search

NeutrinosPlatform / cordova-plugin-mobile-ocr

A cordova plugin that can accept image URI or Base64 data and returns the text present in the image as string without need for network. You can also try :- https://github.com/NeutrinosPlatform/cordova-plugin-ml-text
MIT License
49 stars 30 forks source link

Does not working in html code #4

Closed merbin2012 closed 5 years ago

merbin2012 commented 5 years ago

I am trying to scan the HTML programs, but it doesn't read the symbols properly for example

<ul> <li>one</li> <li>Two</li> </ul> <style> ul{list-style-type:none;} li{display:inline-block;} </style>

ChrisTomAlx commented 5 years ago

Hi @merbin2012

Could you show me the image you captured and the detected response?

Also you could try the cordova-plugin-ml-text plugin. It might give a more accurate response. But it requires firebase.

Cheers, Chris Neutrinos

merbin2012 commented 5 years ago

Thank you very much for the quick response. Please see the below attached file

1545821927900 1545821917949 1545821987801

ChrisTomAlx commented 5 years ago

Are you trying to take the pic from your screen? Could you try by printing the html onto a paper and then taking the pic or using a screen shot of the html code.. Taking pics from screen tend to produce erroneous results.

Also could you show me your code, so I understand the options you have used to run the ocr.

Cheers, Chris Neutrinos

merbin2012 commented 5 years ago

But google translator reading perfectly.

On Wed 26 Dec, 2018, 6:02 PM Christy Thomas Alex <notifications@github.com wrote:

Are you trying to take the pic from your screen? Could you try by printing the html onto a paper and then taking the pic or using a screen shot of the html code.. Taking pics from screen tend to produce erroneous results.

Also could you show me your code, so I understand the options you have used to run the ocr.

Cheers, Chris https://github.com/ChrisTomAlx Neutrinos http://www.neutrinos.co/

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/NeutrinosPlatform/cordova-plugin-mobile-ocr/issues/4#issuecomment-449960306, or mute the thread https://github.com/notifications/unsubscribe-auth/ADhDEOpWt8EgDP_4NucXJxDQSsFUklq4ks5u82xEgaJpZM4ZhpRE .

merbin2012 commented 5 years ago

Hi! Can you fix this? Especially symbols are not working like <,>

ChrisTomAlx commented 5 years ago

There is not much that can be done about accuracy, if you can show me how you used the plugin I can see if there are any performance improvements you can implement.

You can also try the ml-text plugin. In my tests it gave better accuracy. Usage is almost similar.

Sorry I couldn't be of more help.

Cheers, Chris Neutrinos

merbin2012 commented 5 years ago

Thank you very much. Is ml-text plugin uses google's algorithm? Because it required firebase, that's why I asking

ChrisTomAlx commented 5 years ago

Yes it does! Google just took the mobile vision library used in this plugin and integrated it into firebase. They then stopped support for the mobile vision library and only support it through firebase. Hence two plugins that perform similar functionalities were born.

Closing this issue for now. If you decide to use the ml-text plugin and run into problems then raise an issue there.

Cheers, Chris Neutrinos