An Enhanced Multimodal Stacking Scheme for Online Pornographic Content Detection
Source abstract
Abstract
An enhanced multimodal stacking scheme is proposed for quick and accurate online detection of harmful pornographic contents on the Internet. To accurately detect harmful contents, the implicative visual features (auditory features) are extracted using a bi-directional RNN (recurrent neural network) with VGG-16 (a multilayered dilated convolutional network) to implicitly express the signal change patterns over time within each input. Using only the implicative visual and auditory features, a video classifier and an audio classifier are trained, respectively. By using both features together, one fusion classifier is also trained. Then, these three component classifiers are stacked in the enhanced ensemble scheme to reduce the false negative errors in a serial order of the fusion classifier, video classifier, and audio classifier for a quick online detection. The proposed multimodal stacking scheme yields an improved true positive rate of 95.40% and a false negative rate of 4.60%, which are superior values to previous studies. In addition, the proposed stacking scheme can accurately detect harmful contents up to 74.58% and an average rate of 62.16% faster than the previous stacking scheme. Therefore, the proposed enhanced multimodal stacking scheme can be used to quickly and accurately filter out harmful contents in the online environments.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Video Analysis and Summarization · Digital Media Forensic Detection · Advanced Steganography and Watermarking Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Information and Communication Engineering, Inha University, Incheon 22212, Korea
Funding
Inha University
References (30)
Show source-provided references
1. KOREA Communications Commission (2019, December 20). They Try to Block the Distribution Source of Pornographic Content, Available online: http://it.chosun.com/news/article.html?no=2843689.
2. Agarwal, S., Farid, H., Gu, Y., He, M., Nagano, K., and Li, H. (2019, January 15–21). Protecting world leaders against deep fakes. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, Long Beach, CA, USA.
3. Moustaf, N.M. (2015). Applying deep learning to classify pornographic images and videos. arXiv.
4. Pornography Classification: The Hidden Clues in Video Space-Time10.1016/j.forsciint.2016.09.010 ↗
5. Song, K., and Kim, Y. (2017, January 27–28). Pornographic Video Detection Scheme using Video Descriptor based on Deep Learning Architecture. Proceedings of the 4th International Conference on Emerging Trends in Academic Research, Bali, Indonesia.
6. Varges, M., and Marana, A.N. (2019). Spatiotemporal CNNs for Pornography Detection in Videos. Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications, Springer.
7. Pornographic Image Recognition via Weighted Multiple Instance Learning10.1109/TCYB.2018.2864870 ↗
8. Xiao, X., Xu, Y., Zhang, C., Li, X., Zhang, B., and Bian, Z. (2019, January 11–13). A new method for pornographic video detection with the integration of visual and textual information. Proceedings of the IEEE 3rd Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC), Chongqing, China.10.1109/IMCEC46724.2019.8984107 ↗
9. Farooq, M.S., Khan, M.A., Abbas, S., Athar, A., Ali, N., and Hassan, A. (2019, January 16–17). Skin Detection based Pornography Filtering using Adaptive Back Propagation Neural Network. Proceedings of the 8th International Conference on Information and Communication Technologies (ICICT), Karachi, Pakistan.10.1109/ICICT47744.2019.9001915 ↗
10. Kusrini, H.A., Fatta, S.P., and Widiyanto, W.W. (2019, January 24–25). Prototype of Pornographic Image Detection with YCbCr and Color Space (RGB) Methods of Computer Vision. Proceedings of the 2019 International Conference on Information and Communications Technology (ICOIACT), Yogyakarta, Indonesia.10.1109/ICOIACT46704.2019.8938524 ↗
11. Ashan, B., Cho, H., and Liu, Q. (2019, January 20–22). Performance Evaluation of Transfer Learning for Pornographic Detection. Proceedings of the International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD 2019), Kunming, China.10.1007/978-3-030-32591-6_43 ↗
12. Gangwar, A., Fidalgo, E., Alegre, E., and González-Castro, V. (2017, January 13–15). Pornography and child sexual abuse detection in image and video: A comparative evaluation. Proceedings of the 8th International Conference on Imaging for Crime Detection and Prevention (ICDP 2017), Madrid, Spain.10.1049/ic.2017.0046 ↗
13. He, Y., Shi, J., Wang, C., Huang, H., Liu, J., Li, G., Liu, R., and Wang, J. (2019, January 27–28). Semi-supervised Skin Detection by Network with Mutual Guidance. Proceedings of the IEEE International Conference on Computer Vision, Seoul, Korea.10.1109/ICCV.2019.00220 ↗
14. Macedo, J., Costa, F., and dos Santos, J.A. (November, January 29). A benchmark methodology for child pornography detection. Proceedings of the 2018 31st SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI), Parana, Brazil.10.1109/SIBGRAPI.2018.00065 ↗
15. Video pornography detection through deep learning techniques and motion information10.1016/j.neucom.2016.12.017 ↗
16. Fusing audio vocabulary with visual features for pornographic video detection10.1016/j.future.2012.08.012 ↗
17. Pornographic Video Detection Scheme Using Multimodal Features
18. Liu, Q., Feng, C., Song, Z., Louis, J., and Zhou, J. (2019). Deep Learning Model Comparison for Vision-Based Classification of Full/Empty-Load Trucks in Earthmoving Operations. Appl. Sci., 9.10.3390/app9224871 ↗
19. Zeyer, A., Doetsch, P., Voigtlaender, P., Schlüter, R., and Ney, H. (2017, January 5–9). A comprehensive study of deep bidirectional LSTM RNNS for acoustic modeling in speech recognition. Proceedings of the 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), New Orleans, LA, USA.10.1109/ICASSP.2017.7952599 ↗
20. Song, K., and Kim, Y. (2019, January 21–24). A Fusion Architecture of CNN and Bi-directional RNN for Pornographic Video Detection. Proceedings of the 7th International Conference on Big Data Applications and Services, Jeju, Korea.
21. Google Deepmind (2016). Wavenet: A generative model for raw audio. arXiv.
22. Farha, Y.A., and Gall, J. (2019, January 15–20). Ms-tcn: Multi-stage temporal convolutional network for action segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA.10.1109/CVPR.2019.00369 ↗
23. Song, K., and Kim, Y. (2020). Pornographic Contents Detection Scheme using Bi-Directional Relationships in Audio Signal. J. Korea Contents Assoc., accepted.
24. RGB and Hue Color in Pornography Detection
25. Human Body Part Detection Representing Harmfulness of Images
26. Long short-term memory10.1162/neco.1997.9.8.1735 ↗
27. Sak, H., Senior, A.W., and Beaufays, F. (2014, January 14–18). Long short-term memory recurrent neural network architectures for large scale acoustic modeling. Proceedings of the 15th Annual Conference of the International Speech Communication Association, Singapore.10.21437/Interspeech.2014-80 ↗
28. Zhang, X., Yao, J., and He, Q. (2009, January 17–19). Research of STRAIGHT Spectrogram and Difference Subspace Algorithm for Speech Recognition. Proceedings of the 2nd International Congress on Image and Signal Processing, Tianjin, China.10.1109/CISP.2009.5303525 ↗
29. (2020, April 13). Keras: Deep Learning Library for Theano and Tensorflow. Available online: https://keras.io.
30. Kingma, D., and Ba, J. (2017). Adam: A method for stochastic optimization. arXiv.