image_to_lasca.py

import numpy as np
import cv2
import wx

app = wx.App()
frame = wx.Frame(None, -1, 'image_to_lasca.py')
frame.SetSize(0, 0, 200, 50)

openFileDialog = wx.FileDialog(frame, "Open", "", "",
                               "Uncompressed speckle images (*.tif)|*.tif",
                               wx.FD_OPEN | wx.FD_FILE_MUST_EXIST | wx.FD_MULTIPLE)  #

openFileDialog.ShowModal()
file_list = openFileDialog.GetPaths()
openFileDialog.Destroy()

filter_size = 7
kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (filter_size, filter_size)).astype(np.float32)
kernel /= np.sum(kernel)

gray_threshold = 20
max_value = 0.65

save_video = True

if __name__ == '__main__':

    for file in file_list:
        file_name = file.split('\\')[-1]
        print(file_name[:-4])

        buffer = cv2.imread(file)

        output_image = file_name[:-4] + "_LASCA.tif"

        if len(buffer.shape) > 2:
            buffer = cv2.cvtColor(buffer, cv2.COLOR_BGR2GRAY)

        img = buffer.astype(np.float32, copy=False)
        mean_img = cv2.filter2D(img, ddepth=cv2.CV_32F, kernel=kernel)
        mean_img_sq = np.multiply(mean_img, mean_img)
        sq = np.multiply(img, img)
        sq_img_mean = cv2.filter2D(sq, ddepth=cv2.CV_32F, kernel=kernel)
        std = cv2.subtract(sq_img_mean, mean_img_sq)
        std[std <= 0] = 0.000001
        cv2.sqrt(std, dst=std)
        mask = mean_img < gray_threshold
        cv2.pow(mean_img, power=-1.0, dst=mean_img)

        LASCA = cv2.multiply(std, mean_img)
        LASCA -= 0.0
        LASCA /= max_value

        LASCA[LASCA < 0.0] = 0.0
        LASCA[LASCA > 1.0] = 1.0

        LASCA = 1 - LASCA
        LASCA = cv2.GaussianBlur(LASCA, (5, 5), 0)
        LASCA[mask] = 0
        LASCA *= 255
        LASCA = LASCA.astype(np.uint8)
        cv2.filter2D(LASCA, dst=LASCA, ddepth=cv2.CV_8U, kernel=kernel)

        im_color = cv2.applyColorMap(LASCA, colormap=cv2.COLORMAP_JET)
        cv2.imwrite(output_image, im_color)