Sreemathi

Yes thanks alot for sharing it. let me check and post queries further if i have any.

Yes i saw that link, and can we train the handwritten pdf documents using jTessBoxEditor. I am very new to Tesseract. so please guide me on this.

Thanks for your prompt reply. We are in a process to get good configuration servers soon. Meanwhile how to capture the handwritten texts in Tesseract? I came across a topic that we need to train the handwritten fonts. Is that to be done only with Tesseract tool installed?

Yes finally i found that trained data and processed few batches. it's working better now. will try this out on few more batches. Thanks alot for sharing the link. We don't have tesseract installed on our machine so we didn't try running on command line. we have just implemented the tess4j jar on java code as mentioned before. Meanwhile is there any other way to increase the speed of processing the documents on tesseract?

Yes i saw that link. i couldn't see the traineddata for eng. so which one can i use from that link? And also OS is Windows 10

Hi Quan Nguyen. Thanks for the reply. can you please be more specific. please share me the link of tessdata_fast version. Do we have tess4j for that fast version? Already we have eng.traineddata used in our java code The configuration is 8GB RAM, 64 bit OS.

I am currently working on performing ocr using Tesseract. i am using Tess4j version 4.4.1 to scan documents. we notice that any clear documents is taking 20-30 sec to perform ocr. But if the image has some distortions or unclear its taking more than a min to ocr. i would like to understand what are all the major factors that would impact Tess4j performance. All our images would have distortions, have many lines, drawings,handwritten (though hand written need not be OCRed). Also we would like to understand...