Demosaic JPEG Software

Demosaicing is a transformation of Bayer image into the conventional 24/48-bit RGB format. Demosaicing is required because digital cameras usually don’t produce ready-to-go RGB images, instead they store visual information as a set of separate R, B and G pixels derived from the image sensor of the camera, and the actual color of a pixel in that set is determined by interpolating nearby pixel colors. To high quality demosaicing algorithm we have added JPEG compression to fulfill the following:

High quality image demosaicing on NVIDIA GPU, much faster than on any CPU
Fast JPEG compression

What's unique in Demosaic JPEG Software

This is the fastest demosaic and jpeg compression tool
Sophisticated demosaicing algorithms with high PSNR and low artifacts
Simple to use, fast to work and easy to check
Created by company with years of CUDA experience and world-class achievements in the fileld

GPU JPEG Demosaicing Software Features

All CFA Bayer mosaic patterns for input data supported (RGGB, BGGR, GBRG, GRBG)
Read 8/16-bit input CFA image data from HDD/SSD/RAID in PGM format or from PC memory
Output format: 24/48-bit JPEG with variable compression ratio
Minimum image resolution 128×128 pixels
Maximum image resolution up to 16,000×16,000 pixels and more
Demosaic+JPEG performance without host-to-device and device-to-host transfers) on NVIDIA GeForce GTX 1080:
- 2 Mpix 8-bit image (DFPD demosaic, jpeg quality 90%, 4:4:4) – 0.57 ms
- 8 Mpix 8-bit image (DFPD demosaic, jpeg quality 90%, 4:4:4) – 1.7 ms
Strong color moire artifacts suppression
Timing and performance measurements
NVIDIA GPU with Compute Capability >= 3.0
Command-line application for Windows-7/8/10 and Linux Ubuntu/CentOS

Quality for combined Demosaic and JPEG

Both demosaicing and JPEG compression lead to image quality degradation. That's why if we apply these algorithms to image processing pipeline we could get significant quality loss for final image. It happens quite often because high quality demosaicing algorithms for CPU are very slow, so the choice of fast and simple demosaicing algorithm is usually straghtforward. But this is the way to get maximum image degradation.

In tables below we can see averaged PSNR/SSIM results for Kodak image set (total 24 images) which is widely used for quality evaluation in demosaicing. For testing we used standard Baseline JPEG algorithm and Bilinear/HQLI/AHD/DFPD demosaicing algorithms.

Demosaic and JPEG with subsampling 4:4:4

	Bilinear demosaic	HQLI demosaic	AHD demosaic	DFPD demosaic
Demosaic only, PSNR (dB) / SSIM	30.37 / 0.87	35.97 / 0.97	37.38 / 0.97	38.86 / 0.98
Demosaic + JPEG, quality 100%, PSNR (dB) / SSIM	29.36 / 0.87	34.52 / 0.96	36.88 / 0.96	38.24 / 0.97
Demosaic + JPEG, quality 95%, PSNR (dB) / SSIM	29.16 / 0.86	34.18 / 0.95	36.21 / 0.95	37.28 / 0.96
Demosaic + JPEG, quality 90%, PSNR (dB) / SSIM	29.13 / 0.86	33.83 / 0.94	35.53 / 0.94	36.30 / 0.95
Demosaic + JPEG, quality 85%, PSNR (dB) / SSIM	29.15 / 0.86	33.46 / 0.93	34.88 / 0.93	35.46 / 0.94
Demosaic + JPEG, quality 80%, PSNR (dB) / SSIM	29.19 / 0.86	33.05 / 0.92	34.30 / 0.92	34.74 / 0.93
Demosaic + JPEG, quality 75%, PSNR (dB) / SSIM	29.21 / 0.85	32.66 / 0.91	33.78 / 0.91	34.13 / 0.92

Demosaic and JPEG with subsampling 4:2:2

	Bilinear demosaic	HQLI demosaic	AHD demosaic	DFPD demosaic
Demosaic only, PSNR (dB) / SSIM	30.37 / 0.87	35.97 / 0.97	37.38 / 0.97	38.86 / 0.98
Demosaic + JPEG, quality 100%, PSNR (dB) / SSIM	29.86 / 0.88	34.75 / 0.96	36.77 / 0.96	37.96 / 0.97
Demosaic + JPEG, quality 95%, PSNR (dB) / SSIM	29.66 / 0.87	34.30 / 0.95	36.07 / 0.95	37.01 / 0.96
Demosaic + JPEG, quality 90%, PSNR (dB) / SSIM	29.66 / 0.87	33.86 / 0.94	35.38 / 0.94	36.05 / 0.95
Demosaic + JPEG, quality 85%, PSNR (dB) / SSIM	29.70 / 0.87	33.44 / 0.93	34.72 / 0.93	35.21 / 0.94
Demosaic + JPEG, quality 80%, PSNR (dB) / SSIM	29.72 / 0.87	33.00 / 0.92	34.12 / 0.92	34.49 / 0.93
Demosaic + JPEG, quality 75%, PSNR (dB) / SSIM	29.71 / 0.86	32.59 / 0.91	33.58 / 0.91	33.88 / 0.92

Demosaic and JPEG with subsampling 4:2:0

	Bilinear demosaic	HQLI demosaic	AHD demosaic	DFPD demosaic
Demosaic only, PSNR (dB) / SSIM	30.37 / 0.87	35.97 / 0.97	37.38 / 0.97	38.86 / 0.98
Demosaic + JPEG, quality 100%, PSNR (dB) / SSIM	30.46 / 0.90	34.87 / 0.97	36.59 / 0.96	37.66 / 0.97
Demosaic + JPEG, quality 95%, PSNR (dB) / SSIM	30.21 / 0.89	34.31 / 0.95	35.85 / 0.95	36.69 / 0.96
Demosaic + JPEG, quality 90%, PSNR (dB) / SSIM	30.21 / 0.88	33.85 / 0.94	35.13 / 0.94	35.72 / 0.95
Demosaic + JPEG, quality 85%, PSNR (dB) / SSIM	30.19 / 0.88	33.38 / 0.93	34.43 / 0.93	34.88 / 0.94
Demosaic + JPEG, quality 80%, PSNR (dB) / SSIM	30.14 / 0.87	32.91 / 0.92	33.83 / 0.92	34.17 / 0.93
Demosaic + JPEG, quality 75%, PSNR (dB) / SSIM	30.09 / 0.87	32.48 / 0.91	33.29 / 0.91	33.57 / 0.92

Now it's obvious to evaluate image degradation for combined JPEG+demosaic solution and make a right choice for both demosaic algorithm and JPEG quality. This is just an idea how to choose image processing algorithms and their parameters. Full solution should take into account compression ratio, total performance benchmarks, hardware pricing, and much more.

Bilinear demosaicing algorithm has good speed, though quality is moderate according to PSNR and SSIM metrics. Quality loss after bilinear algorithm is significant and JPEG compression doesn't really change PSNR/SSIM after demosaicing. That's why we would suggest to choose demosaicing from HQLI/AHD/DFPD or other high quality algorithms.

What else is interesting in Fastvideo GPU solutions

We are also developing fast image processing software on GPU for high speed and high resolution cameras in realtime: dark frame subtraction, flat field correction, white balance, image filtering, noise removal, color correction, demosaicing, jpeg/j2k compression, etc.