Traffic_Sign_Recognition & CV_YOLO_Object_Detection_and_Trajectory

CV_YOLO_Object_Detection_and_Trajectory/Readme.md

@@ -0,0 +1,24 @@
+# CV Workshop
+
+### Verify python version
+```
+python --version
+```
+### Steps to Debug python path issue 
+1. **Open Environment Variables**
+    - Press ``Win`` and search for Environment Variables
+    - Click on Edit the system environment variables.
+    - In the System Properties window, click the Environment Variables button.
+2. **Check path**
+    - Under the User variables section, find and select the ``Path`` variable, then click Edit.
+    - Select python path and press ``move up`` till it gets at the top.
+
+### Install ultralytics
+```
+pip install ultralytics
+```
+
+### Colour picker 
+- [here](https://colorpicker.me/#b0b0b2)
+
+

CV_YOLO_Object_Detection_and_Trajectory/trackbar.py

@@ -0,0 +1,54 @@
+import cv2
+import numpy as np
+
+
+# cap=cv2.VideoCapture("OpenCV\\Res\\ball_for_tracking.mov")
+img = cv2.imread(r"Res\bit1.png")
+img = cv2.resize(img, (700, 500))
+
+
+def Track(x):
+    pass
+
+
+cv2.namedWindow("Color detectors")
+cv2.resizeWindow("Color detectors", 500, 500)
+cv2.createTrackbar("Upper Hue", "Color detectors", 153, 180, Track)
+cv2.createTrackbar("Upper Saturation", "Color detectors", 255, 255, Track)
+cv2.createTrackbar("Upper Value", "Color detectors", 255, 255, Track)
+cv2.createTrackbar("Lower Hue", "Color detectors", 64, 180, Track)
+cv2.createTrackbar("Lower Saturation", "Color detectors", 72, 255, Track)
+cv2.createTrackbar("Lower Value", "Color detectors", 49, 255, Track)
+
+
+# imgs=cv2.resize(img,(500,500))
+while True:
+    # success,img = cap.read()
+    hsv = cv2.cvtColor(img, cv2.COLOR_BGR2HSV)
+    u_hue = cv2.getTrackbarPos("Upper Hue", "Color detectors")
+    u_saturation = cv2.getTrackbarPos("Upper Saturation", "Color detectors")
+    u_value = cv2.getTrackbarPos("Upper Value", "Color detectors")
+
+    l_hue = cv2.getTrackbarPos("Lower Hue", "Color detectors")
+    l_saturation = cv2.getTrackbarPos("Lower Saturation", "Color detectors")
+    l_value = cv2.getTrackbarPos("Lower Value", "Color detectors")
+
+    print(
+        f"upper\n",
+        {u_hue, u_saturation, u_value},
+        "\n lower",
+        {l_hue, l_saturation, l_value},
+    )
+
+    Upper_hsv = np.array([u_hue, u_saturation, u_value])
+    Lower_hsv = np.array([l_hue, l_saturation, l_value])
+    mask = cv2.inRange(hsv, Lower_hsv, Upper_hsv)
+    res = cv2.bitwise_and(img, img, mask=mask)
+    res = cv2.medianBlur(res, 5)
+
+    cv2.imshow("video", img)
+    cv2.imshow("res", res)
+    if cv2.waitKey(1) & 0xFF == ord("x"):
+        break
+
+cv2.destroyAllWindows

CV_YOLO_Object_Detection_and_Trajectory/tracking.py

@@ -0,0 +1,36 @@
+import cv2
+import numpy as np
+
+video = cv2.VideoCapture(r"C:\Users\mahek\OneDrive\Desktop\CV Workshop\CV_Workshop-main\res\ball_for_tracking.mov")
+cposl = []
+while True:
+    ret, img = video.read()
+    if not ret:
+        break
+
+    hsv = cv2.cvtColor(img, cv2.COLOR_BGR2HSV)
+
+    lower_bound = np.array([0, 170, 90])
+    upper_bound = np.array([4, 255, 255])
+
+    mask = cv2.inRange(hsv,lower_bound,upper_bound)
+    x, y, w, h = cv2.boundingRect(mask)  
+    cv2.rectangle(img, (x - 20, y - 20), (x + w + 20, y + h + 20), (0, 255, 0), 1)
+
+
+
+    center =( ( x+w ), (y + h))
+
+    cposl.append(center)
+    for pos in cposl:
+        cv2.circle(img, pos,2, (255,255,255), -1)
+
+    cv2.imshow("video", img)
+    #cv2.imshow("mask", mask)
+
+    if cv2.waitKey(1) & 0xFF == ord("x"):
+        break
+
+
+
+cv2.destroyAllWindows()

CV_YOLO_Object_Detection_and_Trajectory/yolo.py

@@ -0,0 +1,7 @@
+from ultralytics import YOLO
+
+# Load a pretrained YOLO11n model
+model = YOLO("yolo11n.pt")
+
+# Run model
+model.predict(source=0,conf=0.5,show=True)

TARZAN task 2/Greyscale.py

@@ -0,0 +1,46 @@
+import cv2
+import numpy as np
+
+# Step 1: Read the  image (following the dimensions: 800x600x3)
+image = cv2.imread("photo.jpg")  
+
+ # orignal image display
+cv2.imshow("Original ", image) 
+
+cv2.waitKey(0)  
+cv2.destroyAllWindows()
+
+ # image ko Resize kiya -> 800*600
+image = cv2.resize(image, (800, 600)) 
+cv2.imshow("Resized", image)
+
+cv2.waitKey(0) 
+cv2.destroyAllWindows()
+
+# Step 2: Convert to grayscale (800x600)
+gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
+cv2.imshow("Grayscale", gray_image)
+cv2.waitKey(0)  
+cv2.destroyAllWindows()
+
+# Step 3: Reduce resolution (Resizing grayscale image)
+low_res_gray_image = cv2.resize(gray_image, (128, 128), interpolation=cv2.INTER_LINEAR)
+
+cv2.imshow("Reduced Resolution Grayscale", low_res_gray_image)
+cv2.waitKey(0)
+cv2.destroyAllWindows()
+
+
+# Step 4: Normalize the image (not grayscale) to range [0,1]
+normalized_image = cv2.cvtColor(low_res_gray_image, cv2.COLOR_GRAY2BGR)
+cv2.imshow("Normalized [0-1]", (normalized_image * 255).astype(np.uint8))
+cv2.waitKey(0)
+cv2.destroyAllWindows()
+
+
+# Step 5: Optionally normalize to range [-1,1]
+normalized_minus1_to_1 = (normalized_image - 0.5) * 2  # Scale to [-1,1]
+cv2.imshow("Normalized [-1,1]", normalized_minus1_to_1)
+
+cv2.waitKey(0)
+cv2.destroyAllWindows()

TARZAN task 2/Image proccessing optimization.py

@@ -0,0 +1,59 @@
+import cv2
+import numpy as np
+
+# Step 1: Read the  image (following the dimensions: 800x600x3)
+image = cv2.imread("photo.jpg")  
+
+ # orignal image display
+cv2.imshow("Original ", image) 
+
+cv2.waitKey(0) 
+cv2.destroyAllWindows()
+
+ # image ko Resize kiya -> 800*600
+image = cv2.resize(image, (800, 600)) 
+cv2.imshow("Resized", image)
+
+cv2.waitKey(0)  
+cv2.destroyAllWindows()
+
+# Step 2: Convert to grayscale (800x600)
+gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
+cv2.imshow("Grayscale", gray_image)
+
+cv2.waitKey(0)  
+cv2.destroyAllWindows()
+
+
+# Step 3: Reduce resolution (Resizing grayscale image)
+low_res_gray_image = cv2.resize(gray_image, (128, 128), interpolation=cv2.INTER_LINEAR)
+cv2.imshow("Reduced Resolution Grayscale", low_res_gray_image)
+
+cv2.waitKey(0)
+cv2.destroyAllWindows()
+
+# Step 3: Reduce resolution (Resizing to lower size for faster processing)
+low_res_image = cv2.resize(image, (128, 128), interpolation=cv2.INTER_LINEAR)
+cv2.imshow("Reduced Resolution", low_res_image)
+
+cv2.waitKey(0)  
+cv2.destroyAllWindows()
+
+# Step 4: Normalize the image (not grayscale) to range [0,1]
+normalized_image = low_res_image.astype(np.float32) / 255.0
+cv2.imshow("Normalized [0-1]", normalized_image) 
+
+cv2.waitKey(0)  
+cv2.destroyAllWindows()
+
+# Step 5: Optionally normalize to range [-1,1]
+normalized_minus1_to_1 = (normalized_image - 0.5) * 2  # Scale to [-1,1]
+cv2.imshow("Normalized [-1,1]", normalized_minus1_to_1)
+cv2.waitKey(0)  
+cv2.destroyAllWindows()
+
+
+
+
+
+