ABSTRACT
Indoor scene classification is a challenging problem in computer vision. In order to achieve an accurate solution for this task, a model that can exploit the discriminating information between different scene categories is necessary. In this paper, local feature extraction is suggested with supervised classification techniques for scene recognition in indoor environments. A comparative study between several feature detectors: SIFT, SURF, FAST, ORB, MSER, BRISK, and several local descriptors: SIFT, SURF, ORB, BRISK is presented. Two different classifiers, SVM and k-NN, are used for classification. The different techniques have been tested using the MIT dataset for indoor scenes and all the corresponding performance of each combination has been reported. Upon inspecting obtained results, it is observed that the combination of MSER detector with ORB as descriptor using SVM performs the best, as they give a combination of relatively high accuracy and low complexity; in terms of average execution time and memory space.
- Bay, H., Ess, A., Tuytelaars, T., & Van Gool, L. (2008 June). Speeded-up robust features (SURF). Computer vision and image understanding, 110(3), 346--359. Google ScholarDigital Library
- Bosch, A., Munoz, X., & Mart, R. (2007, June). A review: Which is the best way to organize/classify images by content? Image and Vision Computing, 25(6), 778--791. Google ScholarDigital Library
- Burges, C. J. (1998). A Tutorial on Support Vector Machines for Pattern Recognition. Image and vision computing, 2(2), 121--167. Google ScholarDigital Library
- Calonder, M., Lepetit, V., Strecha, C., & Fua, P. (2010). BRIEF: Binary Robust Independent Elementary Features. Computer Vision-ECCV 2010, 778--792. Google ScholarDigital Library
- Cover, T. M., & Hart, P. E. (1967). Nearest neighbor pattern classification. IEEE Transactions on Information Theory, 13(1), 21--27. Google ScholarDigital Library
- El-gayar, M., & Soliman, H. (2013). A comparative study of image low level feature extraction algorithms. Egyptian Informatics Journal, 14(2), 175--181.Google ScholarCross Ref
- Espinace, P., Kollar, T., Soto, A., & Roy, N. (2010). Indoor scene recognition through object detection. IEEE International Conference on Robotics and Automation (ICRA), (pp. 1406--1413).Google ScholarCross Ref
- Espinace, P., Kollar, T., Soto, A., & Roy, N. (2013, sep). Indoor Scene Recognition Through Object Detection Using adaptive Objects Search. in IEEE International Conference on Robotics and Automation, 61(6), 932--947.Google ScholarDigital Library
- Hassaballah, M., Abdelmgeid, A. A., & Alshazly, H. A. (2016). Image Feature Detection, Description and Matching. In Image Feature Detectors and Descriptors (pp. 11--45). Springer.Google Scholar
- Heinly, J., Dunn, E., & Frahm, J.-M. (2012). Comparative evaluation of binary features. Computer Vision--ECCV 2012 (pp. 759--773). Springer.Google Scholar
- King, R. D., Henery, R., Feng, C., & Sutherland, A. (1994). A Comparative Study of Classification Algorithms: Statistical, Machine Learning and Neural Network. Machine Intelligence 13, (pp. 311--359). Google ScholarDigital Library
- Leutenegger, S., Chli, M., & Siegwart, R. Y. (2011). BRISK: Binary robust invariant scalable keypoints. IEEE International Conference on Computer Vision (ICCV), (pp. 2548--2555). Google ScholarDigital Library
- Lowe, D. G. (2004). Distinctive image features from scale-invariant keypoints. International journal of computer vision, 60(2), 91--110. Google ScholarDigital Library
- Mair, E., Hager, G. D., Burschka, D., Suppa, M., & Hirzinger, G. (2010). Adaptive and Generic Corner Detection Based on the Accelerated Segment Test. In Computer Vision-ECCV'10 (pp. 183--196). Springer. Google ScholarDigital Library
- Obdržálek, D., Basovník, S., Mach, L., & Mikulík, A. (2009). Detecting scene elements using maximally stable colour regions. In Research and Education Robotics-EUROBOT 2009 (pp. 107--115). Springer.Google Scholar
- Quattoni, A. & Torralba, A. (2009). Recognizing indoor scenes. IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2009, (pp. 413--420).Google ScholarCross Ref
- Rublee, E., Rabaud, V., Konolige, K., & Bradski, G. (2011). ORB: an efficient alternative to SIFT or SURF. IEEE International Conference on Computer Vision (ICCV), (pp. 2564--2571). Google ScholarDigital Library
- Stankovic, R. S., & Falkowski, B. J. (2003). The Haar wavelet transform: its status and achievements. Computers and Electrical Engineering, 29(1), 25--44.Google ScholarCross Ref
- Szummer, M., & Picard, R. W. (1998). Indoor-outdoor image classification. In Proceedings of the 1998 IEEE International Workshop on Content-Based Access of Image and Video Databases, (pp. 42--51). Google ScholarDigital Library
- Thamilselvan, P., & Sathiaseelan, J. (2015). A Comparative Study of Data Mining Algorithms for Image Classification. Education and Management Enginerring.Google Scholar
- Torralba, A., Murphy, K. P., Freeman, W. T., & Rubin, M. A. (2003). Context-based vision system for place and object recognition. In Proceedings of the Ninth IEEE International Conference on Computer Vision, (pp. 273--280). Google ScholarDigital Library
- Vailaya, A., Jain, A., & Zhang, H. J. (1998). On image classification: city vs. landscapes. Pattern Recognition, 31(12), 1921--1935.Google ScholarCross Ref
- Viswanathan, D. G. (2011). Features from Accelerated Segment Test (FAST).Google Scholar
Recommendations
Performances of Invariant Feature Detectors in Real-Time Video Applications
IVIC 2013: Third International Visual Informatics Conference on Advances in Visual Informatics - Volume 8237This paper reviews and compares the performance of five well-known detectors, SIFT, SURF, ORB, MSER and STAR, when combined in combination of with using three common descriptors, SIFT, SURF and ORB. To validate the results, these descriptors' ...
A Comparison of Local Detectors and Descriptors for Multi-Object Applications
IPAC '15: Proceedings of the International Conference on Intelligent Information Processing, Security and Advanced CommunicationThere have been a number of evaluations and comparisons between feature detectors and descriptors and between their different implementations. These detectors and descriptors respond differently depending on the image structure. In this paper, we ...
An evaluation of local interest regions for non-rigid object class recognition
Highlights EdgeLap, SURF, HarAff, HesAff, kAS, FAST, IBR, PCBR, HarLap, HesLap, Salient, MSER, and DoG are evaluated. EdgeLap interest region detector reaches to the highest success-rate. A new discriminancy measure based on random region success-rate ...
Comments