I want you to guide me on how to deal with Tesseract jTessBoxEditor to create a training model on 10 images in Arabic and run the model Hello Tesseract with Mahmoud Abdel Aleem I saw your contributions in GitHub about Tesseract and I benefited from you well Thank you for your useful contributions, Tesseract I want you to help me with the following: 1- I have a set of digital images of book covers, 10 images in Arabic, I want to convert them to text using Tesseract 2- The conversion model is inaccurate and does not recognize most of the words ara.traineddata in the tessdata file in Tesseract 3- I created a model ara1.traineddata using jtessboxeditor where I created boxes for each image and modified them in a sample image then created a file ara1.traineddata and put it in the tessdata file in Tesseract and repeated the experiment on the image that was trained on but it did not succeed I think there is an error in the work steps that I am doing using jtessboxeditor If possible Tesseract let me know the correct steps for training and creating a .traineddata file using jtessboxeditor even create a custom model for 10 digital images so that I can make Tesseract recognize them and convert them to text If possible help me by sending an illustrative image of the steps I would be grateful for your cooperation
في الأربعاء، 23 أكتوبر 2024 في تمام الساعة 6:42:25 م UTC+4، كتب [email protected] رسالة نصها: > On Wednesday, October 23, 2024 at 1:13:05 AM UTC-4 [email protected] > wrote: > > I am having an issue with Tesseract splitting text lines incorrectly for > the attached file of a metes and bounds legal description. It returns this: > > [...] > > Any ideas on how to fix this? > > > It would be helpful if you included the version you are using, language > model, the command line, etc. > > The most likely fix is to use a different page segmentation mode on the > command line. > > Tom > -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion visit https://groups.google.com/d/msgid/tesseract-ocr/8d70ca29-3113-4fcc-ae22-e7870e5a02can%40googlegroups.com.
