dirtmaxim/lungs-finder

Library for detecting lungs on chest x-ray images for further processing. It is fast and able to work on embedded devices.

lungs-finder

Library for lungs detection on CXR images

Requirements:
- Python 3;
- OpenCV3.

Installation:
- pip install lungs-finder

You can test your dataset using lungs_viewer.py.
It is a program that is displayed on the screenshot above.
Usage: lungs_viewer.py /your/dataset True
The last parameter defines whether labels will be displayed or not.

You can get lungs using HOG, Haar or LBP.
Example:

import lungs_finder as lf

image = cv2.imread("/your/image/cxr.png", 0)

# Get both lungs image. It uses HOG as main method,
# but if HOG found nothing it uses HAAR or LBP.
found_lungs = lf.get_lungs(image)

if found_lungs is not None:
    cv2.imshow("Found lungs", found_lungs)
    code = cv2.waitKey(0)

# Or you can get left or right lung independently using HOG, Haar or LBP.
right_lung_hog_rectangle = lf.find_right_lung_hog(scaled_image)
left_lung_hog_rectangle = lf.find_left_lung_hog(scaled_image)
right_lung_lbp_rectangle = lf.find_right_lung_lbp(scaled_image)
left_lung_lbp_rectangle = lf.find_left_lung_lbp(scaled_image)
right_lung_haar_rectangle = lf.find_right_lung_haar(scaled_image)
left_lung_haar_rectangle = lf.find_left_lung_haar(scaled_image)

if right_lung_hog_rectangle is not None:
    x, y, width, height = right_lung_hog_rectangle
    right_image = image[y:y + height, x:x + width]
    cv2.imshow("Right lung", right_image)
    code = cv2.waitKey(0)

lungs-finder's Issues

Lungs cannot be detected correctly

Hi,

Thanks for publishing this library, it is super helpful!

One issue I found is that when dealing with a specific data (see attached), it does not return the images correctly. Looks like there is some bug somewhere.

ImportError: cannot import name haar_finder

Fixed with Python 3

The bottom and top of lungs is missed a bit

This library is very amazing. I'm working with lungs detection and analysis but I would like to detect lung until the bottom.

In the image as you see the bottom is missed. I used hog method how can I optimize this?

Documentation

Hi, do you have documentation on how the algorithm works for learning purposes?

Dataset for training

Thanks for sharing the great repo. May I ask what dataset you used to train the detector? (include both positive and negative samples) Thanks!