TAOP: Other Learnings – Exposure Issues
December 28, 2011
After posting my assignment pictures on Flickr and adding to the OCA group on there, I got some feedback regarding the exposure of a couple of my images. Essentially they were underexposed and it was suggested I review the Histograms for the images to recognise this and to allow me to make adjustments as appropriate. Having never looked at Histograms I had no idea how to interpret what I was seeing.
I found the following tutorial which has helped me understand this:
I’ve summarised what I’ve learnt from this below:
Histograms essentially show you how many pixels of an image are at different tonal points (tone is the intensity of colour/light within an image). On my camera these appear as an RGB Histogram and a separate Exposure Histogram. The scale runs from 0 to 255 on the tonal axis. 0 is ‘little’ 255 ‘lots’. The higher the graph, the more pixel of the image are within that tonal range. So, on an Exposure Histogram a peak towards the left of the graph (nearer to 0) would mean there are a large number of dark pixels. A peak towards the right of the graph (nearer to 255) would mean there are a large number of white pixels.
A Histogram’s appearance will depend very much on what you are shooting – shooting something with a lot of contrast may well show peaks towards either end of the graph. Shooting a more balanced scene with less contrast would tend to show a more even distribution, with a peak more towards the mid-point of the graph for the mid-tones.
A quick and dirty example of the Exposure Histogram on the back of my camera is shown below:
There are a lot of dark tones here – from the black on the TV, speakers and media stand (and various black boxes) – so the Histogram is reflecting this with a peak toward the left-hand-side. There are some mid-tones from the browns and other colours – hence the smaller peak in the middle of the graph. Finally, the cream paint for the background is shown by a series of peaks toward the right-hand-side of the graph. I think in this example the image is slightly under-exposed, with detail lacking in the black regions due to the lack of contrast provided by under-exposing the image.
I’m going to leave the Histogram on in my camera for a while so I get used to interpreting it to ensure pictures are properly exposed.