Abstract
Data security is critical. Cryptography and steganography are two of the most common security techniques. The hacker immediately identifies the secret information after following a few paths. This paper provides a new method for introducing secret data into a crowded scene using cryptography and the concept of image steganography, without employing the embedding notion. In this paper, firstly a novel supervised learning framework for detecting abnormalities in varied crowded scenarios is proposed. Visual features, motion features, and energy features are all available for busy settings. These characteristics are derived from spatiotemporal measurements. Three convolutional machines are trained for mid-level feature representation, and then a multimodal fusion model is used to deep understand the crowd patterns. One class support vector machine is used to track and detect abnormalities in a crowded scene based on the results of multimodal fusion. Apply the notion of deep conventional neural network to the image from the first step. Use one of the learned neural network techniques to map secret data into a vector. To obtain a stego image using this method, no mapping or embedding techniques are necessary. After the training phase, we utilize another neural network called an extractor to extract data from the stego image. We may also embed images into other images with two distinct networks using this technique. The secret image is embedded in the original image using the prep network. The hidden image is extracted from the stego image using the reveal network. This suggested technique was trained on a variety of data types and produced better outcomes in terms of embedding rate and extraction rate and payload capacity.
Keywords:
:Deep Learning, Feature Extraction, Multimodal Fusion, Anomalies, Support Vector Machines, Cryptography, Steganography, Deep Conventional Neural Networks
References
[1] V. Mahadevan “Detection of anomalies in crowded scenes for 3D Images,” in International Conference on Security and image processing), pp. 1985–1991, June 2016.
[2] A. A. Sodeemann, M. P. and B. J. Borgghetti, “A review of various anomaly detection in automated surveillance,” security and its applications , , vol. 42, no. 6, pp. 1267–1272, 2016.
[3] M. Sabokrrou, , “anomaly detection and localization in crowded scenes for 2D and 3D images,” in Proceedings of the IEEE Conference on multimedia and computer fusion pp. 2015.
[4] A. Deel Giorno, J., “A discriminative framework for anomaly detection in large videos for identifications,” in Proceedings of the Italian Conferencce onComputer systems (ICSV’16), vol., 2016.
[5] B. Zhaoo, L. Feii-Feii, , “ detection of various unusual events in videos via dynamic sparse coding and conventional coding,” in Interantional Conference on Computer Security
[6] Moore, and M. Shah, “Chaootic invariants trajectories for anomaly detection in crowdedscenes,” in Proceedings of the on Computer Security and Image Processing , pp. 2064–2070, San Francisco, Calif, USA, June 2017.
[7] A. Krizheevsky, I., “ImageNetclassification with deep convolutional neural networks,” Communicationsof the ACM, and FEE vol. 60, no. 6, pp. 87–93, 2018.
[8] R. Giirshick, J., “Rich feaaturehierarchies for accurate object detection , object identification and semanticsegmentation,” in 11th International conference on image processing
[9] C. Feichhtenhofer, A. Pinz, and A “Convolutional of two-stream network fusion for video action recognition and pattern ,” inProceedings of the Interantional Confercne Communciations ), pp. 1933–1941,USA, July 2017
[10] N. Wanng and D.-Y. Yeuung, “Learning and testing deep compact imagerepresentation for visual tracking,” in Advances in Networks and its applciations, pp. 819–827, 2016.
[11] J. Zhaung, “Autoencoder and decoder Networks (CFAN) for real-time face alignment and processing ,” inProceedings of the Itlaian conference on computer security 2016.
[12] D. Xuy, Y. Yaan, J. Siong, and N. Seble, “ Trining deep representationsof appearance and motion for anomalous event detection,”in Proceedings of the Confernce on vision (BMVC’16), Swansea, UK, 2016.
[13] Dr.D.Rathna Kishore, Dr.D.Suneetha A Secure Steganography Approach For Cloud Data Using Ann Along With Private Key Embedding”, International Journal of Computer Science and Information Security (IJCSIS), Vol. 16, No. 6, June2018
[14] Y. Feng, Y Cahm, “Learning deep event models,” Neurocomputing, vol. 219, pp. 558–566, 2017.
[15] T K. Jiayani , Y. Malise, “PCANet:a simple deep learning baseline for image classification and clustering ?” International Confernce on Image Processing, vol. 24, no. 12, pp. 5045-5056 2017.
[16] Wu. Li, V. Mahadeevan, and N. Vascoelos, “Anomaly detection and decentraliazaiiton localization in crowded scenes,” National Confernce on Pattern regocginition in new trends , vol. 36, no. 1, pp. 18–22, 2014.
[17] J.Kimannd Man“: a spacetimeMRF for detecting abnormal activities with incremental and external updates,” in Interantional conference recent ternds, pp. 2921–2928, June 2018.
[18] Dr.D.Rathna Kishore, Dr.D.Suneetha Data Security Model Using Artificial Neural Networks and Database Fragmentation in Cloud Environment”, International Journal of Recent Technology and Engineering (IJRTE) ISSN: 2277- 3878, Volume-8 Issue-2, July 2019
[19] V. Reddy, “Improved anomalydetection in crowded scenes via cell-based analysis of foreground and background pixel speed, size and texture,” in Proceedings of the ComputerSociety Conference on Recent ternds in Computer), pp. 65–71, IEEE, June 2014.
[20] H. Leeko, R. Grosse rosma , “Learning componemts from supervised learning ,” Communications of the international conference on image processing and machine learning, vol. 54,
[21] Dr.D.Rathna Kishore, Dr.D.Suneetha “Deep Convolutional Neural Network based Image Steganogrpahy Technique for Audio- Image Hiding Algorithm” International Journal of Engineering and Advanced Technology (IJEAT), ISSN: 2249 –8958, Volume-9 Issue-4, April 2020
[22] Raja Bala Video Anomaly Detection 14 March 2018 Proceedings of the ComputerSociety Conference on Recent ternds in Computer), pp. 65–71
[23] Jones, M.J.; Ramachandra, A New Dataset And Evaluation Protocol For Video Anomaly Detection MITSUBISHI ELECTRIC RESEARCH LABORATORIES January 19, 2019
[24] Yingying Zhu” Context-Aware Activity Recognition And Anomaly Detection In Video” Ieee Journal Of Selected Topics In Signal Processing, Vol. 7, No. 1, January 2019
