This is a very interesting question with lots of possible answers.

For a measurable approach to detecting colors in bird images, consider

M. Das, R. Manmatha, Automatic segmentation and indexing of specialized databases, University of Massachusetts:

http://ciir.cs.umass.edu/pubfiles/mm-37.pdf

See Fig. 1 (images in bird database) and introduction to the use of segmentation to determine the colors of birds, correctly.   See also the attached image with examples of bird images, in addition to the excellent example given for this thread.   The basic approach to the detection of foreground colors combined with edge information is given in Section 3, starting on page 9.   An overview of the segmentation strategy is given in Fig. 2, page 11.

Thanks James,

I read the attached paper, although I think it is a long way to go for me who doesn't have experience with image processing. 

Note: I have edited my question slightly, to make it clearer. Initially, it may have seem that I looked for an automated tool, but in this case I'm only interested in this individual bird at the given images. 

Well, I agree with James F Peters, there are several possibilities to work with that.

I'd would suggest ImageJ program to look at the intensity of the colours. It is a free open source code.

A tip, I'm not sure whether you know that, but you need to have a image of each of the birds do you wanna automatically identify, like: male or female, and then you can use the standard pics to compare with new ones... or maybe you should mark places around the eyes for instance and the colour you wanna find there. There are a lot of codes for eyes identification ... this may help too. Nice work!

Good luck!

Hi Telma, thanks for reply. I see that my question was not clear a enough and it could seem that I looked for a tool for automatic detection, I apologize this and I edited the question to make it clearer.

In this case I'm not after automatically identification, I'm only interested in the one individual bird at the images.

Learning ImageJ is on my list:)

Ronny, Life is very interesting because I saw the answear you need yesterday in a documentary.  First, the colour is in your brain. I am going to recover the documentary and I am going to send you some information that will help you. At same time I am going to study the subject, because I have a lot of bird photos and I believe to have seen the answer for your question. Wait a moment, please, because

I will be back.

José Walter

Ronny, I could not wait, although is 4:00 AM. First, go to youtube and find "The beauty of diagram" number 3: Newton. Close to the end there is an experiment that is the key for your answer. If you agree with that idea, that in the documentary is very simple for the general people, the next step is to go to laboratory for some experiments with light. Let me know what do you think. I am not sure, but in the Physics Department of my University, we have equipments to study these colours ( I am not sure yet, but the goal are the pigments).

Some photos I will be send to you later, this week.

If your photos are cinfined as the offerd above, yellow-greenish color, blue/grey color, etc., the approach based on "scan line algorithm" for you may be a feasible method.

José, thanks for the interesting documentary. It truly illustrates that our color perception could play us a trick. Looking forward to hear more about your thoughts. 

Jianhua, thanks for the suggested approach, I will read about this method. The photos of interest is of the one individual bird given above (except the two last images which is for comparison). 

Dear Ronny,

Don't know if you still need some help with this. If that is the case, what it can be done is to look at the reflectance pattern of the specific part you want to look at to identify your bird to one or another morph. This is not going to be a correct way to do so and I cannot promise you a definite answer, because you don't seem to have a reference color to correct the color/photo from, but I think it might be possible to get to the detail you need to classify the bird. 

If you don't know how to do this, send me an email to my private address and I can work it out for you and let you know what I think after having a look at the reflectance patterns of the eye ring and mandible. 

Cheers,

J

Thanks Martinez-Padilla, I have know added a image of the parental female, which has a has blue/grey color instead of yellow-greenish. Which could be used as reference color of blue/grey.

Hi, thanks for that.

If you want me to look at the reference pattern, I'd need a photo with a bit more of resolution than the one I can see here. Could you please, send me to my email address the best photo you have with the highest resolution of the specific bird to get the reflectance pattern for you?

Cheers,

J

OK, I got it now. I'll get back to you as soon as I can.

Cheers,

J

Before talking about the colors in digital images, you have to have careful calibration of the sensor (camera) - just to compare apples to apples. Most cameras nowadays are way too clever for their (and our) own good. They do all sorts of weird preprocessing. Colorimetric calibration is absolutely essential.

Thanks Yuri for advice. The camera used is a wildlife camera trap, Reconyx HC900. What is needed to perform a colorimetric calibration?

@Ronny. There is an extensive literature on the subject. See for example,  Thomas W. Pike, Using digital cameras to investigate animal colouration: estimating sensor sensitivity functions. Behav Ecol Sociobiol (2011) 65:849–858. DOI 10.1007/s00265-010-1097-7;

Derya Akkaynak et al., Use of commercial off-the-shelf digital cameras for scientific data acquisition and scene-specific color calibration. J. Opt. Soc. Am. A / Vol. 31, No. 2 / February 2014.

That's a good start.

Thanks Yuri, I will read the suggested papers. 

Ronny - It is unclear from your post the need for fully ascertaining the identity of this bird.  Thus the certainty of its identify based on plumage and exposed skin color from the image may be enough.  I am not sure that a color wavelength analysis of a photo of the soft body parts is necessary to assign this bird in question(unless your question is more rigorous and requires concrete evidence).  If the latter is true then you will need to follow some standardized method  as suggested by Yuri Rzhanov.  If we can believe the images and what they show, then the color of the cere and eyering seem consistent with a dark morph bird; as you have suggested.  To my eye at least the cere and eyering approach the color of the exposed legs which have yellow tones. Also, I assume you are comparing the images with those of reference images (even Ristow et al 1998) so relative to those reference images, you have enough to be fairly certain the identity of this bird.   As for an incorrect color (or the image lying).  The images you have attached are actually quite good and the bird is in subtle light and not harsh contrasting light which could alter color perception.  In all images you have posted, the color of the cere and eyering remains consistent, even as the bird changes positions;  so I do not think that the images are in error.  Hope this helps

Thanks, Andrew. For the bird in question, which is a "intruder" at the nest, I just need to verify which color morph it belongs to and wether it is a female or male. And according to the "reference images" of the parental female and male the color reproduction seem to be true. To my eyesight, for almost all images of the "intruder", it has dull yellowish cere and eyering, although at image IMG_3155 the yellow color is stronger. At image IMG_3155 the bird being very close to the camera.