CODEWITHSUNDEEP
iphoneobjective-ciosxcodeocr
user1645721
user1645721

Reputation: 653

iPhone OCR doesn't recognize all the captured images text, But can recognize the predefined Text images

I have implemented a business card reader image using ocr api by follow the link/ tutorial

http://tinsuke.wordpress.com/2011/11/01/how-to-compile-and-use-tesseract-3-01-on-ios-sdk-5/

It can scan the text of some screen shots of UIalert views / table views. and some images having text like follows.

Can recognize text on this image:

enter image description here enter image description here

[Airoplane mode wi-fi vpn ratch data sounds brightness general wall paper]

but when i capture business card it returns some special charecters and numbers. some times recognized if it is very very clear and big text in captured image.

Can NOT recognize text on follow like images

enter image description here enter image description here

[ xas $- 432 *7 5$#2 $@@#%^ ^&* I( WQ#$ follow like√ √dg @ ]

why how to resolve this issue.

Upvotes: 1

Views: 2095

Answers (1)

V-Xtreme
V-Xtreme

Reputation: 7323

Tesseract library is open source library. I don't think it is that much accurate.OCR gives best results for image more than 150 px resolution with black and white image .The image which you are capture with the camera of iPad or iPhone is saved with 72px that tesseract cannot give best results.If you are more interested then there are some paid libraries with good accuracy you can try out some :here

Upvotes: 1

Related Questions