During World War II the Japanese used some hi-tech codes to prevent their military communications being read by their enemies. However they believed that even if the codes were broken, the inferior foreign devils would not be able to read the Japanese text anyway. On both counts they were overconfident, and code-breaking led to some significant Japanese defeats, such as the battle of Midway, and the shooting down of a plane carrying Admiral Yamamoto.
Let’s assume you have  an interesting-looking Japanese manga that nobody has translated (scanlated). If you are fluent in written Japanese and know hundreds of kanji, you can just read manga, right?
Those who find Japanese text more akin to code may read on:
Here’s how to hack into it. Â Don’t try looking up words in a Japanese-English dictionary. If you have not learned to recognise the kanji (chinese characters) it will take ages.
1) Scan the text page to get a high-resolution .tiff file.
2) Load the image file into a Japanese OCR program (e.g. RealReader Lite 6.0)
3) Tidy up and correct the plain Japanese text output with a Japanese word-processor (e.g. JWPce)
4) Paste the text into a Japanese-English (or other language) translation program. (the web-based Google Translate & Yahoo Babel Fish will do just fine). The result, even if you’ve used machine translation before, can be a severe disappointment… At this point you have to roll up your sleeves and use your sketchy knowledge of Japanese vocabulary and grammar, the machine-translated output(s), and translation of individual kanji (JWPce is particularly quick for this), along with the manga pictures and layout, to arrive at a decent translation.
For the non-expert, it usually helps to write out the Japanese in romaji (English characters). Google Translate does this for you, to save you the trouble of writing it out yourself.
Note that scanning and digitizing the kanji is a major time-saver!