Is it possible in OpenCV to build a local curvature in the form of a heat map representing the "pointiness" object?

Question

Given the threshold image of the drop that you can detect and draw outlines around, is it possible to represent the local curvature as a heat map when drawing the outline?

i.e. (1) it is possible to determine the local curvature on the open cv-contour (2), displaying this curvature in the color space of the heat map (3), we draw the contour as a heat map.

My goal is to measure the "sharpness" of the object so that I can draw a vector from the point side to the opposite non-empty side. From my objects, I know that the pointed side is the top.

If other methods will be more effective at presenting "pointedness" than curvature, feel free to suggest.

Answer 1

EDIT : fixed bug in previous version.

I used the angle between the gradient vectors at the i-th and (i + n) -th points on the contour as an estimate for determining the point accuracy of the point. Code and results below.

import numpy as np import cv2 import pylab as pl def compute_pointness(I, n=5): # Compute gradients # GX = cv2.Sobel(I, cv2.CV_32F, 1, 0, ksize=5, scale=1) # GY = cv2.Sobel(I, cv2.CV_32F, 0, 1, ksize=5, scale=1) GX = cv2.Scharr(I, cv2.CV_32F, 1, 0, scale=1) GY = cv2.Scharr(I, cv2.CV_32F, 0, 1, scale=1) GX = GX + 0.0001 # Avoid div by zero # Threshold and invert image for finding contours _, I = cv2.threshold(I, 100, 255, cv2.THRESH_BINARY_INV) # Pass in copy of image because findContours apparently modifies input. C, H = cv2.findContours(I.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE) heatmap = np.zeros_like(I, dtype=np.float) pointed_points = [] for contour in C: contour = contour.squeeze() measure = [] N = len(contour) for i in xrange(N): x1, y1 = contour[i] x2, y2 = contour[(i + n) % N] # Angle between gradient vectors (gx1, gy1) and (gx2, gy2) gx1 = GX[y1, x1] gy1 = GY[y1, x1] gx2 = GX[y2, x2] gy2 = GY[y2, x2] cos_angle = gx1 * gx2 + gy1 * gy2 cos_angle /= (np.linalg.norm((gx1, gy1)) * np.linalg.norm((gx2, gy2))) angle = np.arccos(cos_angle) if cos_angle < 0: angle = np.pi - angle x1, y1 = contour[((2*i + n) // 2) % N] # Get the middle point between i and (i + n) heatmap[y1, x1] = angle # Use angle between gradient vectors as score measure.append((angle, x1, y1, gx1, gy1)) _, x1, y1, gx1, gy1 = max(measure) # Most pointed point for each contour # Possible to filter for those blobs with measure > val in heatmap instead. pointed_points.append((x1, y1, gx1, gy1)) heatmap = cv2.GaussianBlur(heatmap, (3, 3), heatmap.max()) return heatmap, pointed_points def plot_points(image, pointed_points, radius=5, color=(255, 0, 0)): for (x1, y1, _, _) in pointed_points: cv2.circle(image, (x1, y1), radius, color, -1) def main(): I = cv2.imread("glLqt.jpg", 0) heatmap, pointed_points = compute_pointness(I, n=5) pl.figure() pl.imshow(heatmap, cmap=pl.cm.jet) pl.colorbar() I_color = cv2.cvtColor(I, cv2.COLOR_GRAY2RGB) plot_points(I_color, pointed_points) pl.figure() pl.imshow(I_color) if __name__ == '__main__': main() 

Note that sharper points are brighter in the heat map.

Answer 2

The fact is that "if you bring the contour closer to extended lines, you can see that the point is the point at which the maximum deviation of the angle for the serial line occurs", based on this, you can develop your own algorithm.

You need to do

• Find the outline .

• Find approxPolyDP () for the path.

• Calculate the angle for each consecutive line and save the point at which the maximum deviation occurs.

You can calculate the angle of the line using the equation

  double Angle = atan2(P2.y - P1.y, P2.x - P1.x) * 180.0 / CV_PI;