How can I detect diagram region and extract(crop) it from a research paper enter image description here

  • 2
    $\begingroup$ your question is pretty general and I'm guessing by the down-vote that others feel the same. perhaps research a few methods quickly and learn roughly what they mean, make a list of them and pose a question about which of the methods on your list have worked for others. a quick google of "how to detect and crop a region of an image" will get you started. $\endgroup$ Dec 13, 2019 at 20:55
  • $\begingroup$ There are many methods that extract the image from file and you can apply image detection after that but you did not tell what type of file you need to extract(crop). So, your question needs to be clarified. $\endgroup$ Dec 14, 2019 at 4:36

1 Answer 1


I will consider that you need to extract(crop) the digram from the pdf research paper. You can use PyPDF2 or PyMuPDF to extract the images from the PDF file and then you can apply machine learning to do recognition and classification of the images. There are different types of machine learning solutions for image classification and you can start with Convolutional Neural Network and you can start here.

For more information for Recognition and classification of figures see these and to extract of figures from scholarly documents read this


Not the answer you're looking for? Browse other questions tagged .