrast detection as a function of spatial frequency in the image to correct images in real time. The system includes a head-mountable device that includes a camera and a processor that can provide enhanced images at video framing rates.

Disclosed is a system for image dehazing by modifying a lower bound of a transmission rate and a method therefor, whereby a clear image is obtained by removing components, such as fog or haze, from an image having low image quality due to fog or haze, the system configured such that a lower bound of a transmission rate representing how many rate of haze is mixed is calculated for each pixel, and an initial transmission rate is obtained by an exponentiation operation with the lower bound of transmission rate. By transmission rate correction that reserves an edge showing a large change relative to the initial transmission rate and processes a smooth area showing a small change with a low-pass filter, a final transmission rate is obtained such that a clear image is obtained by removing haze components.

A method of enhancing the visibility of blood vessels in a colour image captured by an image capturing device of a medical device, including, for at least some of the pixels of the image, the steps of: (a) processing data obtained from a first colour channel together with data obtained from a second colour channel to determine a value of a first parameter indicative of the intensity in the red spectrum relative to the total intensity of said pixel; (b) using said value of said first parameter and a first value of a user parameter to alter said pixel, the first value of the user parameter being based on user input, and wherein the strength of the alteration is dependent on both the value of said first parameter and the first value of said user parameter.