(Department of Computer Engineering and Artificial Intelligence, Military Technical College, Cairo, Egypt)(2) Eslam Magdy Mostafa
(Department of Computer Engineering and Artificial Intelligence, Military Technical College, Cairo, Egypt)(3) Mohamed Abdelmoneim Elshafey
(Department of Computer Engineering and Artificial Intelligence, Military Technical College, Cairo, Egypt)*corresponding author
AbstractThe image forgery process can be simply defined as inserting some objects of different sizes to vanish some structures or scenes. Satellite images can be forged in many ways, such as copy-paste, copy-move, and splicing processes. Recent approaches present a generative adversarial network (GAN) as an effective method for identifying the presence of spliced forgeries and identifying their locations with a higher detection accuracy of large- and medium-sized forgeries. However, such recent approaches clearly show limited detection accuracy of small-sized forgeries. Accordingly, the localization step of such small-sized forgeries is negatively impacted. In this paper, two different approaches for detecting and localizing small-sized forgeries in satellite images are proposed. The first approach is inspired by a recently presented GAN-based approach and is modified to an enhanced version. The experimental results manifest that the detection accuracy of the first proposed approach noticeably increased to 86% compared to its inspiring one with 79% for the small-sized forgeries. Whereas, the second proposed approach uses a different design of a CNN-based discriminator to significantly enhance the detection accuracy to 94%, using the same dataset obtained from NASA and the US Geological Survey (USGS) for validation and testing. Furthermore, the results show a comparable detection accuracy in large- and medium-sized forgeries using the two proposed approaches compared to the competing ones. This study can be applied in the forensic field, with clear discrimination between the forged and pristine images.
KeywordsCNN; GAN; Satellite Image Forgery; Transfer Learning; FReLU Activation Function
|
DOIhttps://doi.org/10.26555/ijain.v6i3.548 |
Article metricsAbstract views : 1422 | PDF views : 218 |
Cite |
Full Text Download
|
References
[1] A. Roy, R. Dixit, R. Naskar, and R. S. Chakraborty, Digital Image Forensics: Theory and Implementation, 2020, vol. 755, doi: 10.1007/978-981-10-7644-2.
[2] L. Verdoliva, “Media Forensics and DeepFakes: An Overview,” IEEE J. Sel. Top. Signal Process., vol. 14, no. 5, pp. 910–932, Aug. 2020, doi: 10.1109/JSTSP.2020.3002101.
[3] L. Ghammam, K. Karabina, P. Lacharme, and K. Thiry-Atighehchi, “A Cryptanalysis of Two Cancelable Biometric Schemes Based on Index-of-Max Hashing,” IEEE Trans. Inf. Forensics Secur., vol. 15, pp. 2869–2880, 2020, doi: 10.1109/TIFS.2020.2977533.
[4] K. H. Rhee, “Detection of Spliced Image Forensics Using Texture Analysis of Median Filter Residual,” IEEE Access, vol. 8, pp. 103374–103384, 2020, doi: 10.1109/ACCESS.2020.2999308.
[5] L. Bondi, S. Lameri, D. Guera, P. Bestagini, E. J. Delp, and S. Tubaro, “Tampering Detection and Localization Through Clustering of Camera-Based CNN Features,” in 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2017, pp. 1855–1864, doi: 10.1109/CVPRW.2017.232.
[6] R. Agarwal, D. Khudaniya, A. Gupta, and K. Grover, “Image Forgery Detection and Deep Learning Techniques: A Review,” in 2020 4th International Conference on Intelligent Computing and Control Systems (ICICCS), 2020, pp. 1096–1100, doi: 10.1109/ICICCS48265.2020.9121083.
[7] K. Saeed, S. Datta, and N. Chaki, “A Granular Level Feature Extraction Approach to Construct HR Image for Forensic Biometrics Using Small Training DataSet,” IEEE Access, vol. 8, pp. 123556–123570, 2020, doi: 10.1109/ACCESS.2020.3006100.
[8] C. Chen, X. Zhao, and M. C. Stamm, “Generative Adversarial Attacks Against Deep-Learning-Based Camera Model Identification,” IEEE Trans. Inf. Forensics Secur., pp. 1–1, 2019, doi: 10.1109/TIFS.2019.2945198.
[9] O. Kliuiev, M. Mozhaiev, O. Uhrovetskyi, O. Mozhaiev, and E. Simakova-Yefremian, “Method of Forensic Research on Image for Finding Touch up on the Basis of Noise Entropy,” in 2019 3rd International Conference on Advanced Information and Communications Technologies (AICT), 2019, pp. 76–79, doi: 10.1109/AIACT.2019.8847760.
[10] S. K. Yarlagadda, D. Güera, P. Bestagini, F. Maggie Zhu, S. Tubaro, and E. J. Delp, “Satellite Image Forgery Detection and Localization Using GAN and One-Class Classifier,” Electron. Imaging, vol. 2018, no. 7, pp. 214-1-214–9, Jan. 2018, doi: 10.2352/ISSN.2470-1173.2018.07.MWSF-214.
[11] E. R. Bartusiak et al., “Splicing Detection and Localization In Satellite Imagery Using Conditional GANs,” in 2019 IEEE Conference on Multimedia Information Processing and Retrieval (MIPR), 2019, pp. 91–96, doi: 10.1109/MIPR.2019.00024.
[12] X. Wei, X. Wei, W. Xing, S. Lu, and W. Lu, “An Incremental Self-Labeling Strategy for Semi-Supervised Deep Learning Based on Generative Adversarial Networks,” IEEE Access, vol. 8, pp. 8913–8921, 2020, doi: 10.1109/ACCESS.2020.2964315.
[13] Landsat Science, “National Aeronautics and Space Administration,” 2020. [Online]. Available: https://landsat.gsfc.nasa.gov/. [Accessed: 01-Jul-2020].
[14] Science for a changing world, “U.S. Geological Survey, USGS.gov,” 2020. [Online]. Available: https://www.usgs.gov/. [Accessed: 01-Jul-2020].
[15] A. Kohli, A. Gupta, and D. Singhal, “CNN based localisation of forged region in object-based forgery for HD videos,” IET Image Process., vol. 14, no. 5, pp. 947–958, Apr. 2020, doi: 10.1049/iet-ipr.2019.0397.
[16] Xiaofeng Wang, Kemu Pang, Xiaorui Zhou, Yang Zhou, Lu Li, and Jianru Xue, “A Visual Model-Based Perceptual Image Hash for Content Authentication,” IEEE Trans. Inf. Forensics Secur., vol. 10, no. 7, pp. 1336–1349, Jul. 2015, doi: 10.1109/TIFS.2015.2407698.
[17] A.-J. Gallego, J. Calvo-Zaragoza, and J. R. Rico-Juan, “Insights Into Efficient k-Nearest Neighbor Classification With Convolutional Neural Codes,” IEEE Access, vol. 8, pp. 99312–99326, 2020, doi: 10.1109/ACCESS.2020.2997387.
[18] J. Pomerat, A. Segev, and R. Datta, “On Neural Network Activation Functions and Optimizers in Relation to Polynomial Regression,” in 2019 IEEE International Conference on Big Data (Big Data), 2019, pp. 6183–6185, doi: 10.1109/BigData47090.2019.9005674.
[19] X. Zhang, Y. Zou, and W. Shi, “Dilated convolution neural network with LeakyReLU for environmental sound classification,” in 2017 22nd International Conference on Digital Signal Processing (DSP), 2017, pp. 1–5, doi: 10.1109/ICDSP.2017.8096153.
[20] M. M. Kalayeh and M. Shah, “Training Faster by Separating Modes of Variation in Batch-Normalized Models,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 42, no. 6, pp. 1483–1500, Jun. 2020, doi: 10.1109/TPAMI.2019.2895781.
[21] R. Ashraf et al., “Region-of-Interest Based Transfer Learning Assisted Framework for Skin Cancer Detection,” IEEE Access, vol. 8, pp. 147858–147871, 2020, doi: 10.1109/ACCESS.2020.3014701.
[22] H. Fan, S. Gao, X. Zhang, X. Cao, H. Ma, and Q. Liu, “Intelligent Recognition of Ferrographic Images Combining Optimal CNN With Transfer Learning Introducing Virtual Images,” IEEE Access, vol. 8, pp. 137074–137093, 2020, doi: 10.1109/ACCESS.2020.3011728.
[23] K. Bu, Y. He, X. Jing, and J. Han, “Adversarial Transfer Learning for Deep Learning Based Automatic Modulation Classification,” IEEE Signal Process. Lett., vol. 27, pp. 880–884, 2020, doi: 10.1109/LSP.2020.2991875.
[24] B. Cui, X. Chen, and Y. Lu, “Semantic Segmentation of Remote Sensing Images Using Transfer Learning and Deep Convolutional Neural Network With Dense Connection,” IEEE Access, vol. 8, pp. 116744–116755, 2020, doi: 10.1109/ACCESS.2020.3003914.
[25] S. Obla, X. Gong, A. Aloufi, P. Hu, and D. Takabi, “Effective Activation Functions for Homomorphic Evaluation of Deep Neural Networks,” IEEE Access, vol. 8, pp. 153098–153112, 2020, doi: 10.1109/ACCESS.2020.3017436.
[26] W. Gong, H. Chen, Z. Zhang, M. Zhang, and H. Gao, “A Data-Driven-Based Fault Diagnosis Approach for Electrical Power DC-DC Inverter by Using Modified Convolutional Neural Network With Global Average Pooling and 2-D Feature Image,” IEEE Access, vol. 8, pp. 73677–73697, 2020, doi: 10.1109/ACCESS.2020.2988323.
[27] S. Qiu, X. Xu, and B. Cai, “FReLU: Flexible Rectified Linear Units for Improving Convolutional Neural Networks,” in 2018 24th International Conference on Pattern Recognition (ICPR), 2018, pp. 1223–1228, doi: 10.1109/ICPR.2018.8546022.
[28] Z. Ma et al., “Fine-Grained Vehicle Classification With Channel Max Pooling Modified CNNs,” IEEE Trans. Veh. Technol., vol. 68, no. 4, pp. 3224–3233, Apr. 2019, doi: 10.1109/TVT.2019.2899972.
[29] X. Shen, X. Zhang, L. Lan, Q. Liao, and Z. Luo, “Another Robust NMF: Rethinking the Hyperbolic Tangent Function and Locality Constraint,” IEEE Access, vol. 7, pp. 31089–31102, 2019, doi: 10.1109/ACCESS.2019.2903309.
[30] J. Yang, D. Ruan, J. Huang, X. Kang, and Y.-Q. Shi, “An Embedding Cost Learning Framework Using GAN,” IEEE Trans. Inf. Forensics Secur., vol. 15, pp. 839–851, 2020, doi: 10.1109/TIFS.2019.2922229.
[31] N. Akhtar and U. Ragavendran, “Interpretation of intelligence in CNN-pooling processes: a methodological survey,” Neural Comput. Appl., vol. 32, no. 3, pp. 879–898, Feb. 2020, doi: 10.1007/s00521-019-04296-5.
[32] C.-Y. Lee, P. Gallagher, and Z. Tu, “Generalizing Pooling Functions in CNNs: Mixed, Gated, and Tree,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 40, no. 4, pp. 863–875, Apr. 2018, doi: 10.1109/TPAMI.2017.2703082.
[33] R. Aldana, L. Campos-Macias, J. Zamora, D. Gomez-Gutierrez, and A. Cruz, “Dynamic Learning Rate for Neural Networks: A Fixed-Time Stability Approach,” in 2018 24th International Conference on Pattern Recognition (ICPR), 2018, pp. 1378–1383, doi: 10.1109/ICPR.2018.8546084.
[34] E. Hoffer, I. Hubara, and D. Soudry, “Train longer, generalize better: closing the generalization gap in large batch training of neural networks,” in Advances in Neural Information Processing Systems, 2017, pp. 1731–1741, available at: Google Scholar.
[35] H. Narasimhan and S. Agarwal, “Support Vector Algorithms for Optimizing the Partial Area under the ROC Curve,” Neural Comput., vol. 29, no. 7, pp. 1919–1963, Jul. 2017, doi: 10.1162/NECO_a_00972.
[36] Y. Jiang, “Receiver Operating Characteristic (ROC) Analysis of Image Search-and-Localize Tasks,” Acad. Radiol., vol. 27, no. 12, pp. 1742–1750, Dec. 2020, doi: 10.1016/j.acra.2019.12.020.
[37] W. Ma and M. A. Lejeune, “A distributionally robust area under curve maximization model,” Oper. Res. Lett., vol. 48, no. 4, pp. 460–466, Jul. 2020, doi: 10.1016/j.orl.2020.05.012.
[38] G. Zhang, Z. Wang, and H. Mei, “Sensitivity clustering and ROC curve based alarm threshold optimization,” Process Saf. Environ. Prot., vol. 141, pp. 83–94, Sep. 2020, doi: 10.1016/j.psep.2020.03.029.
[39] A. Jokiel-Rokita and R. Topolnicki, “Estimation of the ROC curve from the Lehmann family,” Comput. Stat. Data Anal., vol. 142, p. 106820, Feb. 2020, doi: 10.1016/j.csda.2019.106820.
[40] L. Omar and I. Ivrissimtzis, “Using theoretical ROC curves for analysing machine learning binary classifiers,” Pattern Recognit. Lett., vol. 128, pp. 447–451, Dec. 2019, doi: 10.1016/j.patrec.2019.10.004.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
___________________________________________________________
International Journal of Advances in Intelligent Informatics
ISSN 2442-6571 (print) | 2548-3161 (online)
Organized by UAD and ASCEE Computer Society
Published by Universitas Ahmad Dahlan
W: http://ijain.org
E: info@ijain.org (paper handling issues)
andri.pranolo.id@ieee.org (publication issues)
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0