PopularMajor Project
Artificial Intelligence
Face Recognition Attendance System
Camera-based attendance with liveness detection and analytics
- Python
- OpenCV
- FaceNet
- Flask
- +2
₹4,900₹5,999
4.8/301 reviews
Artificial Intelligence · Major Project
Image Dehazing with Deep Learning
Removes haze and fog, dark channel prior compared against a CNN
3-4 weeks build
Overview
Implements the atmospheric scattering model twice: once with the classical dark channel prior and guided-filter refinement, and once with a lightweight AOD-Net-style CNN trained on the RESIDE dataset. Both are scored with PSNR and SSIM on synthetic haze and judged visually on real foggy road and city photos where no ground truth exists.
What makes it stand up
- 01Dark channel prior with guided-filter transmission refinement
- 02Lightweight end-to-end dehazing CNN trained on RESIDE
- 03Evaluation on both synthetic and real-world hazy images
- 04Video dehazing mode for dashcam-style clips
- 05Runtime comparison of both methods in the report
Module breakdown
- 01
Scattering model
Transmission and atmospheric light estimation explained in code.
- 02
Dark channel prior
Classical pipeline with guided-filter refinement.
- 03
Dehazing CNN
Compact network trained on RESIDE indoor and outdoor sets.
- 04
Video mode
Frame-by-frame processing with temporal smoothing.
- 05
Evaluation
PSNR, SSIM and timing tables with visual comparisons.
After this, you will be able to
- Explain the atmospheric scattering equation behind haze
- Discuss where the dark channel prior fails, such as bright skies
- Demo dehazing on a foggy photo taken on campus
Technology stack
- Python
- PyTorch
- OpenCV
- NumPy
- Streamlit
- Level
- Major Project
- Domain
- Artificial Intelligence
- Build time
- 3-4 weeks
- Difficulty
You receive
- Complete, commented source code
- 60-90 page project report (IEEE format)
- Editable presentation deck
- Architecture, ER and UML diagrams
- Local setup and deployment guide
- Sample dataset or seed data
- Viva question bank with answers
- One 45-minute walkthrough call
- 30 days of doubt-clearing support