官方的提高tesseract識別成功率的相關方法

標籤：des   http   io   ar   os   sp   for   on   art   

Improving the quality of the output

There are a variety of reasons you might not get good quality output from Tesseract. It‘s important to note that unless you‘re using a very unusual font or a new language retraining Tesseract is unlikely to help.

 

• Improving the quality of the output
• DPI
• Image processing
• Binarisation
• Noise
• Orientation / Skew
• Borders
• Segmentation method
• Dictionaries, word lists, and patterns
• Still having problems?

 

DPI

Tesseract works best with text using a DPI of at least 300 dpi, so it may be beneficial to resize images. For more information see the FAQ.

Image processing[就是預先處理，就是不提opencv，看來opencv也沒有那麼出名]

Tesseract does various image processing operations internally (using the Leptonica library) before doing the actual OCR. It generally does a very good job of this, but there will inevitably be cases where it isn‘t good enough, which can result in a significant reduction in accuracy.

You can see how Tesseract has processed the image by using the configuration variable tessedit_write_images to true when running Tesseract. If the resulting tessinput.tif file looks problematic, try some of these image processing operations before passing the image to Tesseract, whether with a dedicated postprocessing tool like Scan Tailor or unpaper, using a graphics editor like ImageJ or Gimp, with a batch image editor like ImageMagick, or in code using an image processing library like Leptonica.

Binarisation【如果這種東西都能識別，那麼名片什麼的都是弱爆了】

This is converting an image to black and white. Tesseract does this internally, but it can make mistakes, particularly if the page background is of uneven darkness.

Noise

Noise is random variation of brightness or colour in an image, that can make the text of the image more difficult to read. Certain types of noise cannot be removed by Tesseract in the binarisation step, which can cause accuracy rates to drop.

Orientation / Skew

This is when an page has been scanned when not straight. The quality of Tesseract‘s line segmentation reduces significantly if a page is too skewed, which severely impacts the quality of the OCR. To address this rotating the page image so that the text lines are horizontal.

Borders

Scanned pages often have dark borders around them. These can be erroneously picked up as extra characters, especially if they vary in shape and gradation.

Segmentation method

By default Tesseract expects a page of text when it segments an image. If you‘re just seeking to OCR a small region try a different segmentation mode, using the -psm argument. Note that adding a border to the text may also help, see issue 398.【這裡提到識別roi的新方法】

Dictionaries, word lists, and patterns

By default Tesseract is optimised to recognise sentences of words. If you‘re trying to recognise something else, like receipts, price lists, or codes, there are a few things you can do to improve the accuracy of your results, as well as double-checking that the appropriate segmentation method is selected.

Disabling the dictionaries Tesseract uses should increase recognition if most of your text isn‘t dictionary words. They can be disabled by setting the both of the configuration variables load_system_dawg and load_freq_dawg to false.

It is also possible to add words to the word list Tesseract uses to help recognition, or to add common character patterns, which can further help to improve accuracy if you have a good idea of the sort of input you expect. This is explained in more detail in the Tesseract manual.[有manual，在這裡找到的]

If you know you will only encounter a subset of the characters available in the language, such as only digits, you can use thetessedit_char_whitelist configuration variable. See the FAQ for an example.

Still having problems?

If you‘ve tried the above and are still getting low accuracy results, ask on the forum for help, ideally posting an example image.

官方的提高tesseract識別成功率的相關方法