Task 3 : Training the ML model

Task 1/README.md

@@ -0,0 +1,15 @@
+# TASK 1 : 
+# Use any 3 functions/modules in OpenCV, and commit the code in the official Expressando TDoC 2021 Repository.
+
+## Modules Imported
+    + OpenCV
+    + Numpy
+    + MatplotLib
+
+## Operations Performed:
+    + Importing / Reading Images
+    + Resizing Images
+    + Grayscaling the Image
+    + Bluring the Image
+    + Simple thresholding
+

Task 1/Task1.py

@@ -0,0 +1,74 @@
+# TASK 1 : Use any 3 functions/modules in OpenCV,
+# and commit the code in the official Expressando TDoC 2021 Repository.
+
+
+# Importing Modules
+import cv2 as cv
+import numpy as np
+from matplotlib import pyplot as plt
+
+# Reading images and Resizing it as per needs
+
+img = cv.imread('image.jpg')
+width, height = int(img.shape[1] / 2.5), \
+                int(img.shape[0] / 2.5)  # Dynamically allotting the width and height of the image
+
+img = cv.resize(img, (width, height), cv.INTER_CUBIC)  # Resizing the image as per our need
+
+cv.imshow('Flower - Original', img)
+
+# ------------------------ + OPERATIONS ON THE IMPORTED IMAGE + ---------------------------------------------------
+
+# 1. Grayscaling the Image
+img_grey = cv.cvtColor(img, cv.COLOR_BGR2GRAY)
+# cv.imshow('Grayscaled Image', img_grey)
+
+# 2. Bluring Image
+img_blur_simple = cv.blur(img, (5, 5))
+img_blur_Gaussian = cv.GaussianBlur(img, (5, 5), 0)
+# cv.imshow('Simple Blur Image', img_blur_simple)
+# cv.imshow('Gaussian Blur Image', img_blur_Gaussian)
+
+
+titles = ['Original Image', 'Gray Scale', 'Simple Blur Image', 'Gaussian Blur Image']
+images = [img, img_grey, img_blur_simple, img_blur_Gaussian]
+
+# Displaying the ORIGINAL IMAGE, GRAY SCALE IMAGE, SIMPLE BLUR IMAGE, GAUSSIAN BLUR IMAGE
+for i in range(4):
+    plt.subplot(2, 2, i + 1), plt.imshow(images[i], 'gray')
+    plt.title(titles[i])
+    plt.xticks([]), plt.yticks([])
+
+plt.show()
+
+# 3. Simple Threshold
+img_gfg = cv.imread('gfg.jpg', 1)
+img_gfg_grey = cv.cvtColor(img_gfg, cv.COLOR_BGR2GRAY)
+
+ret_1, img_thresh1 = cv.threshold(img_gfg_grey, 120, 255, cv.THRESH_BINARY)
+ret_2, img_thresh2 = cv.threshold(img_gfg_grey, 120, 255, cv.THRESH_BINARY_INV)
+ret_3, img_thresh3 = cv.threshold(img_gfg_grey, 120, 255, cv.THRESH_TOZERO)
+ret_4, img_thresh4 = cv.threshold(img_gfg_grey, 120, 255, cv.THRESH_TOZERO_INV)
+ret_5, img_thresh5 = cv.threshold(img_gfg_grey, 120, 255, cv.THRESH_TRUNC)
+
+# cv.imshow('Original', img_gfg)
+# cv.imshow('Binary Thresh', img_thresh1)
+# cv.imshow('Binary Inverse Thresh', img_thresh2)
+# cv.imshow('Tozero Thresh', img_thresh3)
+# cv.imshow('Tozero Inverse Thresh', img_thresh4)
+# cv.imshow('Trunc Thresh', img_thresh5)
+
+# Displaying the ORIGINAL IMAGE, BINARY THRESHOLD IMAGE, BINARY INVERSE THRESHOLD IMAGE,
+# TOZERO IMAGE, TOZERO INVERSE IMAGE, TRUNC IMAGE
+titles = ['Original Image', 'Binary Thresh', 'Binary Inverse Thresh', 'Tozero Thresh', 'Tozero Inverse Thresh',
+          'Trunc Thresh']
+images = [img_gfg_grey, img_thresh1, img_thresh2, img_thresh3, img_thresh4, img_thresh5]
+
+for i in range(6):
+    plt.subplot(3, 2, i + 1), plt.imshow(images[i], 'gray')
+    plt.title(titles[i])
+    plt.xticks([]), plt.yticks([])
+
+plt.show()
+cv.waitKey(0)
+cv.destroyAllWindows()

Task 2/Task 2.py

@@ -0,0 +1,94 @@
+# Task 2
+
+# Modules Imported
+import cv2
+import os
+import numpy as np
+
+# Checking if the said folders are already present there in the Python file directory or not,
+# and if it's not there then the said folders are created in the same directory.
+if not os.path.exists("data"):
+    os.makedirs("data")
+    os.makedirs("data/train")
+    os.makedirs("data/train/0")
+    os.makedirs("data/train/1")
+    os.makedirs("data/train/2")
+    os.makedirs("data/train/3")
+    os.makedirs("data/train/4")
+    os.makedirs("data/train/5")
+
+mode = 'train'
+directory = 'data/' + mode + '/'
+
+# Capturing video feed from the first web cam connected to the system
+cap = cv2.VideoCapture(0)
+
+while True:
+    _, frame = cap.read()
+    frame = cv2.flip(frame, 1)  # flipping the frame for better visual
+
+    # Putting text onto the frame captured
+    cv2.putText(frame, "SHIRSENDU KONER - TDOC 2021", (175, 460),
+                cv2.FONT_HERSHEY_COMPLEX_SMALL, 1, (0, 150, 255), 2)
+
+    # Creating a dictionary to count the number of images for each finger image captured
+    count = {'zero': len(os.listdir(directory + "/0")),
+             'one': len(os.listdir(directory + "/1")),
+             'two': len(os.listdir(directory + "/2")),
+             'three': len(os.listdir(directory + "/3")),
+             'four': len(os.listdir(directory + "/4")),
+             'five': len(os.listdir(directory + "/5"))}
+
+    # Putting Labels onto the Frame directly
+    cv2.putText(frame, "MODE : " + mode, (10, 50), cv2.FONT_HERSHEY_COMPLEX_SMALL, 1, (225, 255, 255), 1)
+    cv2.putText(frame, "IMAGE COUNT", (10, 100), cv2.FONT_HERSHEY_COMPLEX_SMALL, 1, (225, 255, 255), 1)
+    cv2.putText(frame, "ZERO  : " + str(count['zero']), (10, 120), cv2.FONT_HERSHEY_COMPLEX_SMALL,
+                1, (255, 255, 255), 1)
+    cv2.putText(frame, "ONE   : " + str(count['one']), (10, 140), cv2.FONT_HERSHEY_COMPLEX_SMALL,
+                1, (255, 255, 255), 1)
+    cv2.putText(frame, "TWO   : " + str(count['two']), (10, 160), cv2.FONT_HERSHEY_COMPLEX_SMALL,
+                1, (255, 255, 255), 1)
+    cv2.putText(frame, "THREE : " + str(count['three']), (10, 180), cv2.FONT_HERSHEY_COMPLEX_SMALL,
+                1, (255, 255, 255), 1)
+    cv2.putText(frame, "FOUR  : " + str(count['four']), (10, 200), cv2.FONT_HERSHEY_COMPLEX_SMALL,
+                1, (255, 255, 255), 1)
+    cv2.putText(frame, "FIVE   : " + str(count['five']), (10, 220), cv2.FONT_HERSHEY_COMPLEX_SMALL,
+                1, (255, 255, 255), 1)
+
+    # Separating the Region of Interest for our work
+    # (x1, y1) and (x2, y2) are the two diagonal corner points or coordinates of the rectangle drawn
+    x1 = int(0.5 * frame.shape[1])  # frame.shape[1] gives the width of the frame
+    y1 = 10
+    x2 = frame.shape[1] - 10
+    y2 = int(0.5 * frame.shape[1])
+    cv2.rectangle(frame, (x1 - 1, y1 - 1), (x2 + 1, y2 + 1), (0, 0, 255), 3)  # Drawing a rectangle onto the frame
+    roi = frame[y1:y2, x1:x2]  # slicing a part of the frame
+    roi = cv2.resize(roi, (200, 200))  # resizing the sliced part
+    cv2.putText(frame, "R.O.I", (440, 350), cv2.FONT_HERSHEY_COMPLEX_SMALL, 1, (50, 225, 150), 2)
+    cv2.imshow("Frame", frame)
+
+    roi = cv2.cvtColor(roi, cv2.COLOR_BGR2GRAY)  # Changing Colour Space of our R.O.I
+    # i.e. color frame -> black and white frame
+    # rev, roi = cv2.threshold(roi, 120, 255, cv2.THRESH_BINARY)
+    roi = cv2.adaptiveThreshold(roi, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY, 199, 5)
+    cv2.imshow("ROI", roi)
+
+    # Images are captured and stored when respective keys are pressed
+    interrupt = cv2.waitKey(10)
+    if interrupt & 0xFF == 27:  # 27 refers to the ESC key on our keyboard
+        break
+    if interrupt & 0xFF == ord('0'):  # when 0 is pressed
+        cv2.imwrite(directory + '0/' + str(count['zero']) + '.jpg', roi)
+    if interrupt & 0xFF == ord('1'):  # when 1 is pressed
+        cv2.imwrite(directory + '1/' + str(count['one']) + '.jpg', roi)
+    if interrupt & 0xFF == ord('2'):  # when 2 is pressed
+        cv2.imwrite(directory + '2/' + str(count['two']) + '.jpg', roi)
+    if interrupt & 0xFF == ord('3'):  # when 3 is pressed
+        cv2.imwrite(directory + '3/' + str(count['three']) + '.jpg', roi)
+    if interrupt & 0xFF == ord('4'):  # when 4 is pressed
+        cv2.imwrite(directory + '4/' + str(count['four']) + '.jpg', roi)
+    if interrupt & 0xFF == ord('5'):  # when 5 is pressed
+        cv2.imwrite(directory + '5/' + str(count['five']) + '.jpg', roi)
+
+cap.release()
+cv2.destroyAllWindows()