APPARATUS AND METHOD FOR IMPROVING IMAGE QUALITY OF IMAGE SENSOR
First Claim
1. An apparatus for improving an image quality of an image sensor, the apparatus comprising:
- a first converting unit for converting an RGB color data obtained from bayer data through a de-mosaicking into a YCbCr color data;
a noise removal unit for removing a noise of a Cb and a Cr color data outputted from the first converting unit, and for determining a threshold value to be used during a filtering process by using data located at a center from the Cb and the Cr color data outputted from the first converting unit; and
a second converting unit for converting Cb and Cr color data outputted from the noise removal unit, and a Y color data from the first converting unit, into the RGB color data,wherein, when a statistical characteristic of noise added to adjacent pixels follows a Gaussian distribution, the adjacent pixels being disposed within a predetermined range that is determined by a threshold value for centering on a center pixel, are smoothed, the threshold value being a function of a standard deviation of the noise adjacent to the adjacent pixels.
1 Assignment
0 Petitions
Accused Products
Abstract
Provided are an apparatus and method for improving an image quality of an image sensor, capable of adaptively removing noise occurring in a de-mosaicking that is performed for generating three-channel data of R, G and B from a single channel pixel structure based on a bayer pattern. The apparatus includes a first converting unit for converting RGB color data into a YCbCr color data, the RGB color data being obtained from bayer data through a de-mosaicking, a noise removal unit for removing noise from a Cb and a Cr color data outputted from the first converting unit, and a second converting unit for converting the Cb and the Cr color data from the noise removal unit and a Y data from the first converting unit into the RGB color data.
-
Citations
25 Claims
-
1. An apparatus for improving an image quality of an image sensor, the apparatus comprising:
-
a first converting unit for converting an RGB color data obtained from bayer data through a de-mosaicking into a YCbCr color data; a noise removal unit for removing a noise of a Cb and a Cr color data outputted from the first converting unit, and for determining a threshold value to be used during a filtering process by using data located at a center from the Cb and the Cr color data outputted from the first converting unit; and a second converting unit for converting Cb and Cr color data outputted from the noise removal unit, and a Y color data from the first converting unit, into the RGB color data, wherein, when a statistical characteristic of noise added to adjacent pixels follows a Gaussian distribution, the adjacent pixels being disposed within a predetermined range that is determined by a threshold value for centering on a center pixel, are smoothed, the threshold value being a function of a standard deviation of the noise adjacent to the adjacent pixels. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15)
-
-
16. A method for improving an image quality of an image sensor, the method comprising:
-
converting an RGB color data into a YCbCr color data, the RGB color data being obtained from a bayer pattern through a de-mosaicking; removing a noise from a Cb and a Cr color data among the data converted into the YCbCr color data and determining a threshold value to be used during a filtering process by using data located at a center from the Cb and the Cr color data; and converting, the Cb and the Cr color data outputted after removing the noise from the Cb and the Cr color data, and Y data, into the RGB color data, wherein, when a statistical characteristic of noise added to adjacent pixels follows a Gaussian distribution, the adjacent pixels being disposed within a predetermined range that is determined by a threshold value for centering on a center pixel, are smoothed, the threshold value being a function of a standard deviation of the noise adjacent to the adjacent pixels. - View Dependent Claims (17, 18, 19, 20, 21, 22, 23)
-
-
24. A computer program product executing computer-executable instructions to perform the following acts:
-
removing a noise from a Cb and a Cr color data among data converted into YCbCr color data and determining a threshold value to be used during a filtering process by using data located at a center from the Cb and the Cr color data, wherein, when a statistical characteristic of noise added to pixels adjacent to Cb and Cr color data follows a Gaussian distribution, the adjacent pixels being disposed within a predetermined range that is determined by a threshold value for centering on a center pixel, are smoothed, the threshold value being a function of a standard deviation of the noise adjacent to the adjacent pixels.
-
-
25. An apparatus for improving an image quality of an image sensor, the apparatus comprising:
-
a means for removing a noise from a Cb and a Cr color data among data converted into YCbCr color data and determining a threshold value to be used during a filtering process by using data located at a center from the Cb and the Cr color data, wherein, when a statistical characteristic of noise added to pixels adjacent to Cb and Cr color data follows a Gaussian distribution, the adjacent pixels being disposed within a predetermined range that is determined by a threshold value for centering on a center pixel, are smoothed, the threshold value being a function of a standard deviation of the noise adjacent to the adjacent pixels.
-
Specification