Porn Detection in a Video Streaming Using Hybrid Network of CNN and LSTM

Porn Detection in a Video Streaming Using Hybrid Network of CNN and LSTM

© 2021 by IJETT Journal
Volume-69 Issue-11
Year of Publication : 2021
Authors : Ilham Bintang, Gede Putra Kusuma
DOI :  10.14445/22315381/IJETT-V69I11P231

How to Cite?

Ilham Bintang, Gede Putra Kusuma, "Porn Detection in a Video Streaming Using Hybrid Network of CNN and LSTM," International Journal of Engineering Trends and Technology, vol. 69, no. 11, pp. 248-255, 2021. Crossref,

Porn detection in video streaming needs an efficient way to recognize because it consists of many picture frames that are stitched together to form a movement. Using real-time frame per frame detection is expensive. On the other hand, using fewer frames will lead to the loss of content. Choosing the right n frame to recognize is good, but it will calculate everything from scratch. A great trick to handle that is to use the information from the previous frame to calculate the feature of the next frame in the sequence. One of the most used approaches to process sequential data is long short-term memory (LSTM). In this research, CNN is combined to reduce the feature complexity and feature extraction and LSTM to store previous frame information to calculate the next frame. For the CNN layer, there are 3 types of models: ResNet50, VGG16, Simple CNN. The ResNet50 model can achieve the best accuracy of 98%. However, the best average inference time is achieved by Simple CNN at 90 ms for a 5-second video.

Hybrid Network, CNN model, LSTM model, Porn Recognition, Video Streaming

[1] M. Perez et al., Video pornography detection through deep learning techniques and motion information, Neurocomputing, 230 (2017) 279– 293, doi: 10.1016/j.neucom.2016.12.017.
[2] S. Sharma, B. Sudharsan, S. Naraharisetti, V. Trehan, and K. Jayavel, A fully integrated violence detection system using CNN and LSTM, International Journal of Electrical and Computer Engineering, 11(4) 3374–3380, Aug. 2021, doi: 10.11591/ijece.v11i4.pp3374-3380.
[3] G. Dines, Growing Up with Porn: The Developmental and Societal Impact of Pornography on Children, Dignity: A Journal on Sexual Exploitation and Violence, 2(3) (2017), doi: 10.23860/dignity.2017.02.03.03.
[4] Q. Lan, Z. Wang, M. Wen, C. Zhang, and Y. Wang, High Performance Implementation of 3D Convolutional Neural Networks on a GPU, Computational Intelligence and Neuroscience, 2017(2017), doi: 10.1155/2017/8348671.
[5] I. W. A. Arimbawa, I. G. P. S. Wijaya, and I. Bintang, Comparison of simple and stratified random sampling on porn videos recognition using CNN, 2019 International Conference on Computer Engineering, Network, and Intelligent Multimedia, CENIM 2019 - Proceeding, 2019 (2019), doi: 10.1109/CENIM48368.2019.8973305.
[6] L. Wang, J. Zhang, Q. Tian, C. Li, and L. Zhuo, Porn Streamer Recognition in Live Video Streaming via Attention-Gated Multimodal Deep Features, IEEE Transactions on Circuits and Systems for Video Technology, 30(12) (2020) 4876–4886, doi: 10.1109/TCSVT.2019.2958871.
[7] M. Zufar and B. Setiyono, Convolutional Neural Networks Untuk Pengenalan Wajah Secara Real-Time, Jurnal Sains dan Seni ITS, 5(2). 128862, 2016, doi: 10.12962/j23373520.v5i2.18854.
[8] I. G. P. S. Wijaya, I. B. K. Widiartha, K. Uchimura, M. S. Iqbal, and A. Y. Husodo, Fast pornographic image recognition using compact holistic features and multi-layer neural network, International Journal of Advances in Intelligent Informatics, 5(2)(2019) 89–100, doi: 10.26555/ijain.v5i2.268.
[9] J. A. M. Basilio, G. A. Torres, G. S. Pérez, L. K. T. Medina, and H. M. P. Meana, Explicit image detection using YCbCr space color model as skin detection, Applications of Mathematics and Computer Engineering - American Conference on Applied Mathematics, AMERICAN-MATH`11, 5th WSEAS International Conference on Computer Engineering and Applications, CEA`11, (2011) 123–128.
[10] I. G. P. S. Wijaya, I. B. K. Widiartha, and S. E. Anjarwani, Pornographic Image Recognition Based on Skin Probability and Eigenporn of Skin ROIs Images, TELKOMNIKA (Telecommunication Comput. Electron. Control), 13(3) 985, 2015. Ilham Bintang & Gede Putra Kusuma / IJETT, 69(11), 248-255, 2021 255
[11] N. Sae-Bae, X. Sun, H. T. Sencar, and N. D. Memon, Towards automatic detection of child pornography, 2014 IEEE International Conference on Image Processing, ICIP 2014, no. January, (2014) 5332–5336, doi: 10.1109/ICIP.2014.7026079.
[12] A. Ullah, J. Ahmad, K. Muhammad, M. Sajjad, and S. W. Baik, Action Recognition in Video Sequences using Deep Bi-Directional LSTM with CNN Features, IEEE Access, 6 (2018) 1155–1166, 2017, doi: 10.1109/ACCESS.2017.2778011.
[13] S. Sudhakaran and O. Lanz, Learning to detect violent videos using convolutional long short-term memory, IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS), no. doi: 10.1109/AVSS.2017.8078468, (2017) 1–6.
[14] L. Wang, J. Zhang, M. Wang, J. Tian, and L. Zhuo, Multilevel fusion of multimodal deep features for porn streamer recognition in live video, Pattern Recognition Letters, 140(2020) 150–157, doi: 10.1016/J.PATREC.2020.09.027.
[15] P. Kim, Convolutional Neural Network, in MATLAB Deep Learning, Berkeley, CA: Apress, 2017. doi: 10.1007/978-1-4842-2845-6_6.
[16] Boki Latupono, Implementasi Deep Learning menggunakan Convolution Neural Network untuk Klasifikasi Gambar,” UNIVERSITAS ISLAM INDONESIA, 2018.
[17] K. He, X. Zhang, S. Ren, and J. Sun, Deep residual learning for image recognition, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2016-Decem, (2016) 770– 778, doi: 10.1109/CVPR.2016.90.
[18] K. Simonyan and A. Zisserman, Very deep convolutional networks for large-scale image recognition, 3rd International Conference on Learning Representations, ICLR 2015 - Conference Track Proceedings, (2015) 1–14.
[19] S. K. Borse and D. v Patil, Air Quality Prediction Using Recurrent Neural Network, International Journal on Emerging Trends in Technology (IJETT), 7(1) (2020).