kwsp

from concurrent.futures import ThreadPoolExecutor
from pathlib import Path

from tqdm import tqdm
import cv2
import matplotlib.pyplot as plt
import numpy as np


def find_best_img_match(target_img: np.array, imgpaths: list[Path]):

kwsp

"""
Convert exported, rect OCT images to aligned rect and radial, save as NRRD files.

1. Change the `input_folder` to point to the directory containing images.
2. Change the row range used for correlation by changing `rowStart` and `rowEnd` to make sure they only include the tubing and avoid any tissue in this window. 

Install dependencies:

```
python3 -m pip install numpy matplotlib pillow pynrrd tqdm opencv-python

kwsp

"""
dependencies: numpy pillow pynrrd tqdm
"""

from PIL import Image
from pathlib import Path
from tqdm import tqdm
import nrrd
import numpy as np

kwsp

function(copy_to_target_dir target resource)
    add_custom_command(
      TARGET ${target} POST_BUILD
      COMMAND ${CMAKE_COMMAND} -E copy_directory
        "$<TARGET_PROPERTY:${target},SOURCE_DIR>/${resource}"
        "$<TARGET_FILE_DIR:${target}>/${resource}"
    )
endfunction()

copy_to_target_dir(${EXE_NAME} shaders)

kwsp

#include <iostream>
#include <fstream>
#include <armadillo>


// T is the type of value stored in the binary file.
template <typename T>
auto load_bin(const fs::path &filename) -> arma::Mat<T> {

  std::ifstream file(filename, std::ios::binary | std::ios::ate);

kwsp

Windows

I use git from scoop. Open git bash to access the gpg binary. If you install gpg from scoop it will not work as git doesn't know about it.

Import existing private key

gpg --import private.key

kwsp

jobs:
  build:
    steps:
      - name: Export GitHub Actions cache environment variables
        uses: actions/github-script@v7
 with:

kwsp

"""
typed_mat_mixin.py provides a mixin `InitMixin` that tries
to generate a `.init` method to a typed Dataclass to automatically
deserialize .mat files exported from MATLAB to a Python Dataclass.

Supported types:
- strings
- int, float, np.uint8, and most np typedefs
- np.ndarray (1D and 2D tested)
    - Type arrays as np.array[ndims, dtype]

kwsp

"""
B. S. Reddy and B. N. Chatterji, “An FFT-based technique for translation, rotation, and scale-invariant image registration,” IEEE Transactions on Image Processing, vol. 5, no. 8, pp. 1266–1271, Aug. 1996, doi: 10.1109/83.506761.

Implemented in Numpy/OpenCV and Numpy/Scikit-Image

The OpenCV version is about 4 times faster than the Scikit-Image version.
"""
# Numpy/OpenCV implementation
import cv2
from scipy import signal

kwsp

import numpy as np
import cv2


def polar2cart(img: np.ndarray):
    """
    Convert image (B-scan) from Polar coordinates to Cartesian coordinates
    """
    
    # Resize image to square first