Watermark Embedding in Digital Camera Firmware. Peter Meerwald, May 28, 2008

Similar documents
Watermarking of Raw Digital Images in Camera Firmware: Embedding and Detection

IMPROVEMENTS ON SOURCE CAMERA-MODEL IDENTIFICATION BASED ON CFA INTERPOLATION

Survey On Passive-Blind Image Forensics

Camera identification from sensor fingerprints: why noise matters

Introduction to Video Forgery Detection: Part I

IDENTIFYING DIGITAL CAMERAS USING CFA INTERPOLATION

Forensic Framework. Attributing and Authenticating Evidence. Forensic Framework. Attribution. Forensic source identification

Implementation of a Visible Watermarking in a Secure Still Digital Camera Using VLSI Design

Secure Digital Camera

VISUAL sensor technologies have experienced tremendous

Watermarking-based Image Authentication with Recovery Capability using Halftoning and IWT

A Study on Resource Efficient Digital Multimedia Security Measures in Mobile Devices

Analysis on Color Filter Array Image Compression Methods

Applying the Sensor Noise based Camera Identification Technique to Trace Origin of Digital Images in Forensic Science

Color Filter Array Interpolation Using Adaptive Filter

Different-quality Re-demosaicing in Digital Image Forensics

Simultaneous Capturing of RGB and Additional Band Images Using Hybrid Color Filter Array

Digital Cameras The Imaging Capture Path

Camera Model Identification Framework Using An Ensemble of Demosaicing Features

A Unified Framework for the Consumer-Grade Image Pipeline

Digital photography , , Computational Photography Fall 2017, Lecture 2

A STUDY ON THE PHOTO RESPONSE NON-UNIFORMITY NOISE PATTERN BASED IMAGE FORENSICS IN REAL-WORLD APPLICATIONS. Yu Chen and Vrizlynn L. L.

Artifacts Reduced Interpolation Method for Single-Sensor Imaging System

Chapter 2-Digital Components

Imaging Sensor Noise as Digital X-Ray for Revealing Forgeries

Demosaicing Algorithms

Camera Image Processing Pipeline

ity Multimedia Forensics and Security through Provenance Inference Chang-Tsun Li

Detection of Image Forgery was Created from Bitmap and JPEG Images using Quantization Table

CMOS MT9D112 Camera Module 1/4-Inch 3-Megapixel Module Datasheet

Detection of Hue Modification Using Photo Response Non-Uniformity

Exposing Digital Forgeries from JPEG Ghosts

Cameras. Outline. Pinhole camera. Camera trial #1. Pinhole camera Film camera Digital camera Video camera

Digital Image Authentication from Thumbnails

Cameras. Shrinking the aperture. Camera trial #1. Pinhole camera. Digital Visual Effects Yung-Yu Chuang. Put a piece of film in front of an object.

1982 IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, VOL. 24, NO. 11, NOVEMBER 2014

Multimedia Forensics

Color image Demosaicing. CS 663, Ajit Rajwade

Improved sensitivity high-definition interline CCD using the KODAK TRUESENSE Color Filter Pattern

Image Forgery Detection Using Svm Classifier

An evaluation of debayering algorithms on GPU for real-time panoramic video recording

COLOR demosaicking of charge-coupled device (CCD)

The Raw Deal Raw VS. JPG

High Dynamic Range image capturing by Spatial Varying Exposed Color Filter Array with specific Demosaicking Algorithm

Digital photography , , Computational Photography Fall 2018, Lecture 2

PRIOR IMAGE JPEG-COMPRESSION DETECTION

Forensic Classification of Imaging Sensor Types

Retrieval of Large Scale Images and Camera Identification via Random Projections

SOURCE CAMERA IDENTIFICATION BASED ON SENSOR DUST CHARACTERISTICS

IMAGE RESTORATION WITH NEURAL NETWORKS. Orazio Gallo Work with Hang Zhao, Iuri Frosio, Jan Kautz

Fragile Sensor Fingerprint Camera Identification

Demosaicing Algorithm for Color Filter Arrays Based on SVMs

Image acquisition. In both cases, the digital sensing element is one of the following: Line array Area array. Single sensor

Source Camera Model Identification Using Features from contaminated Sensor Noise

Cameras As Computing Systems

Sapna Sameriaˡ, Vaibhav Saran², A.K.Gupta³

Colour-Decoupled Photo Response Non-Uniformity for. Digital Image Forensics

Local prediction based reversible watermarking framework for digital videos

Fragile Watermarking With Error-Free Restoration Capability Xinpeng Zhang and Shuozhong Wang

2018 IEEE Signal Processing Cup: Forensic Camera Model Identification Challenge

THE DIFFERENCE MAKER COMPARISON GUIDE

Design and Simulation of Optimized Color Interpolation Processor for Image and Video Application

Image Forgery Identification Using JPEG Intrinsic Fingerprints

Cameras. Digital Visual Effects, Spring 2008 Yung-Yu Chuang 2008/2/26. with slides by Fredo Durand, Brian Curless, Steve Seitz and Alexei Efros

TECHNICAL DOCUMENTATION

89% Gold Award. Sep 14, 2016 Oct 16, Aug 25, 2016 Jul 25, 2017 Oct 25, Mid-size SLR Mid-size SLR SLR-style mirrorless

DIGITAL DOCTORED VIDEO FORGERY DETECTION TECHNIQUES

Distinguishing between Camera and Scanned Images by Means of Frequency Analysis

A High Definition Motion JPEG Encoder Based on Epuma Platform

Product Introduction

Scanner Identification Using Sensor Pattern Noise

Camera identification by grouping images from database, based on shared noise patterns

Correlation Based Image Tampering Detection

Lossless Image Watermarking for HDR Images Using Tone Mapping

Introduction. Prof. Lina Karam School of Electrical, Computer, & Energy Engineering Arizona State University

Forensic Hash for Multimedia Information

ABC: Enabling Smartphone Authentication with Built-in Camera

COLOR IMAGE STEGANANALYSIS USING CORRELATIONS BETWEEN RGB CHANNELS. 1 Nîmes University, Place Gabriel Péri, F Nîmes Cedex 1, France.

Noise Reduction in Raw Data Domain

Can We Trust Digital Image Forensics?

Image Tampering Localization via Estimating the Non-Aligned Double JPEG compression

Efficient Estimation of CFA Pattern Configuration in Digital Camera Images

Optimized Image Scaling Processor using VLSI

Exposing Image Forgery with Blind Noise Estimation

Cameras. Outline. Pinhole camera. Camera trial #1. Pinhole camera Film camera Digital camera Video camera High dynamic range imaging

High dynamic range imaging and tonemapping

HDR images acquisition

A JPEG CORNER ARTIFACT FROM DIRECTED ROUNDING OF DCT COEFFICIENTS. Shruti Agarwal and Hany Farid

Resampling and the Detection of LSB Matching in Colour Bitmaps

(12) Patent Application Publication (10) Pub. No.: US 2008/ A1. Kalevo (43) Pub. Date: Mar. 27, 2008

Automatic source camera identification using the intrinsic lens radial distortion

Part Number SuperPix TM image sensor is one of SuperPix TM 2 Mega Digital image sensor series products. These series sensors have the same maximum ima

Figures from Embedded System Design: A Unified Hardware/Software Introduction, Frank Vahid and Tony Givargis, New York, John Wiley, 2002

Camera Image Processing Pipeline: Part II

ISSN: ISO 9001:2008 Certified International Journal of Engineering and Innovative Technology (IJEIT) Volume 2, Issue 3, September 2012

Edge Potency Filter Based Color Filter Array Interruption

Source Camera Identification Using Enhanced Sensor Pattern Noise

Exposing Photo Manipulation with Geometric Inconsistencies

Colored Digital Image Watermarking using the Wavelet Technique

Research Article Discrete Wavelet Transform on Color Picture Interpolation of Digital Still Camera

Transcription:

Watermark Embedding in Digital Camera Firmware Peter Meerwald, May 28, 2008

Application Scenario Digital images can be easily copied and tampered Active and passive methods have been proposed for copyright protection and integrity verification: watermarking and forensics This work: simple watermarking in camera firmware for copyright protection

Camera Forensics [Chen08a] Use Photo-Response NonUniformity (PRNU) for camera identification and integrity verification Detecting forgeries (lighting, interlacing, specular highlights on the eye, JPEG quantization, double compression, chromatic aberration, image statistics, resampling, region, duplication, CFA Interpolation [Popescu05a], [Hani Farid])

Secure Digital Camera : Watermarking [Blythe04a] Capture human iris image through viewfinder, embed in camera image together with camera identification and image hash [Lukac06a] Emboss visible watermark in CFA domain [Mohanty07c] VLSI architecture for robust and fragile watermarking [Nelson05a] CMOS image sensor adds pseudorandom watermark Kodak and Epson offered cameras with watermarking capabilities (2003, discontinued) Patents! Many JPEG-domain algorithms that could be employed

Color Filter Array (CFA) Bayer CFA pattern

Demosaicking Basically interpolation to get full-resolution RGB image Many different approaches, eg. Bilinear, bicubic, smooth hue transition, gradient-based, adaptive color plane, adaptive homogeneitydirected,... Implementation, eg. http://www.cybercom.net/~dcoffin/dcraw/ Camera implementation unknown

Enter CHDK CHDK: firmware add-on for Canon DIGIC II and DIGIC III cameras, http://chdk.wikia.com Adds bracketing of exposure, RAW file support, BASIC scripts, remote camera control, additional data display (histogram, battery life), longer exposure time, faster shutter speed, games,... Linux-hosted cross-compilation, using arm-elf-gcc 3.4.6

CHDK enables Watermarking

Processing Pipeline

Camera Characteristics Canon IXUS 70, 7.1 MP DIGIC III chip (ARM9 core), unknown sensor Sensor resolution 3152x2340 pixels, usable 3112x2328 pixels Camera supports 3072x2304, 2592x1944, 2048x1536, 1600x1200, 640x480 pixel JPEG images (3 quality settings)

Camera Hardware / Software Canon DIGIC III processor aka. TI 32-bit ARM9 core, ~ 200 MHz Lots of custom hardware for JPEG, scaling, histogram computation, color conversion,... VxWorks operating system ~ 3.5 MB firmware ~ 1 MB usable free memory, ~45 MB / sec bandwidth ~ 150 KB CHDK add-on firmware

Watermark Embedding (1) Add pseudo-random sequence {-1, 1} generated by Mersenne Twister to blue channel pixels Want watermarking in 'real-time', before image is saved (so ~ 1 sec delay) 7.1 MP is 9.2 MB (10 bit/pixel, packed) Hard to match runtime requirement! Prefer to spend time on perceptual modelling rather than unpacking/packing bits!

Watermark Embedding (2) First naïve implementation ~ 45 sec: GetPixel()/PutPixel() ~ 15 sec: reduce calls to PRNG Unrolled pixel unpack/pack ~ 2 sec: shifting through two bit buffers Not yet Assembler code, SIMD instructions (?) Use some specialized hardware (?)

Watermark Embedding (3) // process even row p_row_buf = p_out_row_buf = (uint16 *) &buf_pos[pixtobytes(raw_left_margin+4)]; bit_buf = *p_row_buf++; out_bit_buf = bit_buf >> 6; bit_buf = (bit_buf << 16) + *p_row_buf++; pixel = bit_buf >> 12 & 0x3ff; WATERMARK_PIXEL(pixel); out_bit_buf = (out_bit_buf << 10) + pixel; *p_out_row_buf++ = out_bit_buf >> 4; out_bit_buf = (out_bit_buf << 10) + (bit_buf >> 2 & 0x3ff);... ARM9 can move & shift in one instruction! and implemented by shifts ldrh ldr ldrh mov ldr add tst mov mov rsbeq movs mov bne r7, [sl], #2 r2,.l102 r3, [sl], #2 r6, r7, asr #6 r5, [r2, #12] r7, r3, r7, asl #16 r9, r1 r3, r7, asr #12 r4, r3, asl #22 r5, r5, #0 r1, r1, lsr #1 r4, r4, lsr #22.L50

Loading Firmware Add-on

Watermarking Menu

Watermarking Options

Image Quality (RAW vs. JPEG)

Image Quality (RAW vs. Watermarked)

Detection Results SF... superfine, F... fine, N... normal (JPEG quality) Threshold ~ 0.03

Further Work Speedup Simple perceptual shaping Exploit interpolation for detection Image authentication Test different image mosaicking algorithms Compare with PRNU noise [Chen08a]

References [Blythe04a] P. Blythe, J. Fridrich, Secure Digital Camera, Digital Forensic Research Workshop, Aug. 11-13, 2004; Baltimore, MD, USA [Chen08a] M. Chen, J. Fridrich, M. Goljan, J. Lukas, Determining Image Origin and Integrity Using Sensor Noise, IEEE TIFS, 3(1):74-90, Mar. 2008 [Lukac06b] R. Lukac, K. Plataniotis, Camera Image Watermark Transfer by Demosaicking, Proc. ELMAR '06, 9-12, Jun. 2006 [Mohanty07c] S. Mohanty, E. Kougianos, N. Ranganathan, VLSI Architecture and Chip for Combined Invisible Robust and Fragile Watermarking, IET Computers & Digital Techniques, 1(5):600-611, Jun. 2007 [Nelson05a] G. Nelson, G. Jullien, O. Yadid-Pecht, CMOS Image Sensor with Watermarking Capabilities, ISCAS '05, 5326-5329, May. 2005 [Popescu05a] A. Popescu, H. Farid, Exposing Digital Forgeries in Color Filter Array Interpolated Images, IEEE TSP, 53(10):3948-3959, Oct. 2005