Forgery image detection via mask filter banks based CNN

Luyue Wang, Seiichiro Kamata

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we present a new image forgery detection method via a mask filter banks which is consisted with the designed mask filters to extract the features of different channels of image and a modified a ResNet to classify the input image is tempered or not. The proposed model is proved to be capable for copy-move forgery and splicing image detection. In the mask filter layer, we first convert the image from spatial domain to frequency domain, then extract the image edge information of each channel by element-wise with the designed mask matrix. Finally, edge and noise information features of different channels were fused as feature vectors fed to a trained ResNet to do classification. Experiments on three standard datasets: the copy-move forgery image datasets MICC-F220 and MICC-F2000, splicing image manipulation datasets Columbia demonstrate that proposed method get better results than the original colour image as input method and also outperform some existing works.

Original languageEnglish
Title of host publicationTenth International Conference on Graphics and Image Processing, ICGIP 2018
EditorsYifei Pu, Hui Yu, Chunming Li, Zhigeng Pan
PublisherSPIE
ISBN (Electronic)9781510628281
DOIs
Publication statusPublished - 2019 Jan 1
Event10th International Conference on Graphics and Image Processing, ICGIP 2018 - Chengdu, China
Duration: 2018 Dec 122018 Dec 14

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume11069
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

Conference10th International Conference on Graphics and Image Processing, ICGIP 2018
CountryChina
CityChengdu
Period18/12/1218/12/14

Fingerprint

Filter Banks
Filter banks
Mask
Masks
masks
filters
splicing
Filter
Color
Columbia (Orbiter)
Color Image
Feature Vector
Convert
Frequency Domain
Manipulation
manipulators
Experiments
Classify
color

Keywords

  • Image Forgery Detection
  • Mask Banks Filter
  • ResNet

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Computer Science Applications
  • Applied Mathematics
  • Electrical and Electronic Engineering

Cite this

Wang, L., & Kamata, S. (2019). Forgery image detection via mask filter banks based CNN. In Y. Pu, H. Yu, C. Li, & Z. Pan (Eds.), Tenth International Conference on Graphics and Image Processing, ICGIP 2018 [110691P] (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 11069). SPIE. https://doi.org/10.1117/12.2524351

Forgery image detection via mask filter banks based CNN. / Wang, Luyue; Kamata, Seiichiro.

Tenth International Conference on Graphics and Image Processing, ICGIP 2018. ed. / Yifei Pu; Hui Yu; Chunming Li; Zhigeng Pan. SPIE, 2019. 110691P (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 11069).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Wang, L & Kamata, S 2019, Forgery image detection via mask filter banks based CNN. in Y Pu, H Yu, C Li & Z Pan (eds), Tenth International Conference on Graphics and Image Processing, ICGIP 2018., 110691P, Proceedings of SPIE - The International Society for Optical Engineering, vol. 11069, SPIE, 10th International Conference on Graphics and Image Processing, ICGIP 2018, Chengdu, China, 18/12/12. https://doi.org/10.1117/12.2524351
Wang L, Kamata S. Forgery image detection via mask filter banks based CNN. In Pu Y, Yu H, Li C, Pan Z, editors, Tenth International Conference on Graphics and Image Processing, ICGIP 2018. SPIE. 2019. 110691P. (Proceedings of SPIE - The International Society for Optical Engineering). https://doi.org/10.1117/12.2524351
Wang, Luyue ; Kamata, Seiichiro. / Forgery image detection via mask filter banks based CNN. Tenth International Conference on Graphics and Image Processing, ICGIP 2018. editor / Yifei Pu ; Hui Yu ; Chunming Li ; Zhigeng Pan. SPIE, 2019. (Proceedings of SPIE - The International Society for Optical Engineering).
@inproceedings{d6c038d3ed544c648283e3c929811ba3,
title = "Forgery image detection via mask filter banks based CNN",
abstract = "In this paper, we present a new image forgery detection method via a mask filter banks which is consisted with the designed mask filters to extract the features of different channels of image and a modified a ResNet to classify the input image is tempered or not. The proposed model is proved to be capable for copy-move forgery and splicing image detection. In the mask filter layer, we first convert the image from spatial domain to frequency domain, then extract the image edge information of each channel by element-wise with the designed mask matrix. Finally, edge and noise information features of different channels were fused as feature vectors fed to a trained ResNet to do classification. Experiments on three standard datasets: the copy-move forgery image datasets MICC-F220 and MICC-F2000, splicing image manipulation datasets Columbia demonstrate that proposed method get better results than the original colour image as input method and also outperform some existing works.",
keywords = "Image Forgery Detection, Mask Banks Filter, ResNet",
author = "Luyue Wang and Seiichiro Kamata",
year = "2019",
month = "1",
day = "1",
doi = "10.1117/12.2524351",
language = "English",
series = "Proceedings of SPIE - The International Society for Optical Engineering",
publisher = "SPIE",
editor = "Yifei Pu and Hui Yu and Chunming Li and Zhigeng Pan",
booktitle = "Tenth International Conference on Graphics and Image Processing, ICGIP 2018",

}

TY - GEN

T1 - Forgery image detection via mask filter banks based CNN

AU - Wang, Luyue

AU - Kamata, Seiichiro

PY - 2019/1/1

Y1 - 2019/1/1

N2 - In this paper, we present a new image forgery detection method via a mask filter banks which is consisted with the designed mask filters to extract the features of different channels of image and a modified a ResNet to classify the input image is tempered or not. The proposed model is proved to be capable for copy-move forgery and splicing image detection. In the mask filter layer, we first convert the image from spatial domain to frequency domain, then extract the image edge information of each channel by element-wise with the designed mask matrix. Finally, edge and noise information features of different channels were fused as feature vectors fed to a trained ResNet to do classification. Experiments on three standard datasets: the copy-move forgery image datasets MICC-F220 and MICC-F2000, splicing image manipulation datasets Columbia demonstrate that proposed method get better results than the original colour image as input method and also outperform some existing works.

AB - In this paper, we present a new image forgery detection method via a mask filter banks which is consisted with the designed mask filters to extract the features of different channels of image and a modified a ResNet to classify the input image is tempered or not. The proposed model is proved to be capable for copy-move forgery and splicing image detection. In the mask filter layer, we first convert the image from spatial domain to frequency domain, then extract the image edge information of each channel by element-wise with the designed mask matrix. Finally, edge and noise information features of different channels were fused as feature vectors fed to a trained ResNet to do classification. Experiments on three standard datasets: the copy-move forgery image datasets MICC-F220 and MICC-F2000, splicing image manipulation datasets Columbia demonstrate that proposed method get better results than the original colour image as input method and also outperform some existing works.

KW - Image Forgery Detection

KW - Mask Banks Filter

KW - ResNet

UR - http://www.scopus.com/inward/record.url?scp=85065804114&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85065804114&partnerID=8YFLogxK

U2 - 10.1117/12.2524351

DO - 10.1117/12.2524351

M3 - Conference contribution

T3 - Proceedings of SPIE - The International Society for Optical Engineering

BT - Tenth International Conference on Graphics and Image Processing, ICGIP 2018

A2 - Pu, Yifei

A2 - Yu, Hui

A2 - Li, Chunming

A2 - Pan, Zhigeng

PB - SPIE

ER -