Go - Converting an Image to Grayscale

发布时间 2023-10-10 15:51:46作者: ZhangZhihuiAAA

Problem: You want to convert the image to grayscale.


Solution: Convert an image to a grid of pixels. Take each pixel in the grid and convert it to a gray pixel according to the relative luminance formula. Convert the grid of pixels back into an image to get a grayscale image.

 

A grayscale image is an image that has pixels that show different shades of gray representing the amount of light intensity. Black and white represents the opposite ends of the spectrum, with black having the least amount of light and white having the most.

To create a grayscale image from a color image, you can calculate the relative luminance of each pixel from the red, green, and blue values of each pixel. There are a few formulas for calculating this relative luminance, but the simplest is just to take the average of the red, green, and blue values:

L = (R + G + B)/3

Here’s the code to do the conversion:

func   grayscale ( grid   [][] color . Color )   ( grayImg   [][] color . Color )   { 
      xlen ,   ylen   :=   len ( grid ),   len ( grid [ 0 ]) 
      grayImg   =   make ([][] color . Color ,   xlen ) 
      for   i   :=   0 ;   i   <   len ( grayImg );   i ++   { 
          grayImg [ i ]   =   make ([] color . Color ,   ylen ) 
      } 

      for   x   :=   0 ;   x   <   xlen ;   x ++   { 
          for   y   :=   0 ;   y   <   ylen ;   y ++   { 
              pix   :=   grid [ x ][ y ].( color . NRGBA ) 
              gray   :=   uint8 ( float64 ( pix . R ) / 3.0   +   float64 ( pix . G ) / 3.0   + 
                  float64 ( pix . B ) / 3.0 ) 
              grayImg [ x ][ y ]   =   color . NRGBA { gray ,   gray ,   gray ,   pix . A } 
          } 
      } 
      return 
}

The luminosity formula here is the simplest, but there are other formulas defined by various standards to convert color to grayscale. One of these standards is the ITU - R BT.709 standard from the International Telecommunications Union (ITU) Radiocommunication Sector, which produces a better result:

L = 0.2126 * R + 0.7152 * G + 0.0722 * B