Info

Name

Toukir sabugar

Python Version

Description

AI ------------------------------

import cv2 import numpy as np import matplotlib.pyplot as plt from ultralytics import YOLO

def predict_on_image(model, img, conf): result = model(img, conf=conf)[0]

# Check if masks were detected
if result.masks is None:
    print("Warning: No masks detected in the image.")
    return None, None, None, None

# Detection
cls = result.boxes.cls.cpu().numpy()    # cls, (N, 1)
probs = result.boxes.conf.cpu().numpy()  # Confidence score, (N, 1)
boxes = result.boxes.xyxy.cpu().numpy()   # Box with xyxy format, (N, 4)

# Segmentation
masks = result.masks.data.cpu().numpy()     # Masks, (N, H, W)

# Resize each mask individually to match the original image dimensions
resized_masks = []
for mask in masks:
    resized_mask = cv2.resize(mask, (img.shape[1], img.shape[0]), interpolation=cv2.INTER_NEAREST)
    resized_masks.append(resized_mask)
masks = np.array(resized_masks)  # Convert list of masks back to numpy array (N, H, W)

return boxes, masks, cls, probs

def overlay(image, mask, color, alpha): color = color[::-1]

Ensure colored_mask has the same number of dimensions as image

colored_mask = np.expand_dims(mask, -1).repeat(3, axis=-1) * np.array(color)
# Blend the mask with the image
image_overlay = cv2.addWeighted(image, 1 - alpha, colored_mask.astype(np.uint8), alpha, 0)
return image_overlay

def draw_bounding_boxes(image, boxes, color): for box in boxes: x1, y1, x2, y2 = map(int, box) cv2.rectangle(image, (x1, y1), (x2, y2), color, 2)

def filter_and_merge_contours(contours, min_area=100):

Filter contours by area

filtered = [cnt for cnt in contours if cv2.contourArea(cnt) > min_area]

# List to hold merged contours
merged_contours = []
used = [False] * len(filtered)

for i in range(len(filtered)):
    if not used[i]:
        # Create a new contour that will be merged
        merged = filtered[i]
        used[i] = True

        for j in range(i + 1, len(filtered)):
            if not used[j]:
                # Check bounding box distance
                bbox1 = cv2.boundingRect(merged)
                bbox2 = cv2.boundingRect(filtered[j])
                dist = np.sqrt((bbox1[0] - bbox2[0])**2 + (bbox1[1] - bbox2[1])**2)

                if dist < 50:  # Distance threshold for merging
                    merged = np.vstack((merged, filtered[j]))
                    used[j] = True

        merged_contours.append(cv2.convexHull(merged))

return merged_contours

Load the model

model = YOLO('yolov8n-seg.pt')

Load the image using OpenCV

img_path = '/content/20240713_194215.jpg' img = cv2.imread(img_path)

Check if the image was loaded successfully

if img is None: raise ValueError(f"Image at path '{img_path}' could not be loaded. Please check the path and try again.")

Predict using YOLOv8

boxes, masks, cls, probs = predict_on_image(model, img, conf=0.4)

if masks is not None: # Check if masks were returned

Overlay masks on the original image

image_with_masks = np.copy(img)
for mask in masks:
    image_with_masks = overlay(image_with_masks, mask, color=(0, 255, 0), alpha=0.3)

# Draw bounding boxes on the image with masks
draw_bounding_boxes(image_with_masks, boxes if boxes is not None else [], color=(255, 0, 0))

# Convert image to HSV color space
hsv = cv2.cvtColor(img, cv2.COLOR_BGR2HSV)

# Define range for black color in HSV
lower_black = np.array([0, 0, 0])
upper_black = np.array([180, 255, 50])

# Threshold the HSV image to get only black colors
mask = cv2.inRange(hsv, lower_black, upper_black)

# Apply morphological operations to remove noise
kernel = np.ones((5, 5), np.uint8)
mask_cleaned = cv2.morphologyEx(mask, cv2.MORPH_CLOSE, kernel)
mask_cleaned = cv2.morphologyEx(mask_cleaned, cv2.MORPH_OPEN, kernel)

# Find contours in the mask
contours, hierarchy = cv2.findContours(mask_cleaned.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)

# Filter and merge contours
contours = filter_and_merge_contours(contours, min_area=50)  # Adjust min_area as needed

# Draw contours on the original image
rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
cv2.drawContours(rgb, contours, -1, (0, 255, 0), 2)

# Display the results
plt.figure(figsize=(12, 8))
plt.subplot(2, 2, 1)
plt.title('Original Image')
plt.imshow(cv2.cvtColor(img, cv2.COLOR_BGR2RGB))

plt.subplot(2, 2, 2)
plt.title('Mask')
plt.imshow(mask_cleaned, cmap='gray')

plt.subplot(2, 2, 3)
plt.title('Contours on Image')
plt.imshow(rgb)

plt.show()

print('Contours in the image: ', len(contours))

else: print("No masks detected.")

Non_AI ---------------

import cv2 import numpy as np import matplotlib.pyplot as plt

image = cv2.imread('/content/20240713_194215.jpg') gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY) plt.imshow(gray, cmap='gray');

blur = cv2.GaussianBlur(gray, (11,11), 0) plt.imshow(blur, cmap='gray')

canny = cv2.Canny(blur, 90, 200, 3) plt.imshow(canny, cmap='gray')

dilated = cv2.dilate(canny, (1,1), iterations = 22) plt.imshow(dilated, cmap='gray')

(cnt, heirarchy) = cv2.findContours(dilated.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_NONE) rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB) cv2.drawContours(rgb, cnt, -1, (0,255,0), 2)

plt.imshow(rgb)

print('items in the image: ', len(cnt)) plt.show()

perspectivLabs / CountingChallenge

Toukir #22