Capture
How to tell a usable hair photo from a bad one
There are two kinds of bad hair photo. One is bad on its own, and software can catch it. The other is fine on its own and useless next to your last one — and nothing can catch that but you.
The fifteen-second check
Run this before you accept any photo. It is faster to retake now than to spend a month wondering whether a reading was real.
-
Can you see the whole boundary, edge to edge?
If you are measuring a hairline, the entire border must be in frame including both temple corners. A border that runs off the edge of the photo cannot be measured; it can only be guessed at.
-
Is anything covering it?
Hair swept forward, a fringe, a hat, a hood, a hand, a headband. Any of them and you are photographing the object, not the zone.
-
Is there a hard shadow crossing the zone?
The one people miss. A shadow edge running across the scalp reads exactly like a boundary between hair and no hair — to software and, honestly, to you. Look specifically for a line that is straight or that continues off the head onto the background; hair does not make those.
-
Is it actually sharp?
Zoom all the way in on the boundary. If you cannot resolve individual strands, nothing downstream of this photo is real. Phone screens are generous — a photo that looks fine at thumbnail size routinely falls apart at 100%.
-
Is anything blown out or crushed?
Blown-out highlights on the scalp turn into pure white with no detail, which looks like bare skin whether or not it is. Crushed blacks in dark hair turn a textured area into a solid shape. Both destroy information, and neither can be recovered afterwards.
-
Is the head level and facing the lens?
Chin neutral, eyes to the lens, head not rotated. Small rotations systematically favour one temple over the other, which then shows up as the two corners disagreeing.
-
Is it the same as last time?
Same light, same distance, same hair state, same framing. This one is not about the photo's quality at all, which is why it is last and why it is the one that actually decides whether you learn anything.
The two kinds of failure
Steps 1 to 6 are about whether the photo contains the information. Step 7 is about whether the photo is comparable. They fail in completely different ways and only one of them is detectable by a machine.
Too dark. Out of focus. No head in frame. Zone covered or out of shot. Blown highlights. These are properties of a single image, so software can look at that image and refuse it — which is what a good app should do rather than scoring it anyway.
A perfectly exposed, perfectly sharp, well-framed photo taken under a different lamp than last month's, with your chin two degrees lower and your hair damp. Nothing is wrong with it. It is simply not the same measurement, and no amount of analysis of the image alone can know that.
The second kind is where nearly all confusing results come from. People assume a surprising number means the software made a mistake, when far more often the two photographs were never comparable to begin with. That mechanism has its own page, because it is the single most useful thing to understand about tracking anything from photographs.
What Crown checks, and where the checks stop
Two gates run on your phone before a photo is uploaded at all: one rejects a frame too dark to read, one rejects a frame with no head in it. Then, per zone, Crown estimates how clearly that zone is exposed and refuses to score any zone that does not clear the bar — reporting instead which of five things went wrong (covered by hair, headwear, wrong angle, blur, or not pictured) and the angle that would fix it.
Every one of those is a check on a single image. None of them, and none in any other app, can tell you that today's perfectly good photo is not comparable to last month's perfectly good photo. That is step 7, and step 7 is yours.
The shortest version
If you would not be able to point at the boundary with your finger and be confident where it is, neither can anything else. Retake it.
Common questions
What makes a hair photo unusable?
Most commonly: a hard shadow crossing the zone, the boundary running off the edge of the frame, hair or headwear covering it, softness that hides individual strands, or blown-out highlights on the scalp. Any of these removes the information being measured.
Why does a shadow ruin a hairline photo?
Because a shadow edge across the scalp looks exactly like the boundary between hair and bare skin. A straight edge, or one that continues off the head onto the background, is a shadow — hair does not produce those.
Can an app tell if my photo is bad?
It can catch problems within a single image — darkness, blur, no head in frame, a covered zone. It cannot tell that today's good photo was taken under different lighting than last month's good photo, and that is the failure that causes most confusing results.
How sharp does the photo need to be?
Sharp enough to resolve individual strands when you zoom all the way in on the boundary. A photo that looks fine at thumbnail size often falls apart at full magnification, and everything measured from it is only as good as that detail.
Keep reading
- How to photograph your own hairlineThe five variables to hold fixed so that two photos taken months apart are actually comparable.
- Why the same head scores differently in two photosSeven causes of variance, ranked. Your actual hair is seventh, and that is the point.
- How Crown scores a hairlineThe actual score bands, the rule that makes a zone return a blank, and every metric Crown reports.
First scan free. No account, no email, no login. Crown+ is $4.99/week or $39.99/year, each with a 3-day free trial.
Crown measures how hair looks in a photograph. It is not a medical device, it gives no medical advice, and nothing on this page is a substitute for a qualified professional. If something about your hair or scalp worries you, talk to one.