Abstract
Digital image classification assists in distinguishing natural and synthetic images to detect computer-generated objects. However, CGI improvements make it difficult to discern synthetic photos from genuine ones. Researchers suggest multiple deep learning strategies to differentiate these photo sets utilizing thorough feature analysis. These models are either complex or do not handle image sub-components, decreasing efficiency in large-scale applications. These models fail categorically. To address these issues, this work proposes a novel high-density bio-inspired feature analysis deep learning model for natural and synthetic image sub-classification. A YoLo model initially recognizes objects in input image sets. Processed separately, a hybrid LSTM/GRU model predicts high-density feature sets, which are processed by Elephant Herding Optimization (EHO) Models to identify high inter-class variance feature sets. A customized 1D CNN model is used to categorize the desired features into natural and synthetic components. These classification results establish whether the input image is natural, synthetic, or both. In real-time scenarios, the proposed model is able to improve standard classification models with 8.7% greater accuracy, 10.9% higher precision, 3.2% higher recall, and 8.4% higher AUC.
This is a preview of subscription content, log in via an institution to check access.
References
Yake Z, Fang L, Licheng J, Shuyuan Y, Lingling L, Meijuan Y (2020) Discriminative sketch topic model with structural constraint for SAR image classification. IEEE J Select Top Appl Earth Observ Remote Sens 13:5730–5745. https://doi.org/10.1109/JSTARS.2020.3024002
Hongliang Z, Rocky L, Minyi H (2020) Shadow compensation for synthetic aperture radar target classification by dual parallel generative adversarial network. IEEE Sens Lett 4(8):1–4. https://doi.org/10.1109/LSENS.2020.3009179
Hui B, Jiarui D, Tianwen Y, Jian W, Ling W (2021) CNN-based target detection and classification when sparse SAR image dataset is available. IEEE J Select Top Appl Earth Observ Remote Sens 14:6815–6826. https://doi.org/10.1109/JSTARS.2021.3093645
Hongliang Z, Weiye W, Rocky L (2020) SAR target classification based on radar image luminance analysis by deep learning. IEEE Sens Lett 4(3):1–4. https://doi.org/10.1109/LSENS.2020.2976836
Qian W, Biao H, Zaidao W, Zhongle R, Licheng J (2021) Cost-sensitive latent space learning for imbalanced PolSAR image classification. IEEE Trans Geosci Remote Sens 59(6):4802–4817. https://doi.org/10.1109/TGRS.2020.3012276
Nogueira FEA, Marques RCP, Medeiros FNS (2020) SAR image segmentation based on unsupervised classification of log-cumulants estimates. IEEE Geosci Remote Sens Lett 17(7):1287–1289. https://doi.org/10.1109/LGRS.2019.2941075
Xingyu L, Yan W, Wenkai L, Yice C, Ming L (2022) High resolution SAR image classification using global-local network structure based on vision transformer and CNN. IEEE Geosci Remote Sens Lett 19:1–5. https://doi.org/10.1109/LGRS.2022.3151353
Jing G, Licheng J, Fang L, Xiangrong Z, Xu T, Puhua C (2020) Multi-feature weighted sparse graph for SAR image analysis. IEEE Trans Geosci Remote Sens 58(2):881–891. https://doi.org/10.1109/TGRS.2019.2941288
Dou Q, Huiyuan W, Shuang W, Ruiqi L, Baorui D, Yi L, Biao H, Licheng J (2022) Self-distillation feature learning network for optical and SAR image registration. IEEE Trans Geosci Remote Sens 60:1–18. https://doi.org/10.1109/TGRS.2022.3173476
Pengjiang Q, Yangyang C, Jung-Wen K, Yu-Dong Z, Yizhang J, Kaifa Z, Rose AH, Harry F, Atallah B, Feifei Z, Jin UH, Norbert A, Karin H, Rodney E, Bryan T, Robert SJ, Shitong W, Kuan-Hao S, Raymond FM (2020) mDixon-based synthetic CT generation for PET attenuation correction on abdomen and pelvis jointly using transfer fuzzy clustering and active learning-based classification. IEEE Trans Med Imag 39(4):819–832. https://doi.org/10.1109/TMI.2019.2935916
Nupoor Y, Neeraj S, Nikkoo K (2023) Design of a hybrid GWO CNN model for identification of synthetic images via transfer learning process. Int J Intell Eng Syst 16(3):292–301
Zitong W, Biao H, Licheng J (2021) Multiscale CNN with autoencoder regularization joint contextual attention network for SAR image classification. IEEE Trans Geosci Remote Sens 59(2):1200–1213. https://doi.org/10.1109/TGRS.2020.3004911
Rui C, Yong W, Bin Z, Xiaofei Lu (2021) Ship target imaging in airborne SAR system based on automatic image segmentation and ISAR technique. IEEE J Select Top Appl Earth Observ Remote Sens 14:1985–2000. https://doi.org/10.1109/JSTARS.2021.3050108
Acknowledgements
The authors acknowledge and appreciate the effort and time of all the faculty members who have taken part in this questionnaire.
Funding
Currently, there is no funding available for this research.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Yawale, N., Sahu, N. & Khalsa, N. A Multimodal Feature Representation Model for Transfer-Learning-Based Identification of Images. Natl. Acad. Sci. Lett. (2024). https://doi.org/10.1007/s40009-024-01402-7
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s40009-024-01402-7