Converting Colour to Black and White
You may just take all colour from your colour images. That is called desaturation. That means that each pixel gets converted into a grey pixel with the same luminosity that the colour pixel had before.
![]() |
![]() |
That is more or less what classic black and white photographers used to do. They just used some b&w film in their camera. However, that was often way to boring even for b&w photographers. That is why they used colour filters like red and green, which are called B&W filters although they are coloured.
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
Red | Green | Blue |
(The Moire effect in the first line is because I photographed this simulation from some flat computer screen. In real life those filters do not produce any moireés.)
In the lower line I simulated the effect of red, green and blue filters simply by extracting the red, blue and green channels of the RGB image.
Nearly all image manipulation programs have ready-made functions for b&w conversions. Most cameras have build-in picture styles. For Photoshop you will find lots of presets in the community. However, this article focusses on the basic so that you understand what happens behind the scene.
Decompose the Image
This is dead easy using Gimp. Under Colour/Components there is the menu item compose. Then go for "RGB" and Gimp creates a new image file with three layers. Each of the layers represents one of the RGB colour channels.
(I do apologize for the screenshots being in German. However, it gives you an idea on where to look for the features.)
Now you can switch on and off the visibility of each layer and use what remains visible. (The eye symbol left to each layer). You may mix the layers by giving appropriate percentage values to their visibility.
The Channel Mixer
The other important basic function is the channel mixer, which works similar with a wide range of image manipulation software.
![]() |
![]() |
The left example shows a domination of the green channel and the right one shows the red dominating.
The option "Monochrome" turned on cares for the result of the channel mix to be converted into grey values.
The option "Leuchtstärke erhalten" means "Keep Luminosity" and cares for the result to be of the same level of brightness as the original colour image was. Begin with having that one activated. You may want to switch it off if the result is way to dark or to bright.
![]() |
![]() |
Mixing the Colours 
Yo don't have to limit yourself to strict blue, red and white colour filters. You can mix them as you want, which is far advangageous compared to what the ealier b&w programmers had to do.
This is some guidance on how to simulate filters with mix colours:
BW-Filter | Red | Blue | Green |
Red | 100% | 0% | 0% |
Green | 0% | 100% | 0% |
Blue | 0% | 0% | 100% |
Blue-Green | 0% | 66% | 66% |
Yellow | 100% | 0% | 100% |
Orange | 33% | 0% | 66% |
And those look like these:
![]() |
![]() |
|
Original | desaturated | |
![]() |
![]() |
![]() |
Red | Green | Blue |
![]() |
![]() |
![]() |
Bluegreen | Orange | Yellow |
This excellent colourful photograph was contributed by Christoph Stegemann.
Selective Filtering
As seen above BW filtering plays an important role for portraits in black and white. Red filters reduce contrast on the skin and face. They let small veins and blemishes disappear. But freckles disappear alike. If you want to put emphasis on freckles then use some green or blue filter.
Well, you may not want to convert the remaining part of the image, the background, scenery, props etc. in the same manner as you want to convert the face. You may want to be selective here.
![]() |
![]() |
![]() |
![]() |
Original | Red | Green | Blue |
The red one would be way too bright and certainly not match the model's teint. With the green and blue variants the face is far too dark in the shadow. That is why I have chosen this as an example to demonstrate the selective approach.
You have decomposed the image anyway and got the colour layers separated. This time keep the visibility of each with 100%, fully visible. But add a layer mask. Do so by clicking the right mouse button on the layer (in the layers dialog - CTRL-L) and add a layer filled entirely black - fully transparent - for the layers above. (I just did it for the topmost red layer.)
The idea of a layer mask is, that the layer is invisible wherever the mask is black and it is fully visible where the mask is white. Grey values correspond to partly transparent pixels.
Now change the drawing colour to white and choose some soft tool (brush). Set the opacity to something like 30%. Now paint on the mask where ever you want the topmost layer, the bright one (red), to be visible. Draw more often where you want the topmost layer to be more intense.
Let's have a closer look at the layer dialoge. Unfortunately Gimp shows the mask very small here only.
You can make the mask visible for some more detailed look at the mask itself. That may help understanding what is going on on the mask.
And don't get confused by drawing on the mask or the layer itself. That happens frequently to me.
My example looks like this:
And now get started and create your best black and white conversions ever. There are far more images in your portfolio to which some bw-style just suits well.