Samsung Electronics Co., Ltd.

Jongseong Choi

Seungwon Choi

Ildo Kim

Sukhwan Lim

Pilsu Kim

Ji Hwan Park

Ser Hoon Lee

Kyuyul Choi

Dongwoo Lee

Dong-bum Choi

Kai Guo

Na-ri Im

Juhyun Cho

Yun-Seong Jang

Byeongdae Lee

A demosaicing method for an input mosaic image includes: obtaining an input image, for each sub target pixel, estimating a main edge direction based on main color pixel values; determining the main edge direction by performing edge enhancement based on the input mosaic image; generating a main color demosaiced image by performing a main interpolation based on the main edge direction to generate interpolated main color pixel values corresponding to the sub target pixels; and generating a first sub color demosaiced image and a second sub color demosaiced image by performing a sub interpolation to generate interpolated first sub color pixel values and interpolated second sub color pixel values for the sub target pixels and for main target pixels corresponding to the main color pixel values.

Demosaicing method and demosaicing device

A method includes acquiring a current frame, a noise dispersion map for the current frame, and a first denoised frame for a previous frame; generating a weighted first denoised frame based on the noise dispersion map, the current frame, and the first denoised frame using a first neural network; generating an initial fused image based on the current frame and the weighted first denoised frame using a second neural network; and generating a second denoised frame for the current frame based on the initial fused image using a third neural network.

Image processing device, electronic device having the same, and operating method thereof

An image processing device includes; an image sensor including pixels arranged in a non-Bayer pattern and configured to generate non-Bayer pattern image data in response to electrical signals generated by the pixels, and an image signal processor configured to determine an indexing value indicating a color of a target pixel among the pixels and directly process the non-Bayer pattern image data in response to the indexing value.

Image signal processor processing image data having non-bayer pattern and image processing device including same

Provided are image signal processing apparatuses and methods for operating the same. In an embodiment, the image signal processing apparatus is configured to receive, from an image sensor device, an input image. The apparatus is further configured to perform a binning and crop operation on the input image to generate a first image. The apparatus is further configured to perform a Bayer domain processing on the first image to generate a second image. The apparatus is further configured to perform RGB domain processing on the second image to generate a third image. The apparatus is further configured to perform YUV domain processing on the third image to generate an output image. The YUV domain processing comprises at least one of a spatial de-noising operation, a temporal de-noising operation, a motion compensation operation, a tone mapping operation, a detail enhance operation, and a sharpening operation.

Image signal processor, operation method of image signal processor, and image sensor device including image signal processor

A method of reducing noise in an input image by setting, as a local window among color pixels included in the input image, a target pixel and neighboring pixels adjacent to the target pixel, determining color pixel values for the target pixel and each of the neighboring pixels included in the local window, generating local color average values are generated by averaging, color by color, the color pixel values, generating offset color pixel values by converting the color pixel values of the target pixel and the neighboring pixels based on the local color average values, and generating a compensated color pixel value of the target pixel by adjusting the color pixel value of the target pixel based on the offset color pixel values.

Denoising method and denoising device for reducing noise in an image

An image signal processor receives a Bayer image signal from an image sensor and converts the Bayer image signal into a normal image signal. The image signal processor includes a memory configured to store a table including reference brightness values and variation values according to the reference brightness values; a shot noise cancelation unit configured to calculate a reference brightness value of the Bayer image signal, select a variation value in the table of the memory according to the calculated reference brightness value, and perform shot noise cancelling on the Bayer image signal based on the selected variation value to generate a modified Bayer image signal; and an interpolation unit configured to generate the normal image signal by performing interpolation based on the modified Bayer image signal.

Growing community of inventors

Hwaseong-si, South Korea

Jongseong Choi