search

brandonmpetty / Doxa

A Local Adaptive Thresholding framework for image binarization written in C++, with JS and Python bindings. Implementing: Otsu, Bernsen, Niblack, Sauvola, Wolf, Gatos, NICK, Su, T.R. Singh, WAN, ISauvola, Bataineh, Chan and Shafait.
https://brandonmpetty.github.io/Doxa/WebAssembly
Creative Commons Zero v1.0 Universal
167 stars 37 forks source link

skewed large images #25

Closed kevingnet closed 2 years ago

kevingnet commented 2 years ago

Thank you for this great library, it's very well put together and easy to use and understand. I'm doing a personal project, restarting after a very long time and this library will save me time so I can focus on what I'm doing.

When I use a large file, the binary result shows skewed lines thorough out the whole file. After I scaled the file in half it worked ok. The original file was 1800x1200 pixels.

brandonmpetty commented 2 years ago

Sorry to hear about the issue @kevingnet Could you tell me exactly what you are doing? Specifically:

With some of the more recent optimizations, all of the algorithms (except an experimental one) have been retooled to support much larger images, something not possible with the standard Integral Image technique. I'll check it out... thanks for letting me know!

kevingnet commented 2 years ago

Hi Brandon,

actually it works fine. my bad. the issue was that when i convert the image usim gimp the header has a label, however when i use convert fro graphics magic there is no label in the conversion, perhaps that's why, i haven't investigated more, but I'm glad it's working really well. great job and thanks for the prompt reply.

On Mon, May 2, 2022, 4:27 PM Brandon M. Petty @.***> wrote:

Sorry to hear about the issue @kevingnet https://github.com/kevingnet Could you tell me exactly what you are doing? Specifically:

  • Is this happening in C++ or one of the Bindings?
  • How are you reading your image into the framework?
  • I assume this is happening with any binarization method, or is it a specific one?

With some of the more recent optimizations, all of the algorithms (except an experimental one) have been retooled to support much larger images, something not possible with the standard Integral Image technique. I'll check it out... thanks for letting me know!

— Reply to this email directly, view it on GitHub https://github.com/brandonmpetty/Doxa/issues/25#issuecomment-1115464061, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABHJX7C4T5A53N3S4VF23RTVIBQHJANCNFSM5U2VTFZA . You are receiving this because you were mentioned.Message ID: @.***>

brandonmpetty commented 2 years ago

Awesome! Glad to hear. I just converted a 4554x5247 JPG (6.3MB) into an uncompressed PPM (68.3MB) and the framework successfully converted it to a Binary PBM (2.85MB) file without issue.