Project 6: measuring colour

I was directed to make a duplicate print of my personal colour chart and scribe the RGB and HSB values onto it.



Personal Chart showing HSB and RGB values

Typing all these values on was time consuming but it does mean I have a permanent record stored electronically, it also reminded me that to get the ° symbol all I had to do was Alt+0176... handy to know!

Taking three photographs, one to contain neutrals, one skin tones and the final to have vegetation I had to measure and note down the values of the important colours and compare them to the OCA chart and my own personal colour chart.

I began with looking at the skin tones in a portarit of my son...he loves me using his images ;o)




Portrait for checking skin tones

A few issues to begin with as he doesn't have a uniform skin tone, quite an olive complexion on the whole, he does have a healthy pink glow therefore readings were different depending which area I sampled....I chose an area at random and then compared the results. My image was quite close to the OCA chart, on my personal chart they were not taken from actual skin but even so the difference is fairly minimal. Quite pleased with that ;o) Means I got the skin tone fairly accurate!

Comparing Charts



 

Values	Photograph	OCA Chart	Personal Chart
H	15°	20°	11°
S	22%	25%	28%
B	91%	95%	92%
R	231	241	234
G	193	200	180
B	180	180	168

Secondly I looked at the image containing vegetation.



Quite a few of my vegetation shots I had previously used to compile my personal chart therefore thought it would be pointless comparing them as they would have the same values. Instead I chose this tulip shot which had a variety of greens in the stem and background, although not vegetation I thought I'd compare the red too.



Comparing Charts

By eye once again the colours do not appear that dissimilar and the values are fairly close too. The darker greens in the background when sampled were very dark and didn't really match anything on either chart.



Values	Photograph			OCA Chart			Personal Chart
	LG	DG	R	LG	DG	R	LG	DG	R
H	64°	64°	3°	64°	69°		64°	63°	3°
S	100%	99%	99%	93%	63%		76%	84%	98%
B	40%	42%	99%	84%	43%		82%	43%	84%
R	96	100	253	201	100		198	104	213
G	103	107	15	215	110		208	109	15
B	0	1	2	16	41		49	17	4


Finally the neutrals were sampled for a street image taken on the Commercial Road in London



Image to check neutrals

There were not many neutral tones on my personal chart, the only one that I could attempt to match was black.



Comparing Charts

From the values sampled you can tell that my neutrals aren't really neutrals at all. I guess that would be due to a mixture of colour cast and also some of the surfaces could be reflecting other colours. Quite a few areas on the grey road were also made of  different tarmac, some appearing to have a pink mix....



In my photographs the black has a very high saturation but the other neutral tones are close to zero, being 9,4 and 1. RGB levels aren't equal but are quite close. Not sure now what the settings of my camera were it could be I had the incorrect WB set at the time, although not neutral I was surprisingly pleased at how acceptable they were.



Values	Photograph				OCA Chart				P/C
	1	2	3	4	1	2	3	4	1
H	252°	60°	103°	60°	0°	0°	0°	0°	266°
S	136%	9%	4%	1%	0%	0%	0%	0%	44%
B	5%	51%	70%	98%	0%	50%	75%	0%	6%
R	10	129	174	249	0	128	191	255	12
G	19	129	179	249	0	128	191	255	9
B	14	117	172	247	0	128	191	255	16



Once again this exercise has shown how important it is to understand colour cast and noting which areas of an image should be neutral so post processing can be done as accurately as possible. Although pleased that my images are not as far out as I thought they might have been, I can see there is room for improvement.



Project 5: a personal range of colours

This project gave me the opportunity to make a colour chart based upon my own preferences and experiences. Invariably photographers find themselves attracted to a certain tones of colour or a particular topic, some specific locations will also tend to have similar colours. Not sure if I do have a predilection towards certain hues it was fascinating to look at some examples of my work to find out. One of the benefits of this task was the creation of a colour reference point to match against and gain consistent future results.

The main aim was to familiarise myself with the analysis of colours and the identification of those to which I respond the most.

To choose the 20 colours for the grid I decided to look closely at the images produced for my final DPP assignment also some shots of Spring flowers recently taken. It was really interesting to discover that although I hadn't set out to include similar colours/hues throughout the set, when put together and sampled the same colours kept repeating. Is this due to my taste, subliminal or otherwise, or just what nature provided? Below is a montage of the final 10 images for DPP, these being reflected in the first two rows of my colour chart. I can see now how the image colours link, with blues, greens, reds, the creams and blacks. There seems to be a reasonable balance between hot and cold colours in the first two rows.

Colour chart squares 1-10

The bottom two rows are colours picked from photographs of spring flowers taken over the past few weeks. As anticipated most of these are subtle pastel shades; muted greens, yellows and lilacs. There were a few vibrant reds/yellows and  these were similar to the shades in my first set.

Colour chart squares 11-20

Personal colour chart

Printed out this chart will  help with obtain accurate colour capture/representation in future images.

Memory Colours

Colour is an important aspect in our lives, we make not realise it but people seem to remember and associate familiar objects by their colour – a red bus, a green apple, the blue sky.... It appears that colour images are more vivid and easier to remember. Colours and tones can be used to create a style or feel in an image, but some tones have to be "correct" for the audience to connect with it, for example  if a skin tone is not acceptable we will automatically query it, but skin tone is just one of many "memory colours". We know the sky is blue and grass green; simply put memory colours are colours which people find are inseparable from certain common objects or events.

Scientific experiments (Cambridge Research Systems - Research Topics) have shown that colour seems to help people process and store images as well as remembering them. This memory of colour probably harks back to human development when it importantly helped people to find food, such as ripe fruit and remember what is dangerous, such as poisonous berries and stinging insects.

The most familiar are the most important i.e. skin tone and neutrals followed by vegetation and sky tones, by eye we can usually say whether these are right or wrong.

The next part of this section made me look at the colours with more precision, measuring the colours of the charts and on some of my own photographs. If there is a colour cast the Black, White and Grey squares will not have RGB values which are identical, if they are all the same they are described as being "neutral" therefore this is a good way to check for colour cast. Cameras and monitors measure colour solely in RGB however, the human eye perceives colour also in terms of Hue, Saturation and Brightness. Hue being measured in degrees starting with Red at the top at 0°(I never knew that red was always at the top in a spectrum circle!) while Saturation and Brightness are measured in percentages. If I can improve my estimation of the HSB colours that I capture it will assist with my colour correction and colour accuracy of my final images. Practise makes perfect?

It was interesting to discover that my tastes are quite conservative, I don't have that many bold or brash colours and I do like the odd dash of red. That is my favourite colour! I have never heard of memory colours before but can fully appreciate how they work in the "real" world as well as photography and recognise even more how important it is to obtain correct colours. Not only due to the fact that as photographers we like to create true representations but also it could affect the way our audience perceives the images we make.

Project 4: your camera's colour performance

For this project I investigated the colour performance of my camera. Before completion I thought it achieved fairly accurate results and was intrigued to see if this assumption was correct. I had to take a photograph of the colour chart provided and compare it to the digital file on the OCA disk. The photograph was taken in daylight, white balance corrected and on auto metering.

When opening the photograph and comparing it to the digital file OCA_colourchart_as looks it looked awful!


Chart pre colour cast adjustment

Using the colour picker I measured and noted the values of first 5 squares comparing the two charts

Chart	1	2	3
OCA Chart	17,17,17 H 0,S 0, B 7%	52,52,52 H 0,S 0, B 20%	97,97,97 H 0,S 0, B 38%
400D Chart	1,3,1 H 120, S 67, B 1%	57,48,50 H 347, S 16, B 22%	144,140,140 H 0, S 3, B 56%
	4	5
OCA Chart	161,161,161 H 0,S 0, B 63%	243,243,243 H 0,S 0, B 95%
400D Chart	216,217,222 H 230, S 3, B 87%	249,251.251 H 180, S 1, B 98%

Due to the colour cast it was recommended that Enhance was used to correct it. As I am using CS4 the menus and techniques are slightly different, I had to refer to the Adobe help website to find out how to do this.

adjusting cast using curves

Chart after cast adjustment

To the eye it looks much better although there is still a difference. Measuring was tricky as the colour was not uniform and varied across the coloured squares.

Chart	1	2	3
OCA Chart	17,17,17 H 0,S 0, B 7%	52,52,52 H 0,S 0, B 20%	97,97,97 H 0,S 0, B 38%
400D Chart	0,1,0 H 120, S 100, B 1%	49,52,53 H 195, S 8, B 21%	134,134,138 H 240, S 3, B 54%
	4	5
OCA Chart	161,161,161 H 0,S 0, B 63%	243,243,243 H 0,S 0, B 95%
400D Chart	213,4, 85 H 230, S 3, B 87%	254,255,255 H 180, S 1, B 100%

The blacks were registering fairly close to the RGB levels of 0,0,0, that you would expect and the white close to 255,255,255 even at 134,134,138 the grey was not too far off 128,128,128. Not sure what was happening with the Hue,Saturation and Brightness readings though? Maybe I needed to zoom in a little closer due to the uneven tones of the boxes? I think completing this exercise has left me a little confused as to how to interpret the results. I have examined two colour boxes to see if I could figure out the numbers from there....but there didn't seem to be a correlation? There was not more red in them all or more blue....the green box indicated my camera had more R less G and less B which suggested a red cast but then the blue box measured less on all 3? My head hurts :o/ What I have concluded is that adjusting the mid-tones/BW points is something I should be doing as a matter of course even if I have set a custom white balance. I may not be able to fully understand the numbers I am seeing but there was certainly a colour cast to correct. The Saturation and brightness seem to be close but the hues are definately higher.

Working with colour

This section asks us to start taking into consideration the colour of the objects we photograph and the capability of our cameras to accurately capture these. Not thinking about the colour of light at present, but taking this aspect into account the camera should always be set to the correct white balance because the aim of the exercise is to reproduce a scene and its contents accurately. I must admit this was the first exercise that I looked at and groaned ;o) I can see the benefits of completing the tasks but it look so fiddly and time comsuming....sometimes I have to kick myself and remind myself that the mundane and basic experiments often provide the most interesting results and useful information. *KICKS SELF*

The PwDP manual sums it up quite simply (good because some of the scientific sites I looked at went whooshing over my head) reminding us that 'the colour that we see is partly the reflection of the light and partly the properties of the surface itself' and that some objects seem to have 'a different colour under different kinds of light'. This is known as metamerism (suckers for punishment can look here Metamerism if you want to wade through long confusing paragraphs!)

The first stage of checking how to deal with the colour of things is to compare a digital photgraph with the real thing. This is the bit where I laughed and wondered how many people actually unplugged their monitor and all the peripherals to move their pc next to a window? The reason this was suggested is due to the fact that the colour temperature of the window and a monitor should not be far off....luckily my PC isn't too far away from a window......a neutral object was photographed, on a window sill with an Auto white-balance setting. The image was then uploaded to the PC, printed and the colour compared between the real thing, the print and the monitor.

I chose this mug as it has fairly moderate to neutral colouring. The colour match wasn't too bad although the mug itself is a more creamy white than the monitor shows. When the image was printed out it was reasonably close to the monitor hues. I always calibrate my monitor and a few months ago spent quite a while ensuring my colour balance and screen brightness/contrast were as accurate as possible. The image below isn't ideal as I moved the mug from the sill and the print is in front of the monitor but it gives the idea of what the experiment was. The monitor wasn't as blue as this image suggests....

Variations can be caused by the camera, the colour space that the computer and Photoshop are working in and the calibration of the monitor. I make sure I regularly calibrate mine and a few weeks back spent a while making sure the brightness/contrast of my monitor was also set correctly as sometimes my prints were too dark. It is the aim of every photographer (unless into manipulation) that the final printed image is as close to the original scene as possible. However images are most often viewed away from the real object and we have no control over where the photographs will be eventually viewed/displayed. That is another argument altogether so it is a given that common sense should win out in the end otherwise with the alternative lies madness ;o) Seen apart all three are quite acceptable.

In the appendix is an OCA colour chart to study and note the tones, a digital file is also provided but this will differ due to the different range of contrast that paper has compared to a computer screen. You SHOULD be able to see from the two images uploaded here the differences you would expect to see between the monitor images and the printed page.... not sure on my current browser if I can...on my monitor there is a distinct difference....The printed image does not have such pure colours; the colours represent colours likely to be found in real scenes. Intense hues being fairly unusual in nature, these normally are found in advertising etc.

OCA Colour Chart image property of Open College of Arts © 2004 

OCA Colour Chart As Looks image property of Open College of Arts © 2004

The chart Colour Chart mid grey as RGB 128,128,128 zero Hue,zero Saturation and 50% Brightness. Colour charts can be purchased to check the way cameras record colour.

Planning and Thinking Ahead

At the beginning of a new course there are the usual things to think about, time management, materials, learning curve etc but until you actually see what you need to produce there are some things that can't be planned too far in advance. I intended to keep a physical log as well as an online blog as I did for DPP and assumed that it would still be an A5 spiral pad. However looking at some of the research I may do, or as the Harvard referencing guide suggests printing out some of the webpages accessed, it might be better to make it A4...or even a ring binder... I think I'll start off with a ring binder. If this proves unwieldy I may change my mind...something to watch.

I've made contact with Joe and confirmed how he likes assignments to be written up/submitted; almost the same as before but he doesn't mind if jpegs are emailed for assessment rather than uploading to a web gallery. Not a problem :o)

Harvard Referencing System...

Knowing that I will have to write a critical essay and use this system the pdf was duly downloaded and for ease of reference printed out. Browsing through it most of the system makes common sense and I dare say it will be easier to follow and remember when put into practise. During DPP I made sure all websites and books used were acknowledged but realise this needs to be done properly when producing academic work. Some issues became apparent, such as books being a revised edition but not stating what edition, if I come across any of these in my research I need to check how to reference that properly. Some websites don't have published or copyright dates on and it takes me ages to check it isn't hidden anywhere on the site. I can see that noting the reference may take longer than typing out the essay!

To assist me with nailing the system I have been making sure I have a draft post which has all the sources used for research so far and think I shall publish a post at the end of each section.