طبقهبند خودسازمانده هندسی مبتنی بر یادگیری فعال برای نهانکاوی در محیط ویدئو با صرف حداقل برچسب
محورهای موضوعی : مهندسی برق و کامپیوترهادی صدوقی یزدی 1 * , علی محی الدینی شاهم آبادی پور 2 , مرتضی خادمی 3
1 - دانشگاه فردوسی مشهد
2 - دانشگاه شهید باهنر کرمان
3 - دانشگاه فردوسی مشهد
کلید واژه: طبقهبند خودسازمانده پویای شبهناظرنهانکاوی کور ویدئویادگیری شبهناظریادگیری فعال,
چکیده مقاله :
طبقهبند یکی از سه بلوک تشکیلدهنده یک نهانکاو ویدئو است که برای آموزش نیازمند برچسب میباشد. در نهانکاوی کور به دلیل عدم دسترسی به الگوریتمهای نهاننگاری تهیه برچسب مشکل است. در این مقاله از طبقهبند خودسازمانده پویای شبهناظر برای رسیدن به حداقل برچسب استفاده شده و بدین منظور مفهومی به نام افزونگی هندسی گرههای لایه زیرین شبکه خودسازمانده پویای شبهناظر به کار گرفته شده است. نشان داده شده که این افزونگی منجر به ایجاد الگوهای تکراری برای شبکه خواهد شد، پس حذف چنین گرههایی بلامانع است. اثبات شده به دلیل وجود تناظر یک به یک بین گرهها و برچسبها کاهش گرهها منجر به کاهش تعداد برچسب لازم میشود. نکته اساسی این که لازمه وجود افزونگی هندسی در میان تعدادی گره که مفهومی انتزاعی است، تشکیل دسته توسط آنهاست و بنابراین مبنای الگوریتم پیشنهادی شناسایی دستهها و ادغام اعضای آنهاست. طبقهبند به دست آمده بر این مبنا طبقهبند خودسازمانده هندسی نام نهاده شده و اثبات میشود که این طبقهبند میتواند به مقدار بهینه حداقل برچسب دست یابد. نتایج شبیهسازی نشاندهنده برتری چشمگیر طبقهبند نسبت به الگوریتمهای پیشین است.
Classifier is one of the three blocks of a video steganalysis that needs labeled for training. In the blind video steganalysis, due to the lack of access to steganography algorithms, it is difficult to label. In this paper, the semi supervised growing self-organizing map classifier has been used to reach the minimum label. For this purpose, a concept called the geometric redundancy of the lower-layer nodes of the semi supervised self-organizing network has been used. It has been shown that this redundancy will create repetitive patterns of the network, so deleting such nodes is possible. Proven due to the existence of one-to-one correspondence between nodes and labels. Reducing nodes leads to a reduction in the number of labels required. The basic point is the need for a geometric redundancy among a number of nodes, which is a conception of abstraction, is the formation of a group by them. Therefore, the proposed algorithm is based on identifying categories and integrating their members. The classifier obtained on this basis has been named a geometric self-organizing map classifier .It is proven that this classifier can achieve the minimum amount of optimal label. The simulation results show a remarkable superiority over the previous algorithms.
[1] N. Zarmehi and M. A. Akhaee, "Digital video steganalysis toward spread spectrum data hiding," IET Image Processing, vol. 10, no. 1, pp. 1-8, Dec. 2015.
[2] A. Kolakalur, I. Kagalidis, and B. Vuksanovic, "Wavelet based color video steganography," IACSIT International J. of Engineering and Technology, vol. 8, no. 3, pp. 165-169, Jun. 2016.
[3] H. Karimi, M. G. Shayesteh, and M. A. Akhaee, "Steganalysis of JPEG images using enhanced neighbouring joint density features," IET Image Processing, vol. 9, no. 7, pp. 545-552, Jul. 2015.
[4] U. Budia, D. Kundur, and T. Zourntos, "Digital video steganalysis exploiting statistical visibility in the temporal domain," IEEE Trans. on Information Forensics and Security, vol. 1, no. 4, pp. 502-516, Dec. 2006.
[5] J. S. Jainsky, D. Kundur, and D. R. Halverson, "Towards digital video steganalysis using asymptotic memoryless detection," in Proc. of the 9th Int. Workshop on Multimedia and Security, pp. 161-168, Dallas, TX, USA, 20-21 Sept. 2007.
[6] V. Pankajakshan and A. T. S. Ho, "Improving video steganalysis using temporal correlation," in Proc. of the 3rd Int. Conf. on Intelligent Information Hiding and Multimedia Signal Processing, IIH-MSP'07, vol. 1, pp. 287-290, Kaohsiung, Taiwan, 26-28 Nov. 2007.
[7] Y. Su, C. Zhang, L. Wang, and C. Zhang, "A new video steganalysis based on mode detection," in Proc. of the Int. Conf. on Audio, Language and Image Processing, pp. 1507-1510, Shanghai, China, 7-9 Jul. 2008.
[8] K. Tasdemir, F. Kurugollu, and S. Sezer, "Spatio-temporal rich model-based video steganalysis on cross sections of motion vector planes," IEEE Trans. on Image Processing, vol. 25, no. 7, pp. 3316-3328, Jul. 2016.
[9] G. Xuan, et al., "Steganalysis based on multiple features formed by statistical moments of wavelet characteristic functions," in Proc. of the 7th Information Hiding Workshop, Lecture Notes in Computer Science, vol 3727, pp. 262-277, Jun. 2005.
[10] Y. Q. Shi, et al., "Steganalysis based on moments of characteristic functions using wavelet decomposition, prediction-error image, and neural network," in Proc. of the IEEE Int. Conf. on Multimedia and Expo, pp. 269-272, Amsterdam, Netherlands, 6-6 Jul. 2005.
[11] ج. مرتضوی مهریزی، م. خادمي و ه. صدوقی یزدی، "نهانکاوی کور ویدئو با رویکرد یادگیری شبهناظر برای الگوریتمهای نهاننگاری ویدئوی مبتنی بر بردارهای حرکت،" نشريه مهندسي برق و مهندسي كامپيوتر ايران، ب- مهندسي کامپیوتر، سال 13، شماره 2، صص. 100-89، زمستان 1394.
[12] Y. Cao, X. Zhao, and D. Feng, "Video steganalysis exploiting motion vector reversion-based features," IEEE Trans. on Signal Processing, vol. 19, no. 1, pp. 35-38, Nov. 2012.
[13] Y. Deng, Y. Wu, H. Duan, and L. Zhou, "Digital video steganalysis based on motion vector statistical characteristics," International Journal for Light and Electron Optics, vol. 124, no. 14, Jul. pp. 1705-1710, 2013.
[14] Y. S. Su, C. Zhang, and C. Zhang, "A video steganalytic algorithm against motion vector-based steganography," Signal Process., vol. 91, no. 18, pp. 1901-1909, Aug. 2011.
[15] X. Xu, J. Dong, W. Wang, and T. Tan, "Video steganalysis based on the constraints of motion vectors," in Proc. IEEE Int.Conf. on Image Processing, ICIP'13, pp. 4422-4426, Melbourne, VIC, Australia, 15-18 Sept. 2013.
[16] H. A. Aly, "Data hiding in motion vectors of compressed video based on their associated prediction error," IEEE Trans. on Information Forensics and Security, vol. 6, no. 1, pp. 14-18, Nov. 2011.
[17] C. Xu, X. Ping, and T. Zhang, "Steganography in compressed video stream," in Proc. IEEE First Int. Conf. Innovative Computing, Information and Control, ICICIC'06, vol. 1, pp. 269-272, Beijing, China, 30 Aug.-1 Sept. 2006.
[18] D. Y. Fang and L. W. Chang, "Data hiding for digital video with phase of motion vector," in Proc. IEEE Int. Symp. Circuits and Systems, pp. 1422-1425, Island of Kos, Greece, 21-24 May 2006.
[19] D. Alahakoon and K. Halgamunge, "Dynamic self-organizaing maps with controlled growth for knowledge discovery," IEEE Trans. on Neural Networks, vol. 11, no. 3, pp. 601-614, May 2000.
[20] A. Hsu and S. K. Halgamunge, "Class structure visualization with semi-supervised growing self-organizing maps," Neurocomputing, vol. 71, no. 16-18, pp. 3124-3130, Oct. 2008.
[21] B. Settles, Active Learning Literature Survey, Computer Sciences Technical Report 1648 University of Wisconsin-Madison Updated on: Jan. 2010.
[22] B. Settles, R. J. Brachman, W. W. Cohen, and T. Dietterich, Active Learning, Synthesis Lectures on Artificial Intelligence and Machine Learning, Morgan & Claypool Publishers, 2012.
[23] H. Wu and S. Prasad, "Dirichlet process based active learning and discovery of unknown classes for hyperspectral image classification," IEEE Trans. on Geoscience and Remote Sensins, vol. 54, no. 8, pp. 4882-4895, Aug. 2016.
[24] Q. Shi, B. Du, and L. Zhang, "Spatial coherence based batch-mode activelearning for remote sensing image classification," IEEE Trans. Image Process., vol. 24, no. 7, pp. 2037-2050, Jul. 2015.
[25] Z. Zhang, E. Pasolli, H. Lexie Yang, and M. M. Crawford, "Multimetric active learning for classification of remote sensing data," IEEE Geoscience and Remote Sensing Letters, vol. 13, no. 7, pp. 1007-1011, Jul. 2016.
[26] L. Wan, K. Tang, M. Li, Y. Zhong, and A. K. Qin, "Collaborative active and semisupervised learning for hyperspectral remote sensing image classification," IEEE Trans. on Geoscience and Remote Sensing, vol. 53, no. 5, pp. 2384-2396, May 2015.
[27] E. Pasolli, H. L. Yang, and M. M. Crawford, "Active-metric learning for classification of remotely sensed hyperspectral images," IEEE Trans. on Geoscience and Remote Sensing, vol. 54, no. 4, pp. 1925-1939, Apr. 2016.
[28] H. Lin and P. Li, "Circuit performance classification with active learning guided sampling for support vector machines," IEEE Trans. on Computer-Aided Design of Integrated Circuits and Systems, vol. 34, no. 9, pp. 1467-1480, Sep. 2015.
[29] A. Mehrizi and H. Sadoghi Yazdi, "Semi-supervised GSOM integrated with extreme learning machine intelligent data analysis," Intelligent Data Analysis, vol. 20, no. 5, pp. 1115-1132, Sept. 2016.