A method, apparatus, and a non-transitory computer-readable storage medium for temporally stabilizing White Point (WP) information in an Auto White Balance (AWB) process. The method may include receiving lux level, AWB statistics, and input WP. The method also include generating a histogram of the AWB statistics. The method may also include generating, from the histogram, a confidence value for the input WP. The method may also include determining a major colour count. The method may also include determining a threshold value based on the major colour count and the lux level. The method may also include setting an update state to locked when the confidence value is below the threshold value and otherwise setting it to unlocked. The final WP information is obtained from a previous frame when the update state is locked and otherwise from the input WP information.

An in-vehicle imaging device is provided with a detection region setting unit for setting a detection region, which corresponds to a predetermined object to be detected within an imaging screen; a color signal determination unit for setting a specific color corresponding to the object to be detected, and for determining whether the color data of pixels contained in the detection region are close to the specific color; and a gain control unit for averaging the color data of pixels (approximate color pixels) that were determined to be close to the specific color within the detection region and for adjusting the color gain of the image signal on the basis of a differential value of the color data of the specific color and the average value of the color data of the approximate color pixels.

Image analysis includes obtaining, from an image signal processor, image processing information corresponding to a previously processed image, obtaining scene classification information for an input image based on the image processing information, generating a processed image by processing the input image based on the scene classification information, and outputting the processed image. The image processing information includes automatic white balance correction information and obtaining the scene classification information includes obtaining the scene classification information based on the automatic white balance correction information.

An electronic device includes a camera configured to capture an image and generate an electrical image signal by photoelectrically converting incident light, and at least one processor configured to estimate a light source color temperature based on image data of the image signal, identify a target coordinate on a color space based on the light source color temperature, identify a capture coordinate on the color space based on the image, identify a chrominance allowable range based on the target coordinate, obtain a white balance evaluation value based on a first distance between the target coordinate and the chrominance allowable range and a second distance between the capture coordinate and the chrominance allowable range, and perform a white balance correction on the image based on the white balance evaluation value.

Systems and methods for sensor-independent illuminant determination are provided. In one embodiment of the method, the method includes receiving one or more training images in raw-RGB format; generating an input histogram from each of the inputted raw images; generating a learned mapping matrix that map raw images to a learned mapping space by passing the one or more input histograms to a trained first machine learning model; generating one or more mapped images by applying the learned mapping matrix to the one or more training images; generating a mapped histogram from each of the mapped images; and determining the result illuminant by passing the one or more mapped histograms as input into a second machine learning model. A final illuminant for an input image can be determined by applying the result illuminant to the input color space of the input image.

An example of an information processing system includes a moving object including a camera and including a white area in an exterior of the moving object. An image capturing range of the camera includes the white area. It is determined whether or not each of pixels in a preset specified area that includes the white area in an image captured by the camera is white, and if the number of pixels that are white according to the white determination exceeds a predetermined number, white balance is performed to adjust the pixels that are white according to the white determination in the specified area closer to white.

A multi-area white-balance control device includes a component that divides an inputted image into a plurality of division areas, a component that estimates a lighting-source color temperature for each division area, a component that calculates out color information of an areal image for each division area, and a component that acquires information for white-balance processing on the basis of the lighting-source color temperature and the color information of the areal image for each division in such a manner as to weight, in the information for white-balance processing, a multi-area white-balance factor as more significant in reference to a single-area white-balance factor in a case where the color information has a value higher than a predetermined value, than in a case where the color information fails to have the predetermined value. Thereby, an image with natural color reproduction can be generated.

A computer-implemented method for color correcting images captured using a mobile computing device is described. A wavelength of light is emitted from a display of a mobile computing device. An image of the emitted wavelength of light is captured using a camera on the mobile computing device. The image of the wavelength of light emitted from the display of the mobile computing device that is captured using the camera on the mobile computing device is analyzed. At least one image captured using the camera on the mobile computing device is modified based on the analysis of the image of the wavelength of light emitted from the display of the mobile computing device.

A method of ambient light suppression in an imaging system, including illuminating leaves of a multi-leaf collimator (MLC) with a first light of a lighting system inside a housing of the MCL, receiving ambient light inside the housing of the MLC through an aperture of the MLC, and capturing, using an imaging system having optics situated inside the housing of the MLC, an image of the leaves of the MLC illuminated with the first light and the ambient light. The method may further include, suppressing the ambient light in the first image to generate a second image of the leaves of the MLC and detecting a feature of the leaves of the MLC in the second image.

Provided is an exterior inspection device capable of performing inspections with an optimum white balance at all times. This exterior inspection device A is equipped with a camera 1, an inspection reference plane 2, and a light source 3, the exterior inspection device further comprising: a calibration reference plate 4; an advance/retraction mechanism 5 for advancing and retracting the calibration reference plate 4 between a calibration position within an image-taking area of the camera 1 and a retracted position retracted therefrom; and a control unit for issuing a white balance adjustment command to the camera 1 when the calibration reference plate 4 is at the calibration position. On an occasion of a non-inspection state in which no exterior inspection is being executed, the control unit 6 issues to the advance/retraction mechanism 5 a drive command for moving the calibration reference plate 4 to the calibration position, an adjustment command for adjusting light quantity of the light source 3, and a white balance adjustment command. Therefore, the white balance adjustment can be fulfilled automatically without involving human work.