Create haar_classifier.py

This is the 'beginner's guide' to implementing a haar cascading classifier to detect a fact in an image.

Author: TechGirl007

computer_vision/haar_classifier.py

@@ -0,0 +1,68 @@
+"""
+Haar Cascading Classifiers
+
+Objective : Detect Object (Face)
+
+Resources Haar Classifiers :
+https://www.analyticsvidhya.com/blog/2022/04/object-detection-using-haar-cascade-opencv/#:~:text=Haar%20cascade%20is%20an%20algorithm,%2C%20buildings%2C%20fruits%2C%20etc
+https://machinelearningmastery.com/using-haar-cascade-for-object-detection/
+Resource for Haar classifiers: https://github.com/opencv/opencv/tree/master/data/haarcascades
+
+Download photo from :
+https://unsplash.com/
+
+1. Open Colab and create new notebook.
+2. Download a free to download photo from unsplash or use one of your own.
+3. Upload the photo in the model
+4.  Use haar classifier to detect face in image
+"""
+
+from google.colab.patches import cv2_imshow  #to assist with image processing and showing
+import cv2
+import numpy as np
+
+
+# Defining the list of image URLs from Unsplash
+foto = [
+    "https://spash.com/[chosen photo URL]"
+    "https://splash.com/[chosen photo URL]"
+]
+
+def load_image_from_url(url):
+    resp = urlopen(url)
+    image = np.asarray(bytearray(resp.read()), dtype="uint8")
+    image = cv2.imdecode(image, cv2.IMREAD_COLOR)
+    return image
+
+for i, image_url in enumerate(foto):
+    img = load_image_from_url(image_url)
+    print(f"Displaying image {i+1}")
+    cv2_imshow(img)
+#The above code should be in one cell and display the photo you chose.
+
+###########next cell##############
+
+# Loading the pre-trained Haar cascade for face detection
+face_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml') #you can use others but this one is used the most.  Resource link above.
+
+#detecting the face and covert to grayscale
+def detect_faces(image):
+    gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
+    faces = face_cascade.detectMultiScale(gray, scaleFactor=1.1, minNeighbors=20, minSize=(20,20)) #Setting general parameters.  Resouces above to understand more.
+    for (x, y, w, h) in faces:
+        cv2.rectangle(image, (x, y), (x+w, y+h), (255, 0, 0), 2) #blue colour bounding box
+    return image
+
+
+# Processing each loaded and resized image with Haar cascade
+for i, url in enumerate(foto, start=1):
+    img=load_image_from_url(url)
+    img_with_faces = detect_faces(img.copy())
+    print(f"Detecting Face {i}")
+    cv2_imshow(img_with_faces) #face
+
+#this should display the image with a blue bounding box.
+
+
+
+