Help Identifying Blurry Kanji

I’m trying to read some Japanese in the wild, and have downloaded some (paid for!) books. I’m trying to make out this character: in my version of the book, even at 200% increase, I just can’t make it out:

(Hopefully this works, I’ve never uploaded a pic before).

I see a blur, with the radical for stool in the upper right. Apparently it’s pronounced KEI, but I’m not sure how to look that up. Ring any bells?

Hmmm I believe that’s a:


100% sure it’s 「警」. Related jisho link:

Also, I should also add that this is the kanji I hate the most. No matter how much I practice drawing it, I can never make it look good.


If you put けい into the search here, not that many kanji come up. You’d be able to pick it out.


Don’t forget about search by radical :stuck_out_tongue_winking_eye:

You mean like, search for the name of a radical you can see and then look at that radical’s item page? The only one you can really make out is called “winter” here and it’s not really even the same shape as the radical in the blurry kanji.

(Since I was just talking about WK and how to look up by a reading, I didn’t mention other site’s kanji lookup functions, but someone already linked Jisho)



Funny how it shows chinese simplified hanzi in the link title.

On my browser it doesn’t. Perhaps that has to do with your configurations?

Theres so many kanji not on wanikani that I usually resort to typing readings and pressing space looking at all the options of kanji that pop up and use jisho…

As the others said you can use the furigana and compare the kanji proposed by your IME (in my case it was number 59 out of 116, but it changes with your writing habits) with the kanji you see.
Or you can input the radicals you can recognize on, and compare all the remaining kanji.

But there’s a third way for common enough words (i.e. stuff on Jisho) : the website doesn’t say it on the homepage, but it allows you to use some pretty cool search option: Search Options -
You can use wildcards with the kanji or kana before/after your blurred kanji to check for words.
We can’t see it in the image, but there must be something after it.
Let’s assume there’s 察官 after your missing kanji, you can search this:
*察官 - (only 8 possibilities)
or if you know the word starts with your kanji, this:
?察官 - (3 possibilities)

Edit: you can also mix kanji/kana around your missing kanji with readings that the word contains:

Not always fastest, but it’s an option…and you’ll probably look it up on Jisho anyway :man_shrugging:


Edit: After clicking the links in Kurei’s post above I see that they have the same question marks :sweat_smile:

To add to this, you can also search for words containing a kanji using a question mark in the place of the kanji you dont know.

For example if I didnt know the reading for 警 and I saw 警察, I would type ?察 and it would show me all the combinations with that order. This usually helps me find some kanji that have many radicals which would be hard to use the radical search.

1 Like

Yep, “?” is probably more useful if you know where you word starts/ends.
For anyone not used to wildcards:
You use * (asterisk) for an unknown number of characters (kanji, kana, numbers…), so any number of characters (0 included)
You use ? (question mark) for only and exactly one unknown character (kanji/kana or whatever). Not 0, not 2, not 7. Only 1.
The difference is in the number of unknown characters.

If you are thinking there’s only one unknown character, we should use “?”, it’s true, but some times we don’t know where the word begins or ends so using just “?” would exclude all cases where there’s more than one.

For example (bad example, but I couldn’t find a decent one), let’s say we found 宣言型言語 (declarative language) , let’s say we can’t read the second character (言) and we don’t really know where it begins/ends.
We could look up :

宣? #words          (27 results)

but this assumes that our word starts with 宣 (true) and that it’s made up of these two kanji alone (false), so we wouldn’t find our word. It would be excluded from the results.
On the other hand, using

宣* #words            (105 results)

assumes that our word starts with 宣 (true). It doesn’t assume anything about where it ends, so it includes our word.

And so on…it takes a little bit of practice.


I did try–using the little stool, which was the only thing I could make out–but none of the results, to my novice eyes, really resembled the blur :wink:

That wasn’t 又 but 夊

Although, the character is certainly 警 (Level 17): WaniKani / Kanji / 警

1 Like

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.