imgprep
PLAESE NOTE: Development of this project is currently not active.
A small script to prepare the images of my microscopy for analysis.
Reasoning for the script
Many microscopy softwares can't crop images with the standard software. Neither can they put an automatic scale into the image by itself. When processing a huge amount of images, it can be tedious to edit every image by hand.
Tis small python-script tries to do that automatically. Another benefit is learning the python image processing libraries their basic uses.
The script always treats one sample at a time. Multiple images can be associated to one sample. Draw red squares around your images beforehand. The software will recognize the square and crop the image to it.
Also provide a csv file providing the magnification data (e.g. 63x magnification = 0.135mm screen width). The program will read it and use it to automatically add a scale to your images.
Please report any bugs / concerns / improvements.
Script-Progress
- Initialization
- [x] Load the all images of the same object
- [x] Meta-data
- [x] Read image names and file-extensions
- [x] Read the magnification(x20 or x63) from the file-name
- Alteration
- [x] Cropping
- [x] Recognize the red square around the object
- [ ] Recognize a square of any color
- [x] Crop the image to the ROI
- [ ] Crop to the insides of the square
- [ ] Crop to squares with borders of variable width
- [x] Image-Arrangement
- [x] Scale-Bars
- [x] Calculate individual scale-bars for every sub-image
- [x] Calculate positions for scale-bars
- [x] Insert scale-bars
- Saving
- [x] Saving the edited images
- [x] Save the finished arrangement of sample images into one file (.png)
TODO Priorities
- Revise code to shorten it and make it more intuitive
- Rewrite saving methods (export saving from alignment method into regular saving)
- Remove white borders from matplotlib figures (
add_scale and align methods)
- Change all hard-coding to be more flexible
- Magnification table
- Red square -> any square