Coverless image steganography using morphed face recognition based on convolutional neural network

Abstract

In recent years, information security has become a prime issue of worldwide concern. To improve the validity and proficiency of the image data hiding approach, a piece of state-of-the-art secret information hiding transmission scheme based on morphed face recognition is proposed. In our proposed data hiding approach, a group of morphed face images is produced from an arranged small-scale face image dataset. Then, a morphed face image which is encoded with a secret message is sent to the receiver. The receiver uses powerful and robust deep learning models to recover the secret message by recognizing the parents of the morphed face images. Furthermore, we design two novel Convolutional Neural Network (CNN) architectures (e.g. MFR-Net V1 and MFR-Net V2) to perform morphed face recognition and achieved the highest accuracy compared with existing networks. Additionally, the experimental results show that the proposed schema has higher retrieval capacity and accuracy and it provides better robustness.

Image steganalysis algorithm based on deep learning and attention mechanism for computer communication

“…It used steganographic features to encode deformed face information and performed deep learning models for recognition, resulting in higher retrieval capability and accuracy. 8 Sarkar et al used continuous wavelet transform to encode the time series of sensor data into multichannel images and then used spatial attention-assisted CNN to extract high-dimensional features. It also greatly improved the accuracy of image steganographic feature extraction by using genetic algorithm to select the optimal feature set.…”

Convolutional Neural Networks for Face Recognition: A Systematic Literature Review

“…The results are shown in Table 11. al, 2022;Taherkhani et al, 2019;Guo et al, 2017;Qin et al, 2019;Filippidou and Papakostas, 2020;Perti et al, 2020;Sepas-Moghaddam et al, 2020;Zhiqi, 2021;Tran et al, 2022;Luttrell et al, 2017;Nam et al, 2018;Chandran et al, 2018;Choi and Lee, 2020;Zangeneh et al, 2020;Kim et al, 2020) ResNet (He et al, 2016) (Ling et al, 2020;Li et al, 2022;Horng et al, 2022;Li et al, 2022;Kim et al, 2020;Feng et al, 2020;Arafah et al, 2020;Almabdy and Elrefaei, 2019;Setio Aji et al, 2022;Gruber et al, 2017;Hou et al, 2020;Filippidou and Papakostas, 2020;Alkanhal et al, 2023;Zhou et al, 2020) AlexNet (Krizhevsky et al, 2012) (Bukovčiková et al, 2017;Bussey et al, 2017;Liu et al, 2017;Khan et al, 2019b;Han, 2021;H et al, 2023;Almabdy and Elrefaei, 2019;Ma et al, 2018;Alhan...…”

“…It is important to notice that the second most largest category goes to private datasets with 34 papers (Zhou et al, 2018;Ara et al, 2017;Phankokkruad, 2018;Gilani and Mian, 2018;Khan et al, 2019b;Qin et al, 2019;Liu et al, 2019;Peng et al, 2019;Mangal et al, 2020;Lv et al, 2020;Perti et al, 2020;Kim et al, 2017;Irjanto and Surantha, 2020;Arafah et al, 2020;Prasetyo et al, 2021;Moon et al, 2017;Chandran et al, 2018;Yang et al, 2018;Son et al, 2020;Alhanaee et al, 2021;Khan et al, 2020;Nakajima et al, 2021;Talahua et al, 2021;He and Ding, 2023;Karlupia et al, 2023;Bussey et al, 2017;Li et al, 2022;Filippidou and Papakostas, 2020;Bussey et al, 2017;Singh et al, 2022;Setio Aji et al, 2022;Wang et al, 2022;Lestari et al, 2021) creating their own datasets for testing. There are advantages and disadvantages to developing and utilizing private face image datasets for CNN-based face recognition.…”

“…Overall, the high accuracy observed in the study can be attributed to the utilization of pre-trained CNN models, the effectiveness of transfer learning, and the careful selection of optimal features through experimentation and evaluation on diverse datasets. In addition, one more thing we found was that 3 papers (Bussey et al, 2017;Almabdy and Elrefaei, 2019;Li et al, 2022) constructed their own small-scale datasets. That trend is followed in (Li et al, 2022) where the authors construct their own private dataset, however, they don't use pretrained models, but instead design two CNN architectures called MFR-Net V1 and MFR-Net V2.…”

“…In addition, one more thing we found was that 3 papers (Bussey et al, 2017;Almabdy and Elrefaei, 2019;Li et al, 2022) constructed their own small-scale datasets. That trend is followed in (Li et al, 2022) where the authors construct their own private dataset, however, they don't use pretrained models, but instead design two CNN architectures called MFR-Net V1 and MFR-Net V2. Both architectures were able to achieve a 100% accuracy with MFR-Net V1 achieving 99.99%-100% and MFR-Net V2 100%.…”