Abstract
The concept of smart cities has quickly evolved to improve the quality of life and provide public safety. Smart cities mitigate harmful environmental impacts and offences and bring energy-efficiency, cost saving and mechanisms for better use of resources based on ubiquitous monitoring systems. However, existing visual ubiquitous monitoring systems have only been developed for a specific purpose. As a result, they cannot be used for different scenarios. To overcome this challenge, this paper presents a new ubiquitous visual surveillance mechanism based on classification of scene images. The proposed mechanism supports different applications including Soil, Flood, Air, Plant growth and Garbage monitoring. To classify the scene images of the monitoring systems, we introduce a new technique, which combines edge strength and sharpness to detect focused edge components for Canny and Sobel edges of the input images. For each focused edge component, a patch that merges nearest neighbor components in Canny and Sobel edge images is defined. For each patch, the contribution of the pixels in a cluster given by k-means clustering on edge strength and sharpness is estimated in terms of the percentage of pixels. The same percentage values are considered as a feature vector for classification with the help of a Support Vector Machine (SVM) classifier. Experimental results show that the proposed technique outperforms the state-of-the-art scene categorization methods. Our experimental results demonstrate that the SVM classifier performs better than rule and template-based methods.
Similar content being viewed by others
References
Addabbo AD, Refice A, Pasquariello G, Lovergine FP, Capolongo D, Manfreda S (2016) A Bayesian network for flood detection combining SAR imagery and ancillary data. IEEE Trans Geosci Remote Sens 3612–3625
Afsharinejad A, Davy A, Jennings B, Brennan C (2016) Performance analysis of plant monitoring nanosensor networks at THz frequencies. IEEE Internet Things J 59–69
Arabe SB, Gao X, Wang B, Yang F, Brost V (2014) Multi-kernel implicit curve evolution for selected texture region segmentation in VHR satellite images. IEEE Trans Geosci Remote Sens 5183–5192
Bai S (2017) Growing random forest on deep convolutional neural networks for scene categorization. Expert Syst Appl 279–287
Bar, E. O., & Trivedi, M. M. (2017) Are all objects equal? Deep spatio-temporal importance prediction in driving videos”, Patt Recog 425–436
Bastawesy BE, Ali RR, Deocampo DM, Baroudi MSA (2012) Detection and assessment of the waterlogging in the dryland drainage basins using remote sensing and GIS techniques. IEEE J Select Topics Appl Earth Observ Remote Sens 1564–1571
Bosch A, Zisserman A, Munoz X (2008) Scene classification using a hybrid generative/discriminative approach. IEEE Trans Pattern Anal Mach Intell 30(4):712–727
Davis TW, Liang X, Kuo CM, Liang Y (2012). Analysis of power characteristics for sap flow, soil moisture and soil water potential sensors in wireless sensor networking systems. IEEE Sens J 1933–1945
Dornaika F, Moujahid A, Merabt YE, Ruichek Y (2016) Building detection from orthophotos using a machine learning approach: an empirical study on image segmentation and descriptors. Expert System with Applications 58:130–142
Du V, Ling H (2016) Dynamic scene classification using redundant spatial scenelets. IEEE Trans Cybernet 2156–2165
Dunlop H (2010) Scene classification and video via semantic segmentation. In Proc Int Workshop Comput Vis Patt Recog Workshop 72–79
Google API System: https://cloud.google.com/vision/
Guerrero JM, Guijarro M, Montalvo M, Romeo J, Emmi L, Ribeiro A, Pajares G (2013) Automatic expert system based on images for accuracy crop row detection in maize fields. Exp Syst Appl 656–664
Hastie T, Tibshirani R (1998) Classification by pairwise coupling. Ann Stat 26(2):451–471
Hayat M, Khan SH, Bennamoun M, An S (2016) A spatial layout and scale invariant feature representation for Indoor scene classification. IEEE Trans Image Proc 4289–4841
Jang SW, Cha SH (2014) An approach to segmenting initial object movements in visual sensor networks. Int J Distrib Sensor Netw
Karpathy A, Toderici G, Shetty S, Leung T, Sukthankar R, Fei LF (2014) Large-scale video classification with convolutional neural networks. In Proc Int Conf Comput Vis Patten Recog 1725–1732
Li Z, Zheng J (2015) Edge preserving decomposition based single image haze removal. IEEE Trans Image Proc 5432–5441
Liu J, Chen C, Zhu Y, Liu W, Metaxas DN (2016) Video classification via weakly supervised sequence modeling. Comput Vis Image Understand 79–87
Liu W, Wang Z, Liu X, Zeng N, Liu Y, Alsaadi FE (2017) A survey of deep neural network architecture and their applications. Neurocomputing 11–26
Longhi, S., Marzioni, D., Alidori, E. & Buo, G. D. (2012) Solid waste management architecture using wireless senor networks technology. In Proc Int Conf New Technol Mobil Sec 1–5
Lu X, Manduchi R (2011) Fast image motion segmentation for surveillance applications. Image Vis Comput 104–116
Messer H, Sendik O (2014) A new approach to precipitation monitoring: A critical survey of technologies and challenges. IEEE Sig Proc Mag 110–122
Nirmala DE, Vignesh RK, Vaidehi V (2013). Multimodal image fusion in visual sensor networks. In Proc. International Conference on Electronics, Computing and Communication, 1–6
Nogueira, K., Penatti, O. A. B., & Santos, J. A. D. (2017) Towards better exploiting convolutional neural networks for remote sensing scene classification. Patt Recogn 539–556
Olive. A., Modeling the Shape of the Scene: A Holistic Representation of the Spatial Envelope (2001) Int J Comput Vis 42(3):145–175
Qin L, Shivakumara P, Lu T, Pal V, Tan CL (2016) Video scene text frames categorization for text detection and recognition. In Proc Int Conf Patt Recog 3875–3880
Rao VSH, Kumar MN (2012) A new intelligence based approach for computer aided diagnosis of dengue fever. IEEE Trans Info Technol Biomed 112–118
Shaban KB, Kadri A, Rezk E (2016) Urban air pollution monitoring system with forecasting models. IEEE Sens J 2598–2606
Sharma A, Sankar PK (2016) Adapting off-the-self CNNs for word spotting and recognition. In Proc Int Conf Document Anal Recog 986–990
Shou Z, Wang D, Chang SF (2016) Temporal action localization in untrimmed videos via multi-stage CNNs. In Proc Int Conf Comput Vis Patt Recog 1049–1058
Sun X, Liu Z, Hu Y, Zhang L, Zimmermann R (2018) Perceptual multi-channel visual feature fusion for scene categorization. Inf Sci 429:37–48
Tian D, Sun H, Vetro A (2016) Keypoint trajectory coding on compact descriptor for video analysis. In Proc Int Conf Image Proc 171–175
Wu Z, Fu Y, Jiang YG, Sigal L (2016) Harnessing object and scene semantics for large-scale video understanding”, In Proc. Int Conf Comput Vis Patt Recog 3112–3121
Xie L, Tian Q, Zhang B (2016) Simple technique make sense: Feature pooling and normalization for image classification, IEEE Trans Circ Syst Video Proc 1251–1264
Yang DW, Park HW (2016) A new shape feature for vehicle classification in thermal video sequences. IEEE Trans Circ Syst Video Proc 1363–1375
Yi X, Eramian M (2016) LBP based segmentation of defocus blur. IEEE Trans Image Proc 1626–1638
Yuan L, Chen F, Zhou L Hu D (2015) Improve scene classification by using feature and kernel combination. Neurocomputing 213–220
Zhang X, Feng X, Wang W, Xue W (2013) Edge strength similarity for image quality assessment. IEEE Trans Image Proc 319–322
Zhou Q, Zheng B, Zhu W, Latecki LJ (2016) Multi-scale context for scene labeling via flexible segmentation graph. Patt Recog 312–324
Zhu J, Wu T, Zhu SC, Yang X, Zhang W (2016) A reconfigurable tangram model for scene representation and categorization. IEEE Trans Image Proc 150–166
Zuo Z, Wang G, Shuai B, Zhao L, Yang Q (2015) Exemplar based deep discriminative and sharable feature learning for scene image classification. Patt Recog 3004–3015
Acknowledgements
This research work was supported by the Faculty of Computer Science and Information Technology, University of Malaya under a special allocation of Post Graduate Funding for the RP036B-15AET project. The authors also extend their appreciation to the Dean of Scientific Research at King Saud University for funding this work through Research Group Number (RGP-288). The authors convey special thanks to Sangheeta Roy, Faculty of Computer Science and Information Technology, University of Malaya for her help in implementing existing classification methods and conducting experiments for the comparative study.
Author information
Authors and Affiliations
Corresponding authors
Rights and permissions
About this article
Cite this article
Kaljahi, M.A., Palaiahnakote, S., Anisi, M.H. et al. A scene image classification technique for a ubiquitous visual surveillance system. Multimed Tools Appl 78, 5791–5818 (2019). https://doi.org/10.1007/s11042-018-6151-x
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-018-6151-x