Logo and Brand Recognition from Imbalanced Dataset Using MiniGoogLeNet and MiniVGGNet Models

Deep learning model tends to promote models with deep structure. Despite its high accuracy, the model was not practical when high computing power was not available. Thus, deep model with not-so-deep structure or less number of model parameters is needed for low capacity computer. Logo and brand recognition task is an important and challenging problem in computer vision with wide potential applications. The inherent challenge to address this task is not only due to the presence of logo in various direction and clutters as well as imbalanced dataset but also because of high computing workload when deep learning models were adopted. This paper presents empirical results of logo recognition method using MiniVGGNet and MiniGoogleNet models combined with augmentation technique to increase variation and number of samples. The results show that the proposed model combined with augmentation technique increased accuracy of model accuracies and fasten training convergence of both models.

[1]  Rainer Lienhart,et al.  Bundle min-hashing for logo recognition , 2013, ICMR '13.

[2]  Khalid Satori,et al.  Grayscale image encryption using shift bits operations , 2018, 2018 International Conference on Intelligent Systems and Computer Vision (ISCV).

[3]  Giovanni Russello,et al.  $2DCrypt$ : Image Scaling and Cropping in Encrypted Domains , 2016, IEEE Transactions on Information Forensics and Security.

[4]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[5]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[6]  Cordelia Schmid,et al.  Correlation-based burstiness for logo retrieval , 2012, ACM Multimedia.

[7]  Ross B. Girshick,et al.  Fast R-CNN , 2015, 1504.08083.

[8]  Zhuowen Tu,et al.  Aggregated Residual Transformations for Deep Neural Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[9]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[10]  Kaiming He,et al.  Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[11]  Rainer Lienhart,et al.  Scalable logo recognition in real-world images , 2011, ICMR.

[12]  Yee Whye Teh,et al.  A Fast Learning Algorithm for Deep Belief Nets , 2006, Neural Computation.

[13]  Nasharuddin Zainal,et al.  Grey scale image hiding method based on decomposition operation , 2013, 2013 IEEE Student Conference on Research and Developement.

[14]  Salah T. Allawi,et al.  Image encryption based on linear feedback shift register method , 2016, 2016 Al-Sadeq International Conference on Multidisciplinary in IT and Communication Science and Applications (AIC-MITCSA).

[15]  Shaogang Gong,et al.  Deep Learning Logo Detection with Data Expansion by Synthesising Context , 2016, 2017 IEEE Winter Conference on Applications of Computer Vision (WACV).

[16]  Arthur L. Samuel,et al.  Some studies in machine learning using the game of checkers , 2000, IBM J. Res. Dev..

[17]  Roberta Zambrini,et al.  Resolution in rotation measurements , 2005, quant-ph/0503224.

[18]  F. Makedon,et al.  High quality alias free image rotation , 1996, Conference Record of The Thirtieth Asilomar Conference on Signals, Systems and Computers.

[19]  Corneliu Florea,et al.  Local description using multi-scale complete rank transform for improved logo recognition , 2014, 2014 10th International Conference on Communications (COMM).

[20]  Yannis Avrithis,et al.  Scalable triangulation-based logo recognition , 2011, ICMR.

[21]  Cordelia Schmid,et al.  DeepMatching: Hierarchical Deformable Dense Matching , 2015, International Journal of Computer Vision.

[22]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[23]  Raimondo Schettini,et al.  Deep Learning for Logo Recognition , 2017, Neurocomputing.

[24]  Shaozi Li,et al.  Logo detection with extendibility and discrimination , 2013, Multimedia Tools and Applications.

[25]  Ali Farhadi,et al.  You Only Look Once: Unified, Real-Time Object Detection , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[26]  Jian Sun,et al.  Identity Mappings in Deep Residual Networks , 2016, ECCV.

[27]  Wei Liu,et al.  SSD: Single Shot MultiBox Detector , 2015, ECCV.