Thursday 29 May 2014

Colour - Grey Card Test (Foliage Canopy)

Typical Forest Birding!

It is May in a northern deciduous woodland.  Birding in the woods is usually more about songs and calls than crippling views and when that warbler, flycatcher or thrush finally pops up it is generally bathed in vivid green, making ID more of a challenge.  

HERE I summarised white balance corrections, including the limitations of manual white balance using just the Kelvin colour temperature scale, versus a Grey Card, which can correct for all colour casts, including the green channel.  I also unveiled a "pimped up" grey card rig.  Now to put this rig to the test in the field.

Here is the setting, a typical woodland scene in Ireland in May - pretty green!


Here is the test rig shot without white balance correction.


Now here is that shot again with DNG profile and white balance correction completed - that's just two mouse clicks!



White balance involved shifts in both the Colour Temperature (Red and Blue channels) and in what Adobe refers to as "Tint" (the green channel).  Despite an incredibly overwhelming green cast, white balance was a doddle.

Here is a breakdown of the white balance correction in Camera Raw.  First Colour Temperature.


Colour temperature is reading 5050K.  The yellow component of the foliage has been removed but the green channel is unaffected by the colour temperature correction.  As explained HERE manual white balance tools often don't allow for corrections in the green channel.  Grey card corrections however do correct all channels.


Here, a massive correction in the green channel (Tint +77) was required to bring white balance in line.

The purpose of the rig is for a more in depth view so here is the analysis bit.


Overall the white balance performed really well in this setting.  There is very little if any residual colour casting evident in any of the targets.  There is a slight green cast still evident in the shadows (residual foliage influence) and a slight blue cast evident in the highlights (sky influence). 

In summary, I think it is a very good idea to carry a Grey Card into the forest!

Colour - White Balance, A closer look

Pimp my Grey Card!

I decided to create a little mobile laboratory to delve more deeply into the complexities of light.  
More of that in a moment.  First a quick recap on white balance.

The Colour of Light

Sunlight is at it's purest when the sun is highest in the sky.  As the sun approaches the horizon, the colour of sunlight changes from white, through yellow near the horizon, to reddish right on the horizon.  
The reason for this is simple.  Light is scattered by the atmosphere and the denser the atmosphere the greater the scattering of light.  The atmosphere between the observer and the horizon is many times thicker than it is between the observer and zenith (looking straight up).  This causes drastically more scattering of sunlight when the sun is at or near the horizon.  Atmospheric pollution including smoke and ash only serves to increase this scattering affect.  The bluer, shorter wavelength portion of light is scattered the most by the atmosphere, deflected by particles and molecules in the air.  

We observe this phenomena, termed RAYLEIGH SCATTERING, in two ways.  Firstly we observe the sky as blue.  This is the shorter, mainly blue light being scattered in all directions, including towards the earth and out into space.  In effect, the blue sky canopy is a blue light source capable of lighting in all directions, including into deep shadows.  Secondly, we observe the sun and the light coming from it as appearing progressively yellower, then redder, as the sun nears the horizon.  This is just the same phenomenon that gives us the blue sky, except in reverse.  As the blue portion has been progressively scattered, the light that remains appears progressively redder.  

Here I have created a representation showing the sun's position at high noon in Ireland on both the summer and winter solstice.  The difference in the quality of light is remarkable.  Note the density of the atmosphere at the horizon relative to zenith - it is minimum 37.5 times denser, and a lot more if the air is polluted!


My animated gif below gives a representation of the changing nature of daylight over the course of 24 hours here in Ireland during both the summer and winter solstice.  A number of things are particularly striking.

(1) Light temperature is ever-changing during the winter solstice.  It never properly stabilises.
(2) The purity of daylight is already much better come mid-morning in mid-summer than it is at any time during a winter's day, simply because the sun never gets high enough above the horizon in winter.
(3) Colour temperature is very stable for long periods of the day in mid-summer.  While this may seem like an advantage, a high, bright sun brings it's own challenges both for bird observation and photography.




Colour Temperature and Black Bodies

We have a scale for measuring the colour of light.  It is the Kelvin scale.  So, what has a temperature scale got to do with light one might ask?  Well it turns out that the colour appearance of light closely matches the colour appearance of objects (specifically black bodies) as they are heated.  If you heat a pot on an oven it glows red as it gets hot.  If you put it in a furnace you will note the pot's colour change from red to yellow to white and eventually to blue.  As these colours mirror the colours of daylight they represent a useful comparative scale.  It also happens that artificial light sources from candles to light bulbs produces light colours that can be measured along the same scale.  Thus, the Kelvin scale has become the standard benchmark for studying and measuring the colour of light.  It is referred to as COLOUR TEMPERATURE and here is the scale as it is normally displayed.  The numbers are temperatures in degrees Kelvin.


You will notice that the scale contains no violet or magenta, though in my gif above I have included these.  The fact is that light colour varies quite a bit as the sun approaches and drops below the horizon.  There is often a magenta glow at sunset, called ALPENGLOW.  At twilight, as the light fades away the sky is bathed in light of increasingly short wavelength from blue, through indigo, violet, ending in ultra-violet light which is invisible to us.

Colour Temperature and Colour Theory

Colour theory describes all colours in terms of three colour channels, Red, Green and Blue.  Cyan is the opposite of Red, Magenta the opposite of Green and Yellow the opposite of Blue.  You may have noticed that the Kelvin scale does not contain Green or Magenta.  Effectively one could argue the green channel is unaffected by the position of the sun in the sky, except around dusk - the famous green flash and alpenglow (see above).  This has an important bearing when it comes to white balance correcting tools as I will explain in a moment.

White Balance

Why bother with white balance?  Well white balance is not simply an invention of the camera world, it is an intrinsic part of the human visual system, and presumably that of other animals that see in colour.
As an adaptation it makes sense.  If we are on the look out for colourful fruits and hazardous plants marked by different colours, it helps that our brain adjusts our visual senses to cater for ever-changing light temperature.  

The reason we correct white balance in photographs is mainly for aesthetic reasons, to mimic what the human brain sees.  In terms of bird identification it also serves a useful purpose in allowing us to appreciate colours as they might look under more ideal lighting.

Most manual white balance tools only consider the blue and red channels and only make corrections aligned with the Kelvin colour scale.  So if there are other colour "contaminants" in an environment (eg. a vivid green cast created by a foliage canopy (see HERE), or say a bright magenta evening sky) manual white balance correction tools will not solve those lighting problems.  This is where the use of manual white balance tools, based solely on the Kelvin scale fall short.  Not only does the human visual system cater for colour casts along the Kelvin scale, our brain can adjust to any colour cast, including a green or magenta cast.


For instance in MS Office 2010 the Color Tone scale used for manual white balance image correction only utilises the Kelvin scale so it cannot make a colour correction in the green channel.

Grey Card White Balancing

One of the key advantages of using a grey card for white balancing is that it works in all three channels.
The colours in the scene will be normalised, at least in so far as removing any colour casts that influence the grey card.  It is important that the grey card is positioned properly to reflect the lighting conditions on the subject. If the subject is in shade but the grey card is in the sun the white balance correction will not accurately reflect the lighting on the subject.

Pimp My Grey Card

While the grey card and colorchecker are fantastic tools they don't address all of my questions around the strange, changing behaviour of light and how it interacts in our environment.  I decided to bolt on some additional instruments to my colorchecker to gather more information for analysis.  The rig, and these different instruments are explained below.  


Various tests involving this rig are presented HERE.

For a look at manual white balance correction see HERE.



   



Tuesday 20 May 2014

Colour - Birder's Colour Pallet Rev. 1.0

It came as a surprise to discover that the internet lacks any standard colour pallet and appropriate colour nomenclature for birders.  The nearest thing to an ideal nomenclature for our needs is Ridgway's Color Standards and Color Nomenclature (1912).  Sadly, this standard is all but lost and, as of yet, it cannot be digitised properly (see HERE).

sRGB is a colour standard in itself (albeit somewhat limited in terms of our full appreciation of all colours).  With a bit of effort surely there is enough good material online from which to construct a useful colour pallet.  After a few failed attempts I gave up trying to reconstruct Ridgway's colour standard in sRGB colour space but, like a dog with a bone, I have persisted and the result of my efforts is presented below.

First a bit more rationale.  It was while working on the colour profile methodology (HERE) that the realisation struck me that Ridgway and similar colour standards are not suited at all for field birding and field-based digital image work.  Though the means exists to cater for changing ambient light temperature and some in-camera variables (X-rite colorchecker passport and grey card), the variability of ambient light intensity, camera exposure accuracy, and camera dynamic range all continue to play havoc with accurate colour capture from nature.  

The Eureka Moment!

Why bother with luminance in our chosen colour pallet?  Ridgway's plates  consist of just 36 hues and just 5 saturation settings.  That is just 180 colours to start with!  The rest (nearly 1,000 colours) are simply the same 180 colours at different luminance settings.  Ridgway saw a benefit to having these in his standard in that they might allow for an appreciation of the pattern of natural wear and fading in nature.  While this is no doubt of value under ideal, neutral, diffuse lighting in a museum or study setting, it is all but impossible to study and capture these things accurately in nature.  My philosophy is simple, let's do the best we can with what we have and discard anything too cumbersome or likely to discourage the average birder.

Methodology

I have maximized the scope and simplicity of sRGB by selecting 10% of its hues, each hue spaced evenly apart, i.e. 24 hues, starting at hue 0 (red) followed by hue 10, 20, 30 etc.  Next I selected a number of saturation increments, 3 is enough I think - 100%, 66% and 33% saturation (note: 0% saturation = grey scale).  I now have a grid consisting of just 72 colours, evenly spaced across sRGB colour space.

Finally, the hard, but a most enjoyable part, I have researched the web to find and name each if these 72 colours based on some of the most recognisable colour nomenclature used in birding.  And here is the result.



How this standard works is quite simple.  First lets assume a photograph of a bird is already properly calibrated (DNG profile and white balance steps are accurately completed - see HERE).  Lets take a patch of colour from the photo (using the colour sampling technique - see HERE).  Correctly naming this colour swatch using this standard simply requires it's hue, saturation and luminance values.  The naming protocol needs to be ironed out - suggestions welcome.  Right now I am going with a double-barrel combination of two colours if the colour swatch happens to fall right in the middle.  If it is reasonably close to just one of the named colours (say within 3 hue increments and 10 saturation increments) it will be given that name.  Luminance can either be ignored (eg. in the case of a badly over or under-exposed image) or it might be referred to in broad terms, i.e. "light" or "dark".   

Armed with that information, here are a couple of examples.  The colour swatch below is from a desert species.  It falls roughly midway between cinnamon and ochre so the colour is cinnamon-ochre.  It also has a very high luminance. The colour is therefore called Light Cinnamon-Ochre (alternatively Light Tawny-Ochre as both Cinnamon and Tawny are considered interchangeable in this standard).



We have to accept that this is only an approximation but that isn't the main point of the exercise.  We now have a standard to use for discussing and debating colours.  The colour nomenclature is relevant to birding.  The method is easy to use, easily replicated on any computer, and available to everyone.  Last but not least, there is no requirement to calibrate a monitor or even to have proper colour vision to be able to use the method as intended.

Some birds are very aptly named...


I hope it becomes a useful standard and, as always I appreciate feedback.

UPDATE MARCH 2015

A bit more clarity added in the posting linked above.



Wednesday 14 May 2014

Colour - Colour Profiling from Digital Images - A Summary

Summary

Here is a quick summary of my earlier posting.

Challenge: Cameras and computers don't display colour as the human eye sees it.
Solution: None.  We must accept that digital images are only representative and work with that.

Challenge: No two cameras are the same.
Solution: X-rite colourchecker passport DNG profiles will iron out the differences between cameras.
Check out this excellent presentation by Andrew Rodney HERE

Challenge: DNG profiles must be created for different illuminants.
Solution: Correct.  We only need one DNG profile for outdoor use - a Daylight DNG Profile.
Scroll through to point 9 mins. 30 secs. on Andrew Rodney's excellent presentation HERE

Challenge: Birds are highly variable in colour due to a range of factors.
Solution: Very true.  One of the possible uses of this tool would be to try and make sense of the variation within different species, in different places and at different times of year.

Challenge: White balance.
Solution: Grey card.

Challenge: Multiple illuminants in the same environment (shade, foliage, water, flowers, etc.)
Solution: Must be avoided.  Try and photograph out in the open and not under foliage or exposed to other illuminants.  If photographing a bird in shade ensure the grey card is also in shade.

Challenge: Camera exposure.
Solution: This is the big one.  The X-rite colourchecker passport takes care of hue but not saturation and luminosity.  My solution is new.  Create a template for a species/complex with pre-set saturation and luminosity settings.  Paint the template in the correct hues.  The results is not a photograph.  It is a Colour Profile (CP).

Here is an example of a CP template for Chiffchaffs.




To use the Chiffchaff profile apply the following steps:-

(1) Create a daylight DNG Profile for your camera setup by photographing an X-rite colorchecker outdoors in daylight and create the profile using the X-rite software available online.

(2) Bring a Grey Card with you into the field and photograph under the same lighting conditions as your subject (it doesn't have to be in the same frame, thankfully).  Shoot in RAW.

(3) At home apply the DNG profile and white balance correction to your RAW images and save them as PNG files.

(4) Download the Chiffchaff CP from this blog HERE.

(5) Open up the Chiffchaff CP file and a number of your images in your image editing software.  For simplicity I use MS Paint.

(6) For each of the 9 sample points in the Chiffchaff CP sample an equivalent point from your Chiffchaff image(s).  You might need to sample from more than one image to get them all, depending on how cooperative the subject was.  Most important point is to ensure the point is reasonably properly exposed and not in shade or under a colour cast eg. from nearby foliage.

(7) To sample a point use the sampling technique HERE.  I recommend using MS Office postarizing tool "Cutout".  I can't guarantee other postarizing tools will achieve the same result.  Please get in touch if you intend using a different postarizing software tool so we can quickly validate that one against Cutout.

(8) Having sampled and postarised a sample paste it into the box in the Chiffchaff CP template coinciding with the sample location.  Next select the pasted sample with the sample dropper and open up the "Edit Colours" window.  Here note down the Hue number.  Next change the Saturation and Luminance numbers to those pre-defined in the Chiffchaff CP template.  Last step is to select the bucket fill tool and fill the box.  The colour will fill the box plus the relevant parts of the bird.  Repeat for the other 8 sample points.

(9)  Last step is to fill in details about the bird, date, location etc. plus validation information including the DNG profile and Grey Card.

Tuesday 13 May 2014

Colour - Chiffchaff Colour Profile (CP)

A Compromise 

Hopefully the image below looks a bit like a collybita Chiffchaff (Phylloscopus c. collybita).  It is in fact a colour composite created using a new conceptual template which I call a Colour Profile (CP).


Why?

As outlined in my introduction to colour profiling (HERE) there are many challenges to consistently presenting and describing colours from digital bird images.  While it is possible to reasonably accurately and consistently capture the hue of colours in nature, capturing the saturation and luminosity of colours accurately using a digital camera is confounded by the difficulty in obtaining an accurate exposure.  By creating this CP template I have eliminated the need to capture saturation and luminosity at all because these are pre-sets provided for in the template.

As to the broader question of why we might need or use these templates in the first place, I specifically chose Chiffchaffs as my first template because of the ongoing interest in rare Chiffchaff taxa in Western Europe.  A complete understanding of the colour variation within this group is still quite elusive it seems, such that some of the very pale Chiffchaffs which until recently were thought to be possibly P. c. abietinus were found to be genetically P. c. tristis yet we know that many birds giving the classic tristis call are often mid-brown in colour and not especially pale.  Obvious questions that need answering include, how variable are tristis?  Are abietinus occurring in Ireland and Britain and how do we distinguish abietinus from tristis and collybita in the field?  Could there be other hidden taxa involved?  This tool might help us figure some of these things out.  

Many interesting online discussions about colour in Chiffchaffs appear to break down when the question of camera exposure and white balance comes up.  The intention of this template is to apply a standard, reliable methodology and help get past the photographic questions to start to catalogue and organize digital images of vagrant Chiffchaffs into some kind of order mainly based on colour hues.  

How?

Here are the pre-requisites required to use the CP template to create a colour profile for a specific Chiffchaff.

(1)    You must use an X-rite Colorchecker Passport to create a DNG profile for your camera.  The template won’t work reliably without one.  The main problem is cost – roughly €100/$100 for the X-rite.  Note the software used to create DNG profiles is available free from the X-rite website so all you may really need to do is borrow an X-rite colour checker passport for a few minutes to make images of it with your camera and later create your own DNG profiles back at home.  5 birders interested in this project, €20 each – if you catch my drift?

X-rite Colorchecker Passport (formerly Gretagmacbeth) with 
Dan Zetterström's Chiffchaff plate from the large format Collins Bird Guide, 2nd edition.

(2)    Anyone going into the field to take photos of a Chiffchaff with the intention of creating a colour profile will need to bring with them a white balance card (called a grey card).  The X-rite colorchecker passport comes with one but they can be purchased separately and much more cheaply than the full passport.  X-rite is just one manufacturer of grey cards.  The make should not matter provided it is actually properly neutral.  Beware grey cards that are actually not neutral at all.  If you have purchased a grey card and you want to verify it is neutral take a picture of it with an X-rite grey card along side (if you can borrow one).  Correct the white balance in the image using both and there should be no difference in colour temperature recorded.  If there is a difference then one or other of the grey cards is a dud!  Lastly, all pigments are subject to deterioration over time.  All calibration cards will need replacement every few years.  This applies equally to the colour checker as it applies to the grey card.  X-rite recommend replacement every couple of years but it all depends on usage I guess.

X-rite Grey Card with Dan Zetterström's
Chiffchaff plate from the large format Collins Bird Guide, 2nd edition.
Note how similar this image and the one above are.  
All that was required to process these images was the sampling of the white balance 
patch and the selection of the correct DNG profile.  
With two mouse clicks both images were perfectly colour calibrated.
Reason enough to purchase and use this product I think. 

(3)    Your camera must be able to shoot RAW format files.  You need RAW anyway to create a DNG profile and you also need to be using Camera RAW or an equivalent software package to apply the DNG profile to your images.  DNG profiles can't be applied to JPEGs.  It is also the best place to apply your white balance correction to all your images in one go.  Warning, JPEGs are compressed files – colours are also compressed by the process so there is a potential loss of colour accuracy with the switch to JPEG.  Try to work with PNG images rather than JPEGs to avoid the problem of lossy compression.

(4)    In the field you must ensure that your subject is out in the open and not under a foliage canopy or near some other source of “unnatural” illumination.  The only way of properly sampling the colour is by illumination from day light.  A bird sitting in the shade on a sunny day is fine too so long as the grey card is also positioned in the shade for reasons explained HERE.  And, as stated, getting an accurate white balance is essential – don’t forget to take photos of the grey card, positioned appropriately during your session.  If you have forgotten to photograph the grey card or the light has shifted during the shoot you are back to square one.

(5)    So you arrive back home having successfully photographed a confiding Chiffchaff in RAW format files and you have a good, accurate white balance image in the mix as well.  Happy Days!  Open up your images in Camera Raw or a similar raw editing software package and select a white balance standard image plus a series of good, representative shots of the Chiffchaff.  First white balance all the images by reference to the image you took of the grey card, then apply the DNG profile correction to each image on top of that.  You now should have a properly colour-standardised set of images of the Chiffchaff which you photographed.  The hues in your images should be consistent and very accurate to a professional standard.

(6)    Open up these images from Camera Raw in Photoshop Elements or whatever software you use and save them as PNG files without any further correction (not JPEG files as explained above).

(7)    Next open up the Chiffchaff CP template in MS Paint, Photoshop or whatever image editing software you have.  Also separately open up one or more of the properly colour-standardised PNG images of your Chiffchaff.  Now it is time to sample the colour from your digital images and translate those colours to the CP.  Start perhaps by sampling a properly exposed area of the bird’s mantle (refer to my sampling technique HERE).  A properly exposed patch of colour is one which is well illuminated and is neither under nor over-exposed.  Avoid sampling an area which was obviously in shade (eg. the rear mantle, shaded by the wing tip) - it’s white-balance and therefore its hue may not be correct.  Don’t forget to postarize the colour sample then paste the postarized sample into the box linked to the mantle.  Sample the colour with the dropper.  I recommend using the MS Office postarizing tool called Cutout for now if you can access it.  I will need to validate other postarizing tools against Cutout to make sure they all perform the same and arrive at the same colour hue.  If anyone intends using a different postarizing tool and want to validate it please get in touch.

(8)    The next bit is vital, you must change the saturation and luminosity settings for that colour to those listed in the template.  Having set these fill the sample and the rest of the box with the new colour.  I have set up the template in such a way that it will “pump” the colour around the template so to speak and fill in the necessary areas of the bird automatically. 

(9)    You will note that I have left two optional saturation and luminosity pre-sets for the mantle colour.  One is based on the typical characteristics of a mid-toned collybita-type image.  The other suits a typically pale tristis-type.  This is where I have allowed for a small bit of subjectivity.  The fact is that we often encounter birds which appear in life to be exceptionally pale and ghostly but this doesn’t translate well in photographs due to camera exposure and perhaps also dynamic range limitations.  Here is an opportunity to restore some of the ghostly distinctiveness of a bird without compromising the CP template too much.  Note I have also done this for other sample points within the template. 

(10) In the case of the fringes of the remiges and rectrics I have removed the option to sample and paint these parts of the template and simply given some pre-set options for hue, saturation and luminosity.  The reason for this is that it is extremely difficult to accurately sample colours based on so few pixels or micro-structures as one might call them - like these feather fringes for example.  I don’t think this is too much of an issue here as I don’t think the exact colour of these fringes is critical for identification.  I’m open to correction on this point of course and can add more hue options here if required.

(11) Having worked through all the sampling points and painted your CP template the work is done.  I would recommend filling in the hue boxes with text just to confirm the values you arrived at.  Should the profile ever be saved as a JPEG or some other lossy file by accident the actual colours may deteriorate as a result.  At least if the hues are properly recorded in text in the image it would be easy to recreate the image exactly as before using the original template.  I think it is important to add some additional text in the notes section provided including the date and location the bird was photographed plus a record of any calls heard, unusual plumage/colour characteristics noted etc.  Also important to add validation information including details of the DNG profile and grey card used.  When it comes to maintaining files for verification purposes later I would recommend storing the CP together with the RAW and PNG images of the bird used to make the profile including the white balance image used.  

Validation of the method and template

This concept is in its infancy.  I am hoping that the validation of this method will come about through the trial and error of myself and others interested in having a go at using it.  I will also set up a validation page on the blog and populate it with various information as I gather it.

(1)    I am confident that the colour standardisation method is excellent for providing accurate and consistent hue values – professional photographers who use X-rite colorchecker passport can’t be wrong.  Remember we are not talking about recorded colours as the human eye sees them.  This is about using a benchmark to be able to study colour variation with the help of a professional standard.

(2)    By taking saturation and luminosity out of the equation we no longer need to worry about these scales.  It makes the task of creating the colour profile so much simpler.

(3)    I am conscious that the template itself needs some work.  I have created REV. 1.0 and put it out without much fine-tuning for some comment and assessment rather than continuing to work on it.  I would ask those who are far more familiar with Chiffchaff taxonomy to take a look at the sampling points I have chosen and advise me if I need more (or less).  For instance, is it enough just to have two options for Hue/Sat/Lum of the fringes of the rectrices and remiges.  Should I split the primary fringes from the secondary and/or tail fringes to have the option to colour these all separately or is it enough to have all of these lumped together as one sample point?  Should there be more (or fewer) options for colour the supercillium or greater coverts bar for instance.  Is leg colour relevant/necessary?  That said, this needs to be as simple a tool as possible to work with.  Otherwise birders will not bother with it.  I would hope that an average person, reasonably familiar with the tool should be able to process images of an individual Chiffchaff and create a finished colour profile all in under 10 minutes.

(4)    I have intentionally blurred the wing formula as it is not relevant to this profile.  This is all about colours, not structural morphology.  Whether the wing formula is accurate or not is irrelevant.  As stated above, the image at the top of this post may look like a collybita Chiffchaff but it is merely a representation of a Chiffchaff type.  For that reason, I hope that this will be just as useful a template for looking at Iberian, Canary Islands or Caucasian Mountain Chiffchaff colour profiling as it is for Common Chiffchaff taxa.  Willow Warbler and other less closely-related species would be a stretch for this template.

(5)    Next I hope to create a template for Lesser Whitethroat taxa, another remarkably challenging and interesting complex with various taxa reaching Western Europe.

(6)    Lastly just to point out that I would hope to have a final Chiffchaff template complete before September 2014 so that it can be used without further modification this coming autumn and winter when hopefully there will be plenty of rare Chiffchaffs about to sample and profile.  If you have an interest in getting involved with this project please contact me.  Hopefully I can help you get your DNG profile set up and you can be practising using your grey card and having a go using the Chiffchaff CP in its current draft.  For those with an intimate knowledge of Chiffchaff taxonomy I would love to hear what you think of the CP template and what tweaks if any you think it needs.

(7)    Finally, if there are any birders in North America interested in creating a CP for a species or complex from that continent please get in touch.  I might be able to help.

Download

    The template is available for download and use at the link below.  If you are using it I would greatly appreciate your feedback.

Chiffchaff CP Rev 1.0 (c) Mike O'Keeffe


Update March 2015

It has taken me a long time to connect with a tristis-type Chiffchaff.  HERE is a spring comparison of a collybita versus a pale tristis-type.







Colour - Profiling colours from Digital Images

A New Approach


As part of this ongoing project looking into various aspects of bird ID from digital images I have arrived at the prickly question of colour sampling and profiling colours from digital images.  I would have been sceptical about this area beforehand but from my research I now believe it is actually possible to describe colours very accurately, objectively and meaningfully from digital photos.  This could be really useful for forums and hopefully for biological research generally.  

The real beauty of the method proposed below is that a colour blind observer should be able to produce the same results as an observer with perfect colour vision.  There isn't even a requirement to colour calibrate a monitor.  All the calibration is done at the camera end.

Often when there is a discussion involving the subtle colours of a bird in a digital image there is more heat than light generated as it is far too easy to dismiss an argument around a bird on the grounds that the images may not have been properly exposed or white balanced.

What are the challenges to be overcome?

(1)    Digital cameras don’t see the world of colour as we do and possibly never will.  We have to accept therefore that the colours which we describe in digital images are not the actual colours as we see them but a representation of them.  In much the same way, a sound recording is not 100% identical to the sound audible to the human ear, it is still nonetheless a useful representation of it and a properly recorded sound recording can be compared with others recorded in the same way to develop audio profiles for different species and taxa.  As the work of The Sound Approach has shown, this method is now paying dividends in terms of new species discoveries and changes to taxonomy.  Could colour profiling help improve our knowledge of different species, taxa and clines?

(2)    No two cameras are the same.  Individual cameras, even of the same brand and model will vary in terms of how colours are recorded and presented.  This can however be relatively easily (if somewhat expensively) overcome using a professional photographer’s tool called the X-rite Colorchecker passport (formerly Gretagmacbeth).  This handy tool creates camera calibrations (called DNG profiles) under different lighting conditions and effectively calibrates the camera and its images to a professional standard.  It would be nice if all camera manufacturers did this work for us automatically beforehand.  For now it remains the responsibility of the camera owner to properly calibrate their camera kit for the lighting conditions in which they use it.

(3)    Colour in birds is quite variable and subject to many environmental factors including diet, health, metabolism, fading due to strong light and and feather wear and abrasion.  There are also certain structural colours (eg. iridescence) where the colours will in fact change relative to the angle of the observer and the light source, so in the case of iridescence, there is no one true colour of an iridescent feather.  While this is all true I think most experienced observers would know that in many cases there is a high level of consistency involved in how colours are displayed by birds.  What we currently lack is a standard method (in the field at least) to study and understand where and when colour can be used for ID and when it is best to avoid getting too involved with the colours of a particular bird in a particular setting due to the natural deterioration of it's plumage.  The purpose of an objective method to study and discuss colour is to use this tool to peer into this area.  Obviously ringers (banders) and those that regularly study museum specimens will have a much greater understanding of the usefulness (or lack) of colour for ID of different species and taxa.  I certainly don’t claim expertise in this area as I neither ring birds nor regularly visit museum collections.  I am merely proposing a method here to allow field birders and photographers to start to try and make sense of colour based on images obtained in the field.  I would however love to hear from ringers and those who have had the fortune of studying a great many museum specimens (taxonomists and illustrators especially) to see what they make of this new method. 

(4)    White balance is notoriously difficult to get right.  Ambient lighting changes constantly depending on the time of day and whether conditions are sunny, cloudy or the subject is in shade.  Again there is a handy tool to overcome this and manage white balance accuracy.  It is called a White Balance or Grey Card and is simply a neutral grey patch which the photographer takes a photograph of during the course of a photograph session and then later uses it to white balance all the images captured under those lighting conditions.  The grey card and subject don’t have to be in the same scene, the card just has to be photographed at around the same time under the same lighting conditions as the subject.  Note it is important to be mindful of a bird which is in shade on a sunny day.  The grey card must be positioned in the shade also.  The colour temperature of light in a shaded area on a sunny day is not the same as that under an overcast sky though they might appear at first to be quite similar.  Clouds scatter all wavelengths of light so the light from under a cloud canopy is still basically white even in the shadows.  In the shade on a sunny day there is no scattering of white light taking place.  Sunlight is polarized.  The only light source entering the shadows is in fact the sky canopy illumination itself which is blue scattered light.  A grey card can still correct for this provided the card is positioned in the shade, preferably where the subject is/was and not out in the open where it is being illuminated by direct sunlight.

(5)    What about multiple light sources?  If a bird is under a foliage canopy, or feeding among brightly coloured flowers, or drinking at the water’s edge, or near some other highly reflective surface these nearby surfaces will all effectively become new light sources - casting colours onto the subject.  This can dramatically influence how the subject appears both to the eye and to the camera.  The classic example is trying to gauge the accurate colour of a Phylloscopus warbler or Vireo sp. under a canopy of foliage – it may be extremely difficult to do so accurately due to the green colour cast by the leaves.  Removing this unwanted colour cast can be challenging, though not impossible - see HERE.  Ideally the bird should be properly out in the open when photographed, illuminated directly by the sun or under the diffuse light of a cloudy sky.  Shade is okay too provided the grey card is used correctly for shaded circumstances.

Note: the combination of the colour checker passport and grey card provide an extremely accurate and consistent measure of hue (+/- 2 increments on the sRGB colour hue scale by my current reckoning).

(6)    Accurate camera exposure is very challenging.  In any typical series of images taken of a bird there will be some images (and areas within images) which will be over-exposed and/or under-exposed.  This obviously affects the colours captured and presented.  Hue is just one of three scales needed to measure colour accurately.  The other two, Saturation and Luminance, are both affected by exposure.  Colours de-saturate as images are under or over-exposed.  There is also of course a direct relationship between exposure and luminance.  Underexposure reduces the luminance of colours while over-exposure increases the luminance.  Luminance is also dependent on the angle of incident light relative to the observer / camera so as everyone knows when you look at a bird you don’t see a flat subject with uniform colours but rather a 3D subject with gradients of colour from highlights to mid-tones to shadows.  Gauging the mid-point – the correct luminance of a colour is difficult under those circumstances.  Lastly, as if this wasn’t already complex enough, we have Dynamic Range – the limited range a camera has to accommodate areas within a scene which have vastly different luminance values.  For instance, when you are taking a photograph indoors and a window is within the photo, the camera will either over-expose the scene outside the window and subjects in the room will appear normally exposed or visa versa - the outdoor scene will look normal, while the indoor scene will be darkened and lacking in detail.  The human eye appears to work better than any camera or video recorder in being able to quickly compensate for these changes in luminance.  But, then again we must remember that the human eye is only ever focused and observing a small area of a scene and doesn’t need to look at a subject indoors at the same instant as it observes a subject outside.  The iris adjusts dynamically to whatever the person is looking at at any given instant.  A single camera image is not a dynamic environment.  It is fixed both in space and time so is not directly comparable with how the human eye works dynamically.  In fact it has been suggested that a modern digital camera  performs nearly as well as the human eye does in terms of dynamic range in an instant of time.  Video cameras produce dynamic video but are not as quick as the human eye in adapting to sudden changes in luminance, at least not yet. 

Challenge number 6 would at first appear to be insurmountable for anyone hoping to sample colour accurately and consistently, but does it need to be a challenge at all?  How important are saturation and luminance in the context of bird ID?  Or, to put it another way.  What would happen if we decided that we could fix saturation and luminance settings and concern ourselves purely with measuring hue.  Herein lies my discovery and new technique.

I have created a Colour Profile (CP) template for Chiffchaff HERE .  How it works is the observer calibrates their camera (DNG profile) then goes into the field armed with their DSLR camera set to shoot in RAW format, plus a grey card.  The final ingredients are a cooperative Chiffchaff and plenty of patience!

Wednesday 7 May 2014

Colour - Standards and Nomenclature

A colour nomenclature for birders


I have spent a number of weeks researching and attempting (ultimately unsuccessfully) to bring an old colour standard and nomenclature to life on the internet.  Note I removed earlier postings on the topic from the blog to avoid confusion.  Here is a brief summary.

Robert Ridgway's Color Standards and Color Nomenclature (1912) is, on the face of it, the most appropriate colour standard for birding.  It was produced with naturalists, particularly ornithologists in mind.  Unfortunately the standard was not well preserved and as Hamly (1949) illustrated the standard has been in decline for many years.



Firstly on the point of nomenclature.  Why do we need it?  The MUNSELL COLOUR STANDARD is one of the most well known and successful colour standards yet the colours within it are not specifically named.  This is a reasonable point, however on the other hand a great many bird names contain reference to unusual colours which I would guess most birders would struggle to accurately describe.  

Take for example colours like buff, ferruginous, russet, tawny, cerulean and vinaceous.  Ridgway used these terms in his colour standard.  If we all had access to the standard we could begin to reflect upon and learn these colours, and they would hopefully return to common usage, which is the true essence of any standard.

Unfortunately, after numerous attempts I decided to call a halt to my efforts to bring Ridgway's colours to the internet.  Ultimately, my undoing came with the realisation that Ridgway's standard was too broad to fit within the constraints of the internet colour pallet (termed sRGB colour space).

This brings me to the question of the colour standard which we adopt.  Here, online we have no choice.  The colour standard for the foreseeable future is sRGB.  So, what does that mean in reality.  For most, practical purposes we are not too constrained by this standard.  Our cameras, printers and scanners all use this standard so we have continuity.  Some, mainly highly saturated colours cannot be displayed on the internet but when it comes to ID purposes that may not be too big an issue.  The elephant in the room of course is the great difficulty involved in accurately capturing and maintaining colour accuracy from life, through the lens, sensor, processor and internet to your computer screen and mine.  If we are going to benefit at all from the internet we need to work smart, work hard, and compromise a bit. I am currently working on a project to try and develop a method to colour code Chiffchaff images which will hopefully tie all of my efforts on colour together.

Reference
Hamly D. H. 1949.  The Ridgway Color Standards with a Munsell Notation Key.  Journal of the Optical Society of America.  Vol. 39, No. 7