What would be the best combination for processing parameters that needs to be send for obtaining accurate result from the picture of an event ticket? Currently i am using the following parameters Language = English exportFormat = txt and what would be the optimum resolution for an image sent does it need to grey scaled before sending?
First, please try the textExtraction profile for your images. This profile is suitable for extracting all text from the input image.
Also we recommend you to take photo only of the area of interest without other objects and colored background. Please see our tips for taking photo:
Please also refer to the Best Practices section where you can find more recommendations on how to achieve the best recognition results.
As for your image we have cropped it and corrected the geometrical distortion:
Then we have processed this new image using the processImage method with the textExtraction profile and got the following result:
The main data is extracted rather accurately and if you manage to improve the quality of your source images, you can expect to get even better recognition results.
answered 30 Sep '15, 16:36
Oksana Serdyuk ♦♦