Emily_Bartusiak_Master_s_Thesis_Final_Version.pdf (1.55 MB)
An Adversarial Approach to Spliced Forgery Detection and Localization in Satellite Imagery
The widespread availability of image editing tools and improvements in image processing techniques make image manipulation feasible for the general population. Oftentimes, easy-to-use yet sophisticated image editing tools produce results that contain modifications imperceptible to the human observer. Distribution of forged images can have drastic ramifications, especially when coupled with the speed and vastness of the Internet. Therefore, verifying image integrity poses an immense and important challenge to the digital forensic community. Satellite images specifically can be modified in a number of ways, such as inserting objects into an image to hide existing scenes and structures. In this thesis, we describe the use of a Conditional Generative Adversarial Network (cGAN) to identify the presence of such spliced forgeries within satellite images. Additionally, we identify their locations and shapes. Trained on pristine and falsified images, our method achieves high success on these detection and localization objectives.
Funding
Defense Advanced Research Projects Agency (DARPA) and Air Force Research Laboratory (AFRL) through agreement number FA8750-16-2-0173
History
Degree Type
- Master of Science in Electrical and Computer Engineering
Department
- Electrical and Computer Engineering
Campus location
- West Lafayette
Advisor/Supervisor/Committee Chair
Dr. Edward J. DelpAdditional Committee Member 2
Dr. Amy R. ReibmanAdditional Committee Member 3
Dr. Michael D. ZoltowskiUsage metrics
Categories
Keywords
Image ProcessingMachine Learning, Electrical Engineering, Computer Engineering, ECE, Satellite Imagery, Forgery Detection, Forgery Localization, Forged Images, Falsified Imagery, Image Editing, Image Manipulation, Doctored Images, Splicing, Spliced Objects, Spliced Forgery, Image Forensic Analysis, Digital Forensics, Media Forensics, Conditional Generative Adversarial Network, cGANImage ProcessingMachine LearningElectrical EngineeringComputer EngineeringElectrical and Computer EngineeringECESatellite ImagerycGANConditional Generative Adversarial NetworkMedia ForensicsDigital ForensicsImage Forensic AnalysisSpliced ForgerySpliced ObjectsSplicingDoctored ImagesImage ManipulationImage EditingFalsified ImageryForged ImagesForgery LocalizationForgery DetectionImage ProcessingArtificial Intelligence and Image ProcessingKnowledge Representation and Machine Learning
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC