diff --git a/MisplaceAI/core/static/core/fonts/Arial.ttf b/MisplaceAI/core/static/core/fonts/Arial.ttf
new file mode 100644
index 0000000000000000000000000000000000000000..ab68fb197d4479b3b6dec6e85bd5cbaf433a87c5
Binary files /dev/null and b/MisplaceAI/core/static/core/fonts/Arial.ttf differ
diff --git a/MisplaceAI/process_misplaced_manager/views.py b/MisplaceAI/process_misplaced_manager/views.py
index 3724d1075eec4ece8d58242ae8434705508a15dd..14918699cdb51cb52c7221227439e68ea3d2a07a 100644
--- a/MisplaceAI/process_misplaced_manager/views.py
+++ b/MisplaceAI/process_misplaced_manager/views.py
@@ -207,47 +207,32 @@ def display_video_results(request, video_id):
def process_video_for_misplaced_objects(video_path, frame_interval, frame_delay):
- print("Starting object detection for video:", video_path)
cap = cv2.VideoCapture(video_path)
- print("Video capture object created")
frame_width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
- print("Frame width:", frame_width)
frame_height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
- print("Frame height:", frame_height)
fps = int(cap.get(cv2.CAP_PROP_FPS))
- print("FPS:", fps)
-
misplaced_objects_all_frames = []
- print("Misplaced objects list created")
detected_objects_all_frames = []
- print("Detected objects list created")
-
frame_count = 0
annotated_frame_count = 1 # Start frame count from 1 for annotated frames
frame_interval_frames = frame_interval * fps
annotated_frames = []
while cap.isOpened():
- print(f"Processing frame {frame_count}")
ret, frame = cap.read()
- print("Frame read")
if not ret:
break
if frame_count % frame_interval_frames == 0:
- print("Frame read successfully")
image_np = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
- print("Frame converted to RGB")
# Convert the frame to PIL image
image_pil = Image.fromarray(image_np)
detected_objects = run_inference(detection_model, category_index, image_pil)
- print(f"Detected objects in frame {frame_count}: {detected_objects}")
placement_rules = PlacementRules()
misplaced_objects = placement_rules.check_placement(detected_objects)
- print(f"Misplaced objects in frame {frame_count}: {misplaced_objects}")
detected_objects_all_frames.append(detected_objects)
misplaced_objects_all_frames.append(misplaced_objects)
@@ -269,7 +254,6 @@ def process_video_for_misplaced_objects(video_path, frame_interval, frame_delay)
annotated_clip = ImageSequenceClip(annotated_frames, fps=1/frame_delay)
annotated_clip.write_videofile(output_video_path, fps=fps, codec='libx264', audio_codec='aac')
- print("Finished processing video:", output_video_path)
return detected_objects_all_frames, misplaced_objects_all_frames, output_video_path
diff --git a/MisplaceAI/results_viewer/utils.py b/MisplaceAI/results_viewer/utils.py
index 7a6f3a1a97d4add05a1703ca4b4359d7e437e927..6d3a8a759be862b143cc9030d018f010c5a7a25e 100644
--- a/MisplaceAI/results_viewer/utils.py
+++ b/MisplaceAI/results_viewer/utils.py
@@ -1,13 +1,10 @@
# MisplaceAi/results_viewer/utils.py
-from PIL import Image
import matplotlib.pyplot as plt
import matplotlib.patches as patches
-import os
-from item_detector.utils import run_inference
-from placement_rules.utils import PlacementRules
from PIL import Image, ImageDraw, ImageFont
-
-import cv2
+import os
+from django.conf import settings
+
def visualize_misplaced_objects(image_path, detected_objects, misplaced_objects):
"""Visualize misplaced objects with annotations."""
@@ -61,27 +58,37 @@ def visualize_misplaced_objects(image_path, detected_objects, misplaced_objects)
-
-
def visualize_pil_misplaced_objects(image_pil, detected_objects, misplaced_objects, frame_number):
"""Visualize misplaced objects with annotations on a PIL image."""
+
+ # Create a drawing context for the image
draw = ImageDraw.Draw(image_pil)
+
+ # Get the dimensions of the image
width, height = image_pil.size
+ # Create a list of class names for misplaced objects
misplaced_names = [obj["class_name"] for obj in misplaced_objects]
- # Draw frame number on the top-left corner with a larger font size
- frame_text = f"Frame {frame_number}"
- font_size = 300 # Set a larger font size
+ # Load a font using absolute path to the static directory
+ font_size = 40 # Set the font size (increase if needed)
+ font_path = os.path.join(settings.BASE_DIR, 'core/static/core/fonts/Arial.ttf') # Path to the font file
try:
- font = ImageFont.truetype("arial.ttf", font_size)
+ font = ImageFont.truetype(font_path, font_size)
+ print("Font loaded successfully.")
except IOError:
- font = ImageFont.load_default() # Fallback to default font
+ print("Font file not found. Using default font.")
+ font = ImageFont.load_default()
+ # Draw frame number on the top-left corner
+ frame_text = f"Frame {frame_number}"
+ text_width, text_height = draw.textsize(frame_text, font=font)
+ draw.rectangle([(0, 0), (text_width + 20, text_height + 20)], fill="black")
draw.text((10, 10), frame_text, fill="yellow", font=font)
+ # Iterate over all detected objects to draw their annotations
for obj in detected_objects:
ymin, xmin, ymax, xmax = [
obj["ymin"] * height,
@@ -91,7 +98,26 @@ def visualize_pil_misplaced_objects(image_pil, detected_objects, misplaced_objec
]
color = "green" if obj["class_name"] not in misplaced_names else "red"
- draw.rectangle([xmin, ymin, xmax, ymax], outline=color, width=2)
- draw.text((xmin, ymin), f"{'Misplaced: ' if obj['class_name'] in misplaced_names else ''}{obj['class_name']}", fill=color, font=font)
+ draw.rectangle([xmin, ymin, xmax, ymax], outline=color, width=3)
+
+ text = f"{'Misplaced: ' if obj['class_name'] in misplaced_names else ''}{obj['class_name']}"
+ text_width, text_height = draw.textsize(text, font=font)
+
+ # Check available space above and below the bounding box
+ space_above = ymin - text_height - 10
+ space_below = height - ymax - text_height - 10
+
+ if space_above > 0:
+ # Place the label above the bounding box
+ draw.rectangle([(xmin, ymin - text_height - 10), (xmin + text_width + 10, ymin)], fill="black")
+ draw.text((xmin + 5, ymin - text_height - 5), text, fill=color, font=font)
+ elif space_below > 0:
+ # Place the label below the bounding box
+ draw.rectangle([(xmin, ymax), (xmin + text_width + 10, ymax + text_height + 10)], fill="black")
+ draw.text((xmin + 5, ymax + 5), text, fill=color, font=font)
+ else:
+ # Place the label inside the bounding box as a fallback
+ draw.rectangle([(xmin, ymin), (xmin + text_width + 10, ymin + text_height + 10)], fill="black")
+ draw.text((xmin + 5, ymin + 5), text, fill=color, font=font)
return image_pil