joachim/vehicle-counter
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Verbesserte Debug-Visualisierung für Linienschnitt

Browse Source 
Neue Debug-Features:
- Orange Bewegungslinien zeigen die Trajektorie jedes Fahrzeugs
- Größerer grüner Kreis (25px) beim Zählen für bessere Sichtbarkeit
- crossed_line() Helper-Funktion für bessere Code-Lesbarkeit

Hilft beim Debuggen:
- Bewegungslinien zeigen ob Fahrzeug sich bewegt
- Sichtbar ob Bewegungslinie die Zähllinie kreuzt
- Bestätigt visuell dass Tracking funktioniert

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
This commit is contained in:
Joachim Hummel
2025-12-08 16:09:26 +00:00
parent dcb171f093
commit 87a6060a1a
1 changed files with 26 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

32
app.py
View File

@@ -34,6 +34,20 @@ def line_intersect(p1, p2, p3, p4):
return 0 <= t <= 1 and 0 <= u <= 1
def ccw(A, B, C):
"""Check if three points are in counter-clockwise order"""
return (C[1] - A[1]) * (B[0] - A[0]) > (B[1] - A[1]) * (C[0] - A[0])
def crossed_line(prev_pos, curr_pos, line_start, line_end):
"""
Check if movement from prev_pos to curr_pos crossed the line.
Uses orientation check - more robust for frame skipping.
"""
# Check if the two line segments intersect
if line_intersect(prev_pos, curr_pos, line_start, line_end):
return True
return False
@app.route('/')
def index():
return render_template('index.html')
@@ -130,13 +144,16 @@ def detect_objects_from_webcam(line_data):
# If we have a previous position for this track
if track_id in track_positions and track_id not in counted_ids:
prev_x, prev_y = track_positions[track_id]
# Draw the movement line (orange/blue)
cv2.line(frame, (prev_x, prev_y), (center_x, center_y), (255, 100, 0), 2)
# Check if the vehicle crossed the counting line
if line_intersect((prev_x, prev_y), (center_x, center_y), line_start, line_end):
if crossed_line((prev_x, prev_y), (center_x, center_y), line_start, line_end):
counted_ids.add(track_id)
vehicle_count += 1
vehicle_type_counts[c] += 1
# Draw visual feedback when vehicle is counted (green circle)
cv2.circle(frame, (center_x, center_y), 15, (0, 255, 0), 3)
# Draw visual feedback when vehicle is counted (large green circle)
cv2.circle(frame, (center_x, center_y), 25, (0, 255, 0), 5)
# Update position
track_positions[track_id] = (center_x, center_y)
@@ -265,13 +282,16 @@ def detect_objects_from_video(video_path, line_data):
# If we have a previous position for this track
if track_id in track_positions and track_id not in counted_ids:
prev_x, prev_y = track_positions[track_id]
# Draw the movement line (orange/blue)
cv2.line(frame, (prev_x, prev_y), (center_x, center_y), (255, 100, 0), 2)
# Check if the vehicle crossed the counting line
if line_intersect((prev_x, prev_y), (center_x, center_y), line_start, line_end):
if crossed_line((prev_x, prev_y), (center_x, center_y), line_start, line_end):
counted_ids.add(track_id)
vehicle_count += 1
vehicle_type_counts[c] += 1
# Draw visual feedback when vehicle is counted (green circle)
cv2.circle(frame, (center_x, center_y), 15, (0, 255, 0), 3)
# Draw visual feedback when vehicle is counted (large green circle)
cv2.circle(frame, (center_x, center_y), 25, (0, 255, 0), 5)
# Update position
track_positions[track_id] = (center_x, center_y)