Defining/counting filamentous cells using OpenCV

Question

I am attempting to perform some image analysis of filamentous cells. The images are quite busy so I am primarily interested in the cells that are in focus in the images to calculate their dimensions. I have found a lot of documentation of more typical round cells but am having trouble getting adaptations of this code to work with my filamentous cells.

I have used the answer from this thread to form the basis of the image analysis: Irregular shape detection and measurement in python opencv

import numpy as np
import cv2

img = cv2.imread('example.tif')
blur = cv2.GaussianBlur(img, (7, 7), 2)
h, w = img.shape[:2]

kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (7, 7))
gradient = cv2.morphologyEx(blur, cv2.MORPH_GRADIENT, kernel)
cv2.imshow('Morphological gradient', gradient)
cv2.waitKey()

lowerb = np.array([0, 0, 0])
upperb = np.array([15, 15, 15])
binary = cv2.inRange(gradient, lowerb, upperb)
cv2.imshow('Binarized gradient', binary)
cv2.waitKey()

for row in range(h):
    if binary[row, 0] == 255:
        cv2.floodFill(binary, None, (0, row), 0)
    if binary[row, w-1] == 255:
        cv2.floodFill(binary, None, (w-1, row), 0)

for col in range(w):
    if binary[0, col] == 255:
        cv2.floodFill(binary, None, (col, 0), 0)
    if binary[h-1, col] == 255:
        cv2.floodFill(binary, None, (col, h-1), 0)

cv2.imshow('Filled binary gradient', binary)
cv2.waitKey()

foreground = cv2.morphologyEx(binary, cv2.MORPH_OPEN, kernel)
foreground = cv2.morphologyEx(foreground, cv2.MORPH_CLOSE, kernel)
cv2.imshow('Cleanup up crystal foreground mask', foreground)
cv2.waitKey()

kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (17, 17))
background = cv2.dilate(foreground, kernel, iterations=3)
unknown = cv2.subtract(background, foreground)
cv2.imshow('Background', background)
cv2.waitKey()

markers = cv2.connectedComponents(foreground)[1]
markers += 1  # Add one to all labels so that background is 1, not 0
markers[unknown==255] = 0  # mark the region of unknown with zero
markers = cv2.watershed(img, markers)

hue_markers = np.uint8(179*np.float32(markers)/np.max(markers))
blank_channel = 255*np.ones((h, w), dtype=np.uint8)
marker_img = cv2.merge([hue_markers, blank_channel, blank_channel])
marker_img = cv2.cvtColor(marker_img, cv2.COLOR_HSV2BGR)
cv2.imshow('Colored markers', marker_img)
cv2.waitKey()

labeled_img = img.copy()
labeled_img[markers>1] = marker_img[markers>1]  # 1 is background color
labeled_img = cv2.addWeighted(img, 0.5, labeled_img, 0.5, 0)
cv2.imshow('watershed_result.png', labeled_img)
cv2.waitKey()

Included in this post is an example image of some cells similar to what I am describing as well as the output I am receiving. From what I can tell this appears to define my background as my foreground and vice-versa. Instead of defining the empty out of focus spaces between cells I want to define the segmentations of the cells so I can look at cellular dimensions. How could I change the parameters of this framework to focus on the cells instead of the background? Thank you.