Device for removing noise on image using cross-kernel type median filter and method therefor
A device for removing a noise on an image using a cross-kernel type median filter includes a target pixel determination unit configured to determine whether a target pixel is a noise or not; a peripheral pixel determination unit configured to determine degrees of the noise on a peripheral pixel information of a peripheral pixel of the target pixel; and a noise removing unit configured to remove the noise of the target pixel and corrects the image by applying a cross-kernel type median filter based on the peripheral pixel information.
1. A method for removing a noise on an image using a cross-kernel type median filter, comprising the steps of:
determining whether a target pixel is a noise or not;
determining information on a peripheral pixel of the target pixel;
applying the cross-kernel type median filter based on the determined information of the peripheral pixel; and
correcting the image by removing the noise using the cross-kernel type median filter,
wherein the target pixel is disposed on a center of 3×3 region of the image,
wherein a median of 3×3 region is denoted as a first median, and a median of a region using the cross-kernel type median filter is denoted as a second median, and
wherein when an absolute value of a difference value between the first median and a data value of the target pixel is less than a reference value, the first median is outputted as the value of the data where the noise is removed, otherwise, the second median is outputted as the value of data where the noise is removed.
2. The method for removing the noise on the image using the cross-kernel type median filter of claim 1 , wherein the cross-type median filter selects a median from pixels, which are disposed on two lines when the two lines are across on a basis of the target pixel, which are disposed on a center of 3×3 region of the image.
3. The method for removing the noise on the image using the cross-kernel type median filter of claim 1 , wherein the two lines include ‘+’ shape or ‘X’ shape.
4. The method for removing the noise on the image using the cross-kernel type median filter of claim 1 ,
wherein when an absolute value of a difference value between a maximum value of 3×3 region and a minimum value of the 3×3 region is less than a reference value, the first median is outputted as the value of the data where the noise is removed, and otherwise, the second median is outputted as the value of data where the noise is removed.
5. The method for removing the noise on the image using the cross-kernel type median filter of claim 1 ,
wherein when the absolute value of the difference value between the first median and the data value of the target pixel is larger than the reference value, and the absolute value of the difference value between the second median and the data value of the target pixel is less than the reference value, the second median is outputted as the value of data where the noise is removed, and otherwise the data value of the target pixel is outputted as the value of the data where the noise is removed.
6. The method for removing the noise on the image using the cross-kernel type median filter of claim 1 ,
wherein when an absolute value of a difference value between a maximum value of the 3×3 region and a minimum value of the 3×3 region is less than a reference value, the first median is outputted as the value of the data where the noise is removed, and
wherein when the absolute value of the difference value between the maximum value of the 3×3 region and a minimum value of the 3×3 region is larger than the reference value, and the absolute value of the difference value between the data value of the target pixel and the second median is less than the reference value, the second median is outputted as the value of the data where the noise is removed, and otherwise the data value of the target pixel is outputted as the value of the data where the noise is removed.
7. The method for removing the noise on the image using the cross-kernel type median filter of claim 1 , wherein the reference value is a square root of the data value of the target pixel.
8. The method for removing the noise on the image using the cross-kernel type median filter of claim 7 , wherein the reference value is a value acquired by multiplying the square root of the data value of the target pixel by a standard constant.
9. The method for removing the noise on the image using the cross-kernel type median filter of claim 1 , wherein the reference value is a square root of an average value of the 3×3 region.
10. The method for removing the noise on the image using the cross-kernel type median filter of claim 9 , wherein the reference value is a value acquired by multiplying the square root of the average value of the 3×3 region by a standard constant.