In this exercise looking to familiarise yourself with histograms by relating each to the actual image as as aid to recognising the basic characteristics of an image.
Before I continue I think it is worthwhile explaining quickly what a histogram is. Basically a simple graph w
here tone is plotted against quantity. A histogram will show the distribution of brightness values of individual pixels in an image with deep shadows at the far left hand side and towards right hand side will show bright highlights. Useful to show distribution of tones across a wide dynamic range. The vertical height of the graph shows the relative number of pixels at each brightness level between black and white.
If any of the graph shows as spikes on either of the extreme right or left hand side of graph then this indicates that parts of the captured image are outwith the maximum and minimum brightness levels detectable by the camera sensor.
Before I continue I think it is worthwhile explaining quickly what a histogram is. Basically a simple graph w
here tone is plotted against quantity. A histogram will show the distribution of brightness values of individual pixels in an image with deep shadows at the far left hand side and towards right hand side will show bright highlights. Useful to show distribution of tones across a wide dynamic range. The vertical height of the graph shows the relative number of pixels at each brightness level between black and white.If any of the graph shows as spikes on either of the extreme right or left hand side of graph then this indicates that parts of the captured image are outwith the maximum and minimum brightness levels detectable by the camera sensor.
Ok enough of the theory onto the exercises and some practial examples.
High Contrast
For this part of the exercise, I took some pictures for this in Edinburgh's High Street, looking through closes (or alleyways you might call them). The pictures were difficult to expose with bright daylight through the far end of the close and the actual walls and road of the close very dark. I bracketed exposure and so got some under exposed, over exposed and close to well exposed.
When looking at output histogram I could see that the under-exposed photograph had most of the tonal range contained in the left hand side of the graph with a spike at this point. We are seeing lots of pure blacks with some of the image too dark for the sensor. You can see from screenshot below that there are large areas of shadow clipping represented by the blue. For completeness I have also shown picture without the clipping turned on.
The over-exposed photograph does have a wider spread of tone although this is partly due to the amount of the frame which is over exposed in relation to rest of frame. There is however a large spike at the right hand side of the histogram which I will expand on below. The red areas in the screenshot above show the areas of highlight clipping in the frame. for completenss I have shown the same image with clippingturned off.
High Contrast
For this part of the exercise, I took some pictures for this in Edinburgh's High Street, looking through closes (or alleyways you might call them). The pictures were difficult to expose with bright daylight through the far end of the close and the actual walls and road of the close very dark. I bracketed exposure and so got some under exposed, over exposed and close to well exposed.
When looking at output histogram I could see that the under-exposed photograph had most of the tonal range contained in the left hand side of the graph with a spike at this point. We are seeing lots of pure blacks with some of the image too dark for the sensor. You can see from screenshot below that there are large areas of shadow clipping represented by the blue. For completeness I have also shown picture without the clipping turned on.
The over-exposed photograph does have a wider spread of tone although this is partly due to the amount of the frame which is over exposed in relation to rest of frame. There is however a large spike at the right hand side of the histogram which I will expand on below. The red areas in the screenshot above show the areas of highlight clipping in the frame. for completenss I have shown the same image with clippingturned off.
If I crop frame just to show the over-exposed section (shown below) then it can be seen that the histogram shows a spike at the right hand side of the graph, at the opposite side of the graph from an under-exposed photograph. The graph below shows that the section of the photograph in bright sunlight has lots of highlights or pure white and that the spike shows that the captured image is too bright for the camera sensor.
Lastly I show a picture which is close to being correctly exposed. As picture is very high contrast, it still shows small amount of both highlight and shadow clipping. As these are quite small I could adjust the shadows and highlights to remove this clipping. The histogram tells me that the image is still too dark.
The tonal spread in this shot covers a wider range than in both the previous examples although still tends towards black. As an aside. I have played with this picture and can remove the clipping and make the tunnel far lighter. As it happens I think the darkness adds to the picture.
As a final illustration, I have adjusted shadows and highlights within Photoshop and the resultant image with histogram is shown below.
No comments:
Post a Comment