Abstract
This study presents a system that can automatically colorize grayscale images in large quantities. To enable big data training, divide-and-conquer support vector machines (SVMs) also proposed as classifiers are frequently used in this study. The system is composed of two components—image classification and local-descriptor classification. The former firstly analyzes an input by using a classifier, so that the system can determine which class should serve as the knowledge base. After the class is decided, the latter stage subsequently uses this knowledge base as the reference to colorize the input. Experimental results showed that the accuracy of classification in image classification could reach 90.50 %. Moreover, in the local-descriptor classification, the majority of pixels were successfully assigned correct colors. During the efficiency test, the proposed divide-and-conquer SVM enhanced computational speed while maintaining the accuracy. Such findings demonstrate the effectiveness of the proposed method and the feasibility of our idea.
Similar content being viewed by others
References
Levin A, Lischinski D, Weiss Y (2004) Colorization using optimization. ACM Trans Graph 23(3):689–694
Yatziv L, Sapiro G (2005) Fast image and video colorization using chrominance blending. IEEE Trans Image Process 15(5):1120–1129
Charpiat G, Bezrukov I, Hofmann M, Altun Y, Schölkopf B (2010) Machine learning methods for automatic image colorization. In: Lukac R (ed) Computational photography: methods and applications. CRC Press, Boca Raton, pp 395–418
Sousa A, Kabirzadeh R, Blaes P (2014) Automatic colorization of grayscale images. Department of Electrical Engineering, Stanford University, Stanford, CA
Divide and conquer algorithms, Wikipedia, The Free Encyclopedia. http://en.wikipedia.org/wiki/Divide_and_conquer_algorithms. Accessed 16 Jan 2015
Kung S-Y (2014) Kernel methods and machine learning. Cambridge University Press, Cambridge
Chen B-W, Chen C-Y, Wang J-F (2013) Smart homecare surveillance system: behavior identification based on state transition support vector machines and sound directivity pattern analysis. IEEE Trans Syst Man Cybern Syst 43(6):1279–1289
Guo, Q, Chen, B-W, Jiang F, Ji X, Kung S-Y (2015) Efficient divide-and-conquer classification based on feature-space decomposition. http://arxiv.org/abs/1501.07584. Accessed 29 Jan 2015
Paul A, Chen B-W, Bharanitharan K, Wang J-F (2013) Video search and indexing with reinforcement agent for interactive multimedia services. ACM Trans Embed Comput Syst 12(2)
Kung S-Y, Wu P-Y (2012) On efficient learning and classification kernel methods. In: Proceedings of 2012 IEEE international conference on acoustics, speech, and signal processing (ICASSP 2012), Kyoto, Japan, 25–30 March 2012, pp 2065–2068
Chang EY, Zhu K, Wang H, Bai H, Li J, Qiu Z, Cui H (2007) PSVM: parallelizing support vector machines on distributed computers. In: Proceedings of 21st annual conference on neural information processing system (NIPS 2007), Vancouver, British Columbia, Canada, 03–06 Dec 2007, pp 257–264
Gu Q, Han J (2013) Clustered support vector machines. In: Proceedings of 16th international conference on artificial intelligence and statistics (AISTATS), Scottsdale, Arizona, United States, 29 Apr–01 May 2013, pp 307–315
Zhang Y, Duchi J, Wainwright M (2013) Divide and conquer kernel ridge regression. In: Proceedings of conference on learning theory (Colt 2013), Princeton, New Jersey, United States, 12–14 June 2013, pp 592–617
Hsieh C-J, Si S, Dhillon IS (2013) A divide-and-conquer solver for kernel support vector machines. In: Proceedings of 31st international conference on machine learning (ICML 2013), Beijing, China
Lin C-J, Moré JJ (1999) Incomplete Cholesky factorizations with limited memory. SIAM J Sci Comput 21(1):24–45
Oliva A, Torralba A (2001) Modeling the shape of the scene: a holistic representation of the spatial envelope. Int J Comput Vis 42(3):145–175
Canny J (1986) A computational approach to edge detection. IEEE Trans Pattern Anal Mach Intell 8(6):679–698
Duda RO, Hart PE (1972) Use of the Hough transformation to detect lines and curves in pictures. Commun ACM 15(1):11–15
Yu Y, Mckelvey T, Kung S-Y (2013) A classification scheme for ’high-dimensional-small-sample-size’ data using SODA and ridge-SVM with microwave measurement applications. In: Proceedings of 2013 IEEE international conference on acoustics, speech, and signal processing (ICASSP 2013), Vancouver, British Columbia, Canada, 26–31 May 2013, pp 3542–3546
Sugiyama M (2006) Local Fisher discriminant analysis for supervised dimensionality reduction. In: Proceedings of 23rd international conference on machine learning, Pittsburgh, Pennsylvania, United States, 25–29 June 2006, pp 905–912
Sugiyama M (2007) Dimensionality reduction of multimodal labeled data by local Fisher discriminant analysis. J Mach Learn Res 8:1027–1061
Platt JC (1999) Fast training of support vector machines using sequential minimal optimization. In: Schölkopf B, Burges CJC, Smola AJ (eds) Advances in Kernel methods: support vector learning. MIT Press, Cambridge
Lab color space, Wikipedia, The Free Encyclopedia. http://en.wikipedia.org/wiki/Lab_color_space. Accessed 16 Dec 2014
Lowe DG (1999) Object recognition from local scale-invariant features. In: Proceedings of 7th IEEE international conference on computer vision, Kerkyra, Greece, 20–27 Sep 1999, pp 1150–1157
Bay H, Ess A, Tuytelaars T, Gool LV (2008) Speeded-up robust features (SURF). Comput Vis Image Underst 110(3):346–359
Delong A, Osokin A, Isack HN, Boykov Y (2011) Fast approximate energy minimization with label costs. Int J Comput Vis 96(1):1–27
Wang JZ, Li J, Wiederhold G (2001) SIMPLIcity: semantics-sensitive integrated matching for picture libraries. IEEE Trans Pattern Anal Mach Intell 23(9):47–63
Acknowledgments
This work was supported in part by the Ministry of Science and Technology, the Republic of China under Grant No. 103-2917-I-564-058. Part of this research is sponsored by the Beijing Key Laboratory of Mobile Computing and Pervasive Device, Institute of Computing Technology, Chinese Academy of Sciences, under the open project No. 2015-4. Furthermore, the Department of Electrical Engineering, Princeton University, supported this research during 2014/04-2015/05. No financial fund is given by Sungkyul University, South Korea. Dr. S. Rho does not participate in any creation and development of the proposed methods.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Chen, BW., He, X., Ji, W. et al. Large-scale image colorization based on divide-and-conquer support vector machines. J Supercomput 72, 2942–2961 (2016). https://doi.org/10.1007/s11227-015-1414-z
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11227-015-1414-z