Commit 8b19938

committed

Circle detection

1 parent 08b7bdd commit 8b19938Copy full SHA for 8b19938

File tree

+51

-0

lines changed

+51

-0

lines changed

Lines changed: 26 additions & 0 deletions

Original file line number	Diff line number	Diff line change
`@@ -0,0 +1,26 @@`
	`1`	`+# from turtle import circle`
	`2`	`+import cv2`
	`3`	`+import numpy as np`
	`4`	`+from tkinter.filedialog import *`
	`5`	`+import tkinter as tk`
	`6`	`+`
	`7`	`+photo = askopenfilename()`
	`8`	`+img = cv2.imread(photo,cv2.IMREAD_GRAYSCALE)`
	`9`	`+blurred = cv2.medianBlur(img,5)`
	`10`	`+edges = cv2.Canny(blurred,50,150,apertureSize=3)`
	`11`	`+# the image read in grayscale format is blurred and edges are detected in it`
	`12`	`+circles = cv2.HoughCircles(edges,cv2.HOUGH_GRADIENT,1,30,param1=50,param2=31,minRadius=0,maxRadius=0)`
	`13`	`+# the parameters in the above function needs to be adjusted for each image accordingly.`
	`14`	`+circles = np.uint16(np.around(circles))`
	`15`	`+`
	`16`	`+colorImg = cv2.cvtColor(img,cv2.COLOR_GRAY2BGR)`
	`17`	`+for i in circles[0,:]:`
	`18`	`+ cv2.circle(colorImg,(i[0],i[1]),i[2],(0,255,0),2)`
	`19`	`+ # drawing the circle`
	`20`	`+ cv2.circle(colorImg,(i[0],i[1]),2,(0,0,255),3)`
	`21`	`+ # drawing its center`
	`22`	`+`
	`23`	`+cv2.imwrite('output.jpg',colorImg)`
	`24`	`+cv2.imshow('output',colorImg)`
	`25`	`+cv2.waitKey(2000)`
	`26`	`+cv2.destroyAllWindows()`

Lines changed: 25 additions & 0 deletions

Original file line number	Diff line number	Diff line change
`@@ -0,0 +1,25 @@`
	`1`	`+# Circle detection`
	`2`	`+This python script will allow us to identify circles in an image using Hough circle transform`
	`3`	`+`
	`4`	`+## Setup Instructions`
	`5`	`+### Install python3`
	`6`	`+sudo apt-get install python3`
	`7`	`+### Install pip (package installer for python)`
	`8`	`+sudo apt-get install python3-pip`
	`9`	`+### Install Numpy library with pip`
	`10`	`+pip3 install numpy`
	`11`	`+### Install OpenCV library with pip`
	`12`	`+pip3 install opencv-python`
	`13`	`+### Install tkinter library`
	`14`	`+sudo apt-get install python3-tk`
	`15`	`+`
	`16`	`+## Details/Output`
	`17`	`+User is prompted to select an image and this script detects circles in it using the hough circle transform.`
	`18`	`+`
	`19`	`+Note Each image needs its own set of parameters in cv2.HoughCircles() function. What worked for one might not work for another image.`
	`20`	`+minDist(minimum distance between two centers), param1(parameter 1), param2(parameter 2) are to be adjusted until we get desired output.`
	`21`	`+`
	`22`	`+The output image is written/stored in the current folder.`
	`23`	`+`
	`24`	`+## Author`
	`25`	`+Github: invigorzz313`

5.96 KB

17.8 KB

Comments

(0)