Edge crispening color interpolation method based on gradient

Abstract

The invention discloses an edge enhancement color interpolation method based on gradient which comprises the following procedures: (1) a gradient threshold is set, the value range of the gradient threshold is between 0 and 255; (2) the interpolation is made to all defect green components and obtain a horizontal gradient and a vertical gradient of a position corresponding to an interpolation point which is ready to interpolate, the green components are obtained by edge enhancement if both are more than the threshold in Step (1); (3) corresponding color interpolation is made to the current point as non-green components to obtain the gradients of two diagonals of the current point, corresponding components are obtained by edge enhancement if both are more than the threshold in Step (1); (4) the interpolation of red component and blue component is made to the current point as green components to obtain a horizontal gradient and a vertical gradient of a position which the current point corresponds to, corresponding components are obtained by edge enhancement if both are more than the threshold in Step (1). The method of the invention can eliminate the edge noise brought by color interpolation and can also increase the effect of edge sharpeness.

Description

technical field [0001] The invention relates to a color interpolation method, in particular to a gradient-based edge enhancement color interpolation method. Background technique [0002] Among various existing imaging devices, most of them use CFA image sensors. Each photosensitive element in this type of CCD or CMOS sensor can only collect a specific component of the pixel color, and in the entire sensor matrix, there is only one photoreceptor for each point, that is to say, only one photoreceptor can be used for each pixel. One color component is captured, while the other two components are lost. Therefore, a method must be provided to compensate for the lost two components so that each output pixel has red (R), green (G), and blue (B) component outputs. [0003] At present, there is a relatively simple method, which is to use the average value of corresponding surrounding color components to calculate the current defect component. [0004] Such as figure 1 As shown, t...

AI Technical Summary

Problems solved by technology

However, since the three coefficients α, β, and γ are calculated by comparing MSE, the processed picture is closest to the original picture, so it does not have the effect of edge sharpening.

Images