4

Over the last couple of years, with the introduction of the electronic passport gates at Heathrow (LHR) - I've found I've had a 0% success rate of getting through them, with the system always flagging me to see a human - who manually checks my passport and lets me through.

Until now, I've assumed this was just bad luck (or personal incompetence at using these scanners).

However, when recently applying for an ESTA to visit the USA - I tried to use their "Upload passport" functionality, that asks specifically for the Machine Readable section at the bottom of a passport. Every time I tried, it was refused with an error about not being readable. (Just to be clear, my passport itself is definitely not blocked - as inputting the information manually was approved without issues, and the ESTA was approved).

It's got me wondering, are there any known bans on certain British Passports (e.g. ban lists, or codes embedded into the Machine Readable section) that forces the holder to always see a human for entry, or otherwise blocks use of the Machine Readable section? Or is this just a case of repeat bad luck/user incompetence?

Note, this is in reference to the Machine Readable section, of written characters, on the main picture-page. Not an embedded electronic/biometric chip.

  • 3
    If the machines don't accept your passport but you never have any problems when speaking to a person, and trying to use the MRZ for your ESTA resulted in an error saying it was "unreadable" but putting the details in manually resulted in no problems... what on earth leads you to think the problem is anything other than the MRZ on your passport being (for whatever reason) unreadable? – Chris H Nov 27 '19 at 12:30
  • 1
    @ChrisH It's a really pretty new passport, and so I assumed that the MRZ (as it's entirely undamaged/marked) would have been printed correctly/readably. But if printing errors (in that section) aren't unheard of for British Passports - that would explain it (I guess I'd assumed it wouldn't have been sent out if it wasn't? Plus, it "looks" fully readable. But I'm guessing these was bad assumptions then.) – Bilkokuya Nov 27 '19 at 12:42
  • 1
    If you take a look at this wikipedia page, there's not really anywhere for any kind of codes to go in the MRZ anyway. I guess in theory the "personal number" could include some kind of code. But if it were something in that, that would still leave the question of why having a "banned" passport would cause no actual problems for you while travelling. I might try to turn this into a real answer later, but don't have time to do a good job of it right now. – Chris H Nov 27 '19 at 12:57
  • 3
    @Bilkokuya I tried to upload the machine readable section of my passport to the ESTA website 10+ times, each time getting the same rejection message as you did. I finally managed the upload by using a photocopy my passport. My conclusion was that some slight angle problem or a reflection off the laminate was preventing the upload. – Traveller Nov 27 '19 at 12:58
  • 3
    I am not sure if that is the case here, but it is known to cause problems with certain readers if the MRZ background has too prominent security features (guilloché or other patterns). The ICAO document specification is unfortunately rather vague on how the MRZ background is supposed to look. – Tor-Einar Jarnbjo Nov 27 '19 at 13:22
7

The short answer is that I would consider this to be very unlikely.

Let's start by taking a look at what the machine readable zone actually contains. From wikipedia, the first line contains:

 Positions || Length || Characters || Meaning
 1         || 1      || alpha      || P, indicating a passport
 2         || 1      || alpha + <  || Type (for countries that distinguish between different types of passports)
3–5        || 3      || alpha + <  || Issuing country or organization (ISO 3166-1 alpha-3 code with modifications)
6–44       || 39     || alpha + <  || Surname, followed by two filler characters, followed by given names. Given names are separated by single filler characters

and the second line contains:

 Positions !! Length !! Characters  !! Meaning
 1–9       || 9      || alpha+num+< || Passport number
  10       || 1      || numeric     || Check digit over digits 1–9
  11–13    || 3      || alpha+<     || Nationality (ISO 3166-1 alpha-3 code with modifications)
  14–19    || 6      || numeric     || Date of birth (YYMMDD)
  20       || 1      || numeric     || Check digit over digits 14–19
  21       || 1      || alpha+<     || Sex (M, F or < for male, female or unspecified)
  22–27    || 6      || numeric     || Expiration date of passport (YYMMDD)
  28       || 1      || numeric     || Check digit over digits 22–27
  29–42    || 14     || alpha+num+< || Personal number (may be used by the issuing country as it desires)
  43       || 1      || numeric+<   || Check digit over digits 29–42 (may be < if all characters are <)
  44       || 1      || numeric     || Check digit over digits 1–10, 14–20, and 22–43

(apologies for the formatting, stack exchange sites don't allow any good way of doing tables as far as I can tell)

You can see that more or less everything in the MRZ is fixed. The only free field is the "personal number".

In principle, is it possible that the personal number contains some kind of code? Sure. And if it did, that's the kind of information that I wouldn't necessarily expect to be publicly stated. But I don't believe it's at all likely that there's some kind of secret code hidden here which is used to reject your passport by the machines, for one main reason:

When the machines reject you, you go and see a human. And they check your passport, and let you through. If we assume there is some code, it doesn't seem to be having any effect (I guess you'd have mentioned if you seemed to face heightened scrutiny after the machines reject you). That could be explained by
a) the agents can quickly perform some look up that shows you're not a concern
b) agents can't tell the code is there
c) there is no code

a) seems implausible, as if the lookup is so quick you can't tell it's going on it's almost certainly something that just shows the agents a yes/no answer, and the machines should be able to do it themselves
b) would mean the code is completely and utterly pointless in the first place
c) seems compatible with everything we know

Similar logic applies to your ESTA application. If the MRZ contained some kind of code suggesting you're a danger, it wouldn't make any sense for it to be described as "unreadable" and then you have no trouble getting the ESTA when you input your details manually. The MRZ would surely be read just fine, and quite likely set some internal flag to alert them to the problem so it can be considered when they decide whether or not to grant you the ESTA.

I think the above lays out a few reasons that the idea of the MRZ having some code marking your passport as banned doesn't make much sense, given your experiences. Ultimately I see no reason to believe that there's anything going on beyond what the ESTA application told you: your MRZ is unreadable. I can't give you a certain answer as to why that is, but some possibilities have been raised by various users in the comments.

A printing error seems unlikely but isn't entirely out of the question. It might be recognisable if you took a close look at your MRZ (perhaps with reference to the font).

An error in the check digits (as @phoog mentioned) also seems rather unlikely (and I would expect it to lead to an error describing it as "invalid" rather than "unreadable", but you never know), but certainly wouldn't be easily recognisable to humans.

The pattern behind the text may be causing some difficulty for the OCR process. This would presumably have been noticed if it were widespread, but the pattern varies across the MRZ - it's conceivable that a particular character becomes difficult to identify only when it occurs at some specific location in the MRZ.

@Traveller mentioned having similar experience of a passport repeatedly being rejected, but a photocopy of that passport being accepted. Even a small imperfection on the surface could in principle cause weird reflections that interfere with the OCR process but aren't easily visible when looking at it under normal lighting conditions.

  • 1
    Thank you for this extremely comprehensive answer. It clears everything up. – Bilkokuya Nov 28 '19 at 10:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.