Issue in extracting text from images

Hi,
I am writing this mail in behalf of amit.kumar@cantata.in and this is the account details

I want OCR functionality in my application to read text from images. I have question-answer based website and I want my users to upload a question from textbooks or notes by simply clicking it and uploading to our site. Then I require to convert that image into text format.
I have registered on your site for a 30 day free trial. Now when I use your given sample code i.e. ocr-java-examples to read text from images, the output doesn't match the text in the images.
I am attaching some images and their outputs are as below.

image19.jpg output --> A rUnraC* 1r-ead cr a uesu lreead Rannor re lesenm

image4.png output --> W
image9.jpg output --> `Theyiustmovedtomew Vork and it's kvnaa empty and
resume.jpg output --> Exception in thread "main" java.lang.OutOfMemoryError: Java heap space
at com.aspose.ocr.internal.azy.b(Unknown Source)
at com.aspose.ocr.internal.azy.a(Unknown Source)
at com.aspose.ocr.internal.avv.a(Unknown Source)
at com.aspose.ocr.OcrEngine.a(Unknown Source)
at com.aspose.ocr.OcrEngine.process(Unknown Source)
at com.aspose.ocr.examples.WorkingWithOcr.PerformOCROnImage.main(PerformOCROnImage.java:36)
[Note - all images are attached in zip file tmp.7z]

When I analyse this type of output, I came to know, image quality is 96 dpi so I have a question

A] CAN YOUR SOFTWARE CORRECTLY HANDLE IMAGE WITH 96 DPI QUALITY? Because, our users take image from mobile camera
and mobile gives upto 96 dpi images. or provide me guidelines about how to increase image quality ?



Hi,

Thank you for your inquiry.

Please note that the archive file shared in 7z format is not working. Current implementation of the Aspose.OCR API works well with images having resolution of at least 300 DPI and the accuracy rate tends to decrease by decreasing the resolution. On the other hand, if you intend to get some specific contents from a portion of the image, you can use the custom recognition blocks to get better accuracy.

We have already logged the issue of OCR operation on low DPI images with ID OCR-34250 in our system. We will update you about the progress via this thread.

The issues you have found earlier (filed as ) have been fixed in this Aspose.Words for JasperReports 18.3 update.