Abstract: An image noise reduction method in frequency domain is provided, comprising selecting a first image and applying a Fourier transform thereon to acquire a first frequency spectrum, wherein each pixel of the first frequency spectrum has a real part X and a imaginary part Y, calculating a first energy of each pixel and a first mean energy of all first energies of the pixels, calculating a first mean value and a first standard deviation of the real part of the pixels, calculating a second mean value and a second standard deviation of the imaginary part of the pixels, determining a first and a second predetermined values, when the first energy exceeds the first mean energy, the first and second predetermined value are replaced with X and Y, and when the first energy does not exceed the first mean energy, X and the Y are reserved.

Abstract: A method for correcting skew images. The method first groups a scanned image into blocks with N*N pixels, and detects all the blocks to find out the edge areas with the larger luminance or chrominance difference afterward. Next, gradient angles of the edge areas are calculated, and a predominant angle is determined from among the calculated gradient angles; and (c) rotating the image according to the predominant gradient angle. The predominant gradient angle is taken as the skew angle for the image to be rotated. In consequence, a higher accuracy of image correction is achieved.

Abstract: An image noise reduction method in frequency domain is provided, comprising selecting a first image and applying a Fourier transform thereon to acquire a first frequency spectrum, wherein each pixel of the first frequency spectrum has a real part X and a imaginary part Y, calculating a first energy of each pixel and a first mean energy of all first energies of the pixels, calculating a first mean value and a first standard deviation of the real part of the pixels, calculating a second mean value and a second standard deviation of the imaginary part of the pixels, determining a first and a second predetermined values, when the first energy exceeds the first mean energy, the first and second predetermined value are replaced with X and Y, and when the first energy does not exceed the first mean energy, X and the Y are reserved.

Abstract: A method for correcting skew images is provided. The method first groups a scanned image into blocks with N*N pixels, and detects all the blocks to find out the edge areas with the larger luminance or chrominance difference afterward. Next, gradient angles of the edge areas are calculated to sum up their weightings. The gradient angle with the highest weighting is taken as the skew angle for the image to be rotated. In consequence, the method achieves a higher accuracy of image correction.

Abstract: A color image processing provides a thin-line enhancement module for a color image for finding and keeping thin lines in the color image. A boundary enhancement module is provided for finding and keeping the boundary in the color image. A de-background module is provided for removing noise pixels from the white background. The color image is processed to avoid missing and/or twisting the thin lines, boundary, background, and texts in the color image after the scanning procedure.

Abstract: A color image processing provides a thin-line enhancement module for a color image for finding and keeping thin lines in the color image. A boundary enhancement module is provided for finding and keeping the boundary in the color image. A de-background module is provided for removing noise pixels from white background. The color image is processed to avoid missing and/or twisting the thin lines, boundary, background, and texts in the color image after the scanning procedure.