Comments (7)
@FUF1 can you send it via email? Send it over to: karol.zak at hotmail.com
from boxdetect.
karol.zak at hotmail.com
Yes, I sent it. 👍
from boxdetect.
Tweaking the values with the suggestions of Karol resulted in a much higher success rate. Thanks again.
from boxdetect.
Hi @FUF1,
Glad you found my package useful :)
BoxDetect should be perfect for your type of input. Try the following:
# 2nd step would be to try making sure this width and height ranges are correct for your size of boxes
# it's best to open something like paint or so and just manually validate that your boxes are in fact within this pixel range
cfg.width_range = (25,42)
cfg.height_range = (25,42)
# 1st step would be to change scaling_factors to something like [0.8] or [0.7]
cfg.scaling_factors = [0.75]
# you know you're looking just for squares so you can limit wh_ratio_range to something like: (0.8, 1.2)
cfg.wh_ratio_range = (0.8, 1.2)
cfg.dilation_iterations = 0
Do let me know how it went! Good luck!
from boxdetect.
Hi @FUF1, Glad you found my package useful :) BoxDetect should be perfect for your type of input. Try the following:
# 2nd step would be to try making sure this width and height ranges are correct for your size of boxes # it's best to open something like paint or so and just manually validate that your boxes are in fact within this pixel range cfg.width_range = (25,42) cfg.height_range = (25,42) # 1st step would be to change scaling_factors to something like [0.8] or [0.7] cfg.scaling_factors = [0.75] # you know you're looking just for squares so you can limit wh_ratio_range to something like: (0.8, 1.2) cfg.wh_ratio_range = (0.8, 1.2) cfg.dilation_iterations = 0Do let me know how it went! Good luck!
Hi @karolzak
Sorry it took me so long to respond. I had very limited time and tried your suggestion but didn't succeed. The program detected less checkboxes when i changed the scaling factor to something else than 0.5. I checked the size of the checkboxes and it was fine. So I had to write a test first, that checked through all sensible possibilities. My initial values seemed to be pretty close. scaling_factor = 0.5 and the selected width & height_range gave me the most detections for the testfile (but many other combinations did equivalent as well). Is there anything else to consider that it might work better? Thanks a lot!
from boxdetect.
@FUF1 could you send me an example of one of your images? I'll try to run it with my code and see if I can crack it
from boxdetect.
@karolzak Thank you for the generous offer! Is there a way to send it to you non publicly since the tests should not be distributed unnecessarily? I tried to erase the content, but then the program seems to detect different things.
from boxdetect.
Related Issues (20)
- As a user I want to automatically get optimal configuration based on provided ground truth
- AttributeError: module 'boxdetect.config' has no attribute 'update_num_iterations'
- Strategies for getting accurate checkboxes on documents with Serif Font HOT 3
- Is there way to extract coordinates HOT 3
- Check box mapping with text HOT 1
- challenging case on checkbox crossing outside box HOT 2
- Failed detection of cropped image HOT 2
- Cumulative results? HOT 2
- Removing noise while preserving the boundary of the checkbox HOT 4
- Not detecting all the boxes HOT 1
- Default config for vertical grouping has bad results for vertically aligned checkboxes HOT 2
- New release for scikit-learn installation HOT 3
- Failure in UnitTests HOT 1
- Which configurations should I use?
- Can't detect table cells
- using boxdetect in a lambda errors due to GUI artifacts
- AttributeError: module 'boxdetect.config' has no attribute 'update_num_iterations'. Did you mean: 'dilation_iterations'? HOT 1
- Add missing docstrings
- Add full tests coverage
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from boxdetect.