Latent topics models have become a popular paradigm in many computer vision applications due to their capability to discover semantics in visual content. Various knowledge based object discovery algorithms for the classification problem in dependent images are appearing in the literature. However, these algorithms mostly suffer from the following two problems: image metadata and time measures. To overcome this kind of problem, this paper presents a Probabilistic Randomized Hough Transform (PRHT) with Deep Learning Classification Algorithm (DLC) algorithm performs the object discovery and localization used by deep learning classification algorithm. The proposed method of object regions are efficiently matched across images using a Probabilistic Randomized Hough Transform with Deep Learning Classification that evaluates the confidence for each candidate correspondence considering both appearance and spatial consistency. The achieved PRHT-DLC has high accuracy and performance increases compared to the previous method of Pipeline method and Latent Dirichlet allocation (LDA) algorithms.

Image Mining, Image Retrieval, Probabilistic Randomized Hough Transform, Deep learning, Unsupervised object discovery

[1] A. Joulin, F. Bach, and J. Ponce. Discriminative clustering for image co-segmentation. In CVPR, 2010.
[2] D. Mimno, H. M. Wallach, E. Talley, M. Leenders, and A. McCallum, “Optimizing semantic coherence in topic models,” in Proc. EMNLP, 2011, pp. 262–272.
[3] T. Deselaers, B. Alexe, and V. Ferrari, “Weakly supervised localization and learning with generic knowledge,” Int. J. Comput. Vis., vol. 100, no. 3, pp. 275–293, 2012.
[4] Z. Chen, A. Mukherjee, B. Liu, M. Hsu, M. Castellanos, and R. Ghosh, “Leveraging multi-domain prior knowledge in topic models,” in Proc. IJCAI, 2013, pp. 2071–2077.
[5] M. Rubinstein, J. Kopf, C. Liu, and A. Joulin, “Unsupervised joint object discovery and segmentation in Internet images,” in Proc. CVPR, Jun. 2013, pp. 1939–1946.
[6] A. Faktor and M. Irani, “Clustering by composition’—Unsupervised discovery of image categories,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 36, no. 6, pp. 1092–1106, 2014.
[7] A. Joulin, K. Tang, and L. Fei-Fei, “Efficient image and video co-localization with Frank–Wolfe algorithm,” in Proc. ECCV, 2014, pp. 253–268.
[8] Z. Niu, G. Hua, X. Gao, and Q. Tian, “Semi-supervised relational topic model for weakly annotated image recognition in social media,” in Proc. CVPR, 2014, pp. 4233–4240.
[9] L. Haldurai and V. Vinothini, “Parallel Indexing on Color and Texture Feature Extraction using R-Tree for Content Based Image Retrieval”, International Journal of Computer Sciences Engineering, Volume-03, Issue-11, pp.11-15, 2015.
[10] C. Wang, K. Huang, W. Ren, J. Zhang, and S. Maybank, “Largescale weakly supervised object localization via latent category learning,” IEEE Trans. Image Process., vol. 24, no. 4, pp. 1371–1385, 2015.
[11] M. Cho, S. Kwak, C. Schmid, and J. Ponce, “Unsupervised object discovery and localization in the wild: Part-based matching with bottomup region proposals,” in Proc. CVPR, pp. 1201–1210, 2015.
[12] L. Haldurai, T. Madhubala and R. Rajalakshmi, “A Study on Genetic Algorithm and its Applications”, International Journal of Computer Sciences Engineering, Volume-04, Issue-10, pp.139-143, 2016.
[13] Zhenzhen Wang ; Junsong Yuan, “Simultaneously Discovering and Localizing Common Objects in Wild Images”, IEEE Transactions on Image Processing, Vol: 27, Issue: 9, 2018.
[14] Mereena Johny and L. Haldurai, “A Brief Survey on Dynamic Topic Model for Unsupervised Object Discovery and Localization”, International Journal of Computer Sciences Engineering, Volume-06, Issue-09, pp.567-571, 2018.
[15] Zhenxing Niu, Gang Hua, Le Wang, Member, and Xinbo Gao, “Knowledge-Based Topic Model for Unsupervised Object Discovery and Localization”, IEEE transactions on image processing, vol. 27, no. 1, 2018