Merge pull request #2598 from invigorzz313/borderextract

border extraction

Author: avinashkranjan

Border Extraction/Border_extraction.py

@@ -0,0 +1,32 @@
+import cv2
+import numpy as np
+import tkinter as tk
+from tkinter.filedialog import *
+
+window = tk.Tk()
+window.title("Border extraction")
+window.geometry('300x100')
+label = tk.Label(window, text="Choose an image").grid(row=0, column=0)
+
+photo = askopenfilename()
+img = cv2.imread(photo)
+n,l,m = img.shape
+size = (n,l)
+gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
+
+threshold_value = 20 # this value needs to be adjusted for every image
+ret, thresh = cv2.threshold(gray, threshold_value, 255, 0)
+
+kernel = np.ones((5,5),np.uint8)
+# dilation and erosion are applied to binary images
+dilated = cv2.dilate(thresh, kernel, iterations=1)
+eroded = cv2.erode(thresh, kernel, iterations=1)
+
+bordered = dilated - eroded
+bordered = cv2.resize(bordered, size)
+cv2.imshow("Border/outline", bordered)
+cv2.imwrite("Output.png", bordered)
+cv2.waitKey(0)
+cv2.destroyAllWindows()
+
+window.mainloop()

Border Extraction/Output.png

@@ -0,0 +1,22 @@
+# Border extraction
+This python script will allow us to extract borders of objects in an image.
+
+## Setup Instructions
+### Install python3
+sudo apt-get install python3
+### Install pip (package installer for python)
+sudo apt-get install python3-pip
+### Install Numpy library with pip
+pip3 install numpy
+### Install OpenCV library with pip
+pip3 install opencv-python
+### Install tkinter library
+sudo apt-get install python3-tk
+
+## Details/Output
+User should select an image and the script returns an output image of borders of the objects in it.  
+The input image is converted into binary format and then we use morphological transformation of dilation minus erosion.  
+**Note** The threshold value while converting an image into binary format needs to be adjusted for every image.
+
+## Author
+Github: invigorzz313