Spaces:
Sleeping
Sleeping
File size: 3,827 Bytes
993db33 8b83fe8 993db33 ea7c227 89530ea 993db33 c153693 8b83fe8 993db33 88cc205 993db33 c153693 993db33 c153693 993db33 c153693 993db33 c153693 993db33 c153693 993db33 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 |
import cv2
import spaces
import numpy as np
import gradio as gr
import tempfile
import os
os.environ['CUDA_VISIBLE_DEVICES'] = "0"
# Function to preprocess each frame
def preprocess_frame(frame):
resized_frame = cv2.resize(frame, (224, 224)) # Adjust size based on your model's input shape
normalized_frame = resized_frame / 255.0
return np.expand_dims(normalized_frame, axis=0) # Add batch dimension
# Function to draw pretty label on the frame
def draw_label(frame, label, position=(50, 50), font_scale=1, thickness=2):
# Define label properties
if label == 'Drowsy':
color = (0, 0, 255) # Red for Drowsy
bg_color = (0, 0, 100) # Darker background for Drowsy
else:
color = (0, 255, 0) # Green for Alert
bg_color = (0, 100, 0) # Darker background for Alert
font = cv2.FONT_HERSHEY_SIMPLEX
text_size = cv2.getTextSize(label, font, font_scale, thickness)[0]
# Define rectangle background dimensions
text_x, text_y = position
rect_start = (text_x, text_y - text_size[1] - 10) # Adjust y to account for text height
rect_end = (text_x + text_size[0] + 10, text_y + 10)
# Draw rectangle background
cv2.rectangle(frame, rect_start, rect_end, bg_color, -1)
# Add border around text
cv2.putText(frame, label, position, font, font_scale, (255, 255, 255), thickness + 2, lineType=cv2.LINE_AA)
# Add the main colored text
cv2.putText(frame, label, position, font, font_scale, color, thickness, lineType=cv2.LINE_AA)
@spaces.GPU(duration=120)
def predict_drowsiness(video_path):
# Open the video file
import tensorflow as tf
print(tf.config.list_physical_devices("GPU"))
model = tf.keras.models.load_model('cnn.keras')
cap = cv2.VideoCapture(video_path)
frame_width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
frame_height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
fps = int(cap.get(cv2.CAP_PROP_FPS))
# Calculate frame skipping interval based on 0.5 seconds
skip_interval = int(fps * 0.5) # Skip frames to achieve 1 frame every 0.5 seconds
# Create a temporary file for the output video
with tempfile.NamedTemporaryFile(suffix='.mp4', delete=False) as temp_output:
temp_output_path = temp_output.name
# Output video settings
out = cv2.VideoWriter(temp_output_path, cv2.VideoWriter_fourcc(*'mp4v'), fps, (frame_width, frame_height))
frame_count = 0
while cap.isOpened():
ret, frame = cap.read()
if not ret:
break
# Only process frames at the specified interval
if frame_count % skip_interval == 0:
# Preprocess frame
preprocessed_frame = preprocess_frame(frame)
# Use the model to predict drowsiness
prediction = model.predict(preprocessed_frame)
drowsiness = np.argmax(prediction)
# Add label to frame with improved visibility
label = 'Drowsy' if drowsiness == 0 else 'Alert'
draw_label(frame, label, position=(50, 50)) # Use the draw_label function
# Write the frame (whether labeled or not) to the output video
out.write(frame)
frame_count += 1
# Release resources
cap.release()
out.release()
return temp_output_path # Return the path to the temporary output video
# Gradio interface
interface = gr.Interface(
fn=predict_drowsiness,
inputs=gr.Video(), # Video input from webcam or upload
outputs="video", # Return a playable video with predictions
title="Drowsiness Detection in Video",
description="Upload a video or record one, and this tool will detect if the person is drowsy.",
)
# Launch the app
if __name__ == "__main__":
interface.launch() |