Data Augmentation and Clustering for Vehicle Make/Model Classification

Vehicle shape information is very important in Intelligent Traffic Systems (ITS). In this paper we present a way to exploit a training data set of vehicles released in different years and captured under different perspectives. Also the efficacy of clustering to enhance the make/model classification is presented. Both steps led to improved classification results and a greater robustness. Deeper convolutional neural network based on ResNet architecture has been designed for the training of the vehicle make/model classification. The unequal class distribution of training data produces an a priori probability. Its elimination, obtained by removing of the bias and through hard normalization of the centroids in the classification layer, improves the classification results. A developed application has been used to test the vehicle re-identification on video data manually based on make/model and color classification. This work was partially funded under the grant.

[1]  Jun-Wei Hsieh,et al.  Symmetrical SURF and Its Applications to Vehicle Detection and Vehicle Make and Model Recognition , 2014, IEEE Transactions on Intelligent Transportation Systems.

[2]  James M. Ferryman,et al.  Vehicle subtype, make and model classification from side profile video , 2015, 2015 12th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS).

[3]  Meena Abdelmaseeh,et al.  Car Make and Model recognition combining global and local cues , 2012, Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012).

[4]  Nick Pears,et al.  Automatic make and model recognition from frontal images of cars , 2011, 2011 8th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS).

[5]  Afshin Dehghan,et al.  View Independent Vehicle Make, Model and Color Recognition Using Convolutional Neural Network , 2017, ArXiv.

[6]  Jonathan Krause,et al.  3D Object Representations for Fine-Grained Categorization , 2013, 2013 IEEE International Conference on Computer Vision Workshops.

[7]  Tobias Glasmachers,et al.  Vehicle Shape and Color Classification Using Convolutional Neural Network , 2019, ArXiv.

[8]  Ahmet Emir Dirik,et al.  Deep Learning Based Vehicle Make-Model Classification , 2018, ICANN.

[9]  Adam Herout,et al.  BoxCars: 3D Boxes as CNN Input for Improved Fine-Grained Vehicle Recognition , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[10]  Ignacio Parra,et al.  Vehicle model recognition using geometry and appearance of car emblems from rear view images , 2014, 17th International IEEE Conference on Intelligent Transportation Systems (ITSC).

[11]  Timothy F. Cootes,et al.  Analysis of Features for Rigid Structure Vehicle Type Recognition , 2004, BMVC.

[12]  Richard Szeliski,et al.  Car make and model recognition using 3D curve alignment , 2014, IEEE Winter Conference on Applications of Computer Vision.

[13]  Michael G. Madden,et al.  Multi-Class and Single-Class Classification Approaches to Vehicle Model Recognition from Images , 2005 .

[14]  Eleftherios Kayafas,et al.  Vehicle model recognition from frontal view image measurements , 2011, Comput. Stand. Interfaces.

[15]  Maurice Milgram,et al.  Multi-class Vehicle Type Recognition System , 2008, ANNPR.

[16]  Gérard G. Medioni,et al.  3-D model based vehicle recognition , 2009, 2009 Workshop on Applications of Computer Vision (WACV).