For those new to gestalt and gestalt keys please read the introductory post (HERE) and the essential principles of gestalt keys (HERE) before proceeding.
In the posting on Primary Projection (HERE) I described a useful method for aligning the body and wings properly in side profile. By looking at the bird from a slight elevation the tips of various feathers of the opposing wing can act as a useful guide to line up the body parallel with the plane of the camera.
Of course, having the body in side profile is absolutely no guarantee that the head will be aligned similarly. The head and neck move very independently of the body.
Whereas the wings provided a handy clue to the alignment of the body relative to the camera lens, the head doesn't give up such secrets easily! There is only one structure that we can rely on to definitively say the head is in side profile. That is the bill. Everything in side profile is foreshortened, with the possible exception of the length of the bill measured along the culmen. So how do we confirm that the head and bill are in side profile?
Nares
In some species it may be possible to look straight through the nares (nostrils) when the bill is viewed in perfect side profile. Rather like the sight in a rifle, the narrow bore of the nasal passage restricts our view to within a very narrow angle. So there is little margin for error. This method takes care of the problem of rotation about both the X and Y axes. This makes it one of the more effective tools for judging side profile of the head and bill.
Glaucous Gull (Larus hyperboreus) here and Iceland Gull (Larus glaucoides) are sometimes separated using complex formula involving bill and eye proportions (from New Approach To Identification, Grant and Mullarney). It's useful to be able to confirm that the bill is in perfect side profile by being able to peer through the nares. Note it may be convenient to photograph a bird backlit as it is here, to illuminate the nares chamber and help verify the angle of view.
Eye to Bill Ratio
In the posting reviewing the limitation of this measurement (HERE) I highlighted some of the issues with this commonly-used tool. As with most comparative measurements in common usage the eye and bill don't actually align along the same axis. So we cannot hope to measure an anatomically accurate eye to bill ratio from our images. What we in fact measure is a two dimensional projection of these structures. Rarely is an attempt made to actually define whether it is the eye or the bill that should be in profile for this test. I think it's pretty clear that the bill should get priority as explained above. But very often the images we see tend to show the eye more in profile, because this is more aesthetically pleasing perhaps. Or, because birds, being curious creatures, tend to look at the camera/observer much of the time, which means the image captured is often of the eye in side profile.
For birds which possess a symmetrical bill, and provided one has a good series of shots to work with, it should be possible to establish side profile closely based upon eye to bill ratios. Ideally the available shots should capture a bird in the process of turning it's head through the parallel axis, i.e. from looking slightly towards the camera to looking slightly away (as shown above). Video can be useful for capturing this perfectly. With luck, at least one of the images will show a perfect side profile view.
This method doesn't provide confirmation of a side profile view but it should certainly very closely approach it. Another slight flaw with this method is that it doesn't deal with rotation about the X axis. For example in the top image inset of an Eastern Olivaceous Warbler (Iduna elaeiaca), while the bill can be aligned closely to the X axis, it is clear that the angle of view is from below (we can see part of the underside of the bill for instance). Similarly with this Ring-billed Gull (Larus delawarensis) the bill is rotated slighting towards the camera along the X axis in all images. This is because I was elevated above the bird when I shot these images. So, unlike the Glaucous Gull image above we can't see straight through the nares in this instance. This viewing angle is further confirmed by the fact that we can see the base of the culmen in all images, something which wouldn't necessarily be possible if we were looking at the bird's head perfectly from the side. This doesn't ultimately affect the eye to bill ratio but it may impact on a measurement involving the Y or Z axis (such as those used in the Glaucous V Iceland assessment tool).
Mapping of Features
These methods alone may not give us the required confidence in some cases to take reliable measurements in side profile. Once again we turn to Gestalt Keys. In the introductory postings I showcased a couple of prototype keys illustrating structural angles of the head. With thanks to some valuable input from Pete Dunten I have done further work on the Dowitcher Loral Angle key to improve side profile alignment. It is clear that in order for a feature like loral angle to succeed the head and bill must be in perfect side alignment. The key requires a number of hard and soft loci to achieve this and that indeed may prove to be the case with most gestalt keys involving the head.
I am not quite ready to make this key widely available through the blog as I feel it needs more testing and refinement, but if anyone is keen to have a go using the current version "warts and all", drop me an email and I will send it on to you.
Nares
In some species it may be possible to look straight through the nares (nostrils) when the bill is viewed in perfect side profile. Rather like the sight in a rifle, the narrow bore of the nasal passage restricts our view to within a very narrow angle. So there is little margin for error. This method takes care of the problem of rotation about both the X and Y axes. This makes it one of the more effective tools for judging side profile of the head and bill.
Glaucous Gull (Larus hyperboreus) here and Iceland Gull (Larus glaucoides) are sometimes separated using complex formula involving bill and eye proportions (from New Approach To Identification, Grant and Mullarney). It's useful to be able to confirm that the bill is in perfect side profile by being able to peer through the nares. Note it may be convenient to photograph a bird backlit as it is here, to illuminate the nares chamber and help verify the angle of view.
Eye to Bill Ratio
In the posting reviewing the limitation of this measurement (HERE) I highlighted some of the issues with this commonly-used tool. As with most comparative measurements in common usage the eye and bill don't actually align along the same axis. So we cannot hope to measure an anatomically accurate eye to bill ratio from our images. What we in fact measure is a two dimensional projection of these structures. Rarely is an attempt made to actually define whether it is the eye or the bill that should be in profile for this test. I think it's pretty clear that the bill should get priority as explained above. But very often the images we see tend to show the eye more in profile, because this is more aesthetically pleasing perhaps. Or, because birds, being curious creatures, tend to look at the camera/observer much of the time, which means the image captured is often of the eye in side profile.
For birds which possess a symmetrical bill, and provided one has a good series of shots to work with, it should be possible to establish side profile closely based upon eye to bill ratios. Ideally the available shots should capture a bird in the process of turning it's head through the parallel axis, i.e. from looking slightly towards the camera to looking slightly away (as shown above). Video can be useful for capturing this perfectly. With luck, at least one of the images will show a perfect side profile view.
This method doesn't provide confirmation of a side profile view but it should certainly very closely approach it. Another slight flaw with this method is that it doesn't deal with rotation about the X axis. For example in the top image inset of an Eastern Olivaceous Warbler (Iduna elaeiaca), while the bill can be aligned closely to the X axis, it is clear that the angle of view is from below (we can see part of the underside of the bill for instance). Similarly with this Ring-billed Gull (Larus delawarensis) the bill is rotated slighting towards the camera along the X axis in all images. This is because I was elevated above the bird when I shot these images. So, unlike the Glaucous Gull image above we can't see straight through the nares in this instance. This viewing angle is further confirmed by the fact that we can see the base of the culmen in all images, something which wouldn't necessarily be possible if we were looking at the bird's head perfectly from the side. This doesn't ultimately affect the eye to bill ratio but it may impact on a measurement involving the Y or Z axis (such as those used in the Glaucous V Iceland assessment tool).
Mapping of Features
These methods alone may not give us the required confidence in some cases to take reliable measurements in side profile. Once again we turn to Gestalt Keys. In the introductory postings I showcased a couple of prototype keys illustrating structural angles of the head. With thanks to some valuable input from Pete Dunten I have done further work on the Dowitcher Loral Angle key to improve side profile alignment. It is clear that in order for a feature like loral angle to succeed the head and bill must be in perfect side alignment. The key requires a number of hard and soft loci to achieve this and that indeed may prove to be the case with most gestalt keys involving the head.
I am not quite ready to make this key widely available through the blog as I feel it needs more testing and refinement, but if anyone is keen to have a go using the current version "warts and all", drop me an email and I will send it on to you.
No comments:
Post a Comment