An Empirical Analysis of Deep Learning Architectures for Vehicle Make and Model Recognition

Abstract

Vehicle make and model recognition is an important component of the Intelligent Transport System (ITS), which plays an essential role in vehicle surveillance and traffic monitoring. In this paper, we evaluated the performance of the recent deep neural networks for vehicle recognition to identify 196 different types of vehicles based on their make, model, and year using Stanford Cars data. Transfer learning has been employed to reduce the training time and also added a dropout of 0.5 at the last dense layer before the output layer for improved generalization performance on an applied dataset. Different augmentation methods such as random erasing, horizontal flip, random crop, resize, rotate, shear, and sharpen were applied to the training dataset to increase the number of images. Besides, optimization methods namely mixup and cyclical learning rate were also investigated to train these deep neural networks to avoid overfitting. Finally, the proposed ensemble learning framework using k-fold cross-validation has resulted in the improved accuracy of image classification for the vehicle make and model recognition. The final achieved accuracy is 93.96% through ensembling homogeneous models of DenseNet201, which is higher than previous stateof-the-art approaches for the vehicle make and model recognition. Simultaneously, it has also improved the performance of other models where MobileNetV2 has shown the highest improvement with an increase of 2.8% in the overall accuracy using the proposed ensemble learning technique.INDEX TERMS deep neural networks, vehicle make and model recognition, ensemble learning, and k-fold cross-validation

Vehicle make and model recognition using mixed sample data augmentation techniques

“…The existing studies are either based on manual features extraction [3] or multiple ensemble models [9] resulted in reduced performance during inference. The proposed solution is robust during inference but has some limitations during training.…”

“…Their work is based on emergency vehicle type classification and had images of fire trucks, police cars, ambulances and standard cars [8]. Hassan et al compared different classifiers with cyclic learning rate and used the MixUp image augmentation technique to achieve an accuracy of 93.96% through ensembling homogeneous models of DenseNet201 [9]. Though the CNN-based model has gained much attention in recent years, manual feature-based classification is still being studied recently.…”

Vehicle images dataset for make and model recognition

“…We conducted the experiment on the same test, train split as it is given by the dataset, for validation data we further split training data to 90% training and 10% validation. We load all the models with the ImageNet [9] weights and used the transfer learning technique which reduce the training time of the models and train the models on 50 epochs initially [1] , [6] . We used accuracy as a performance parameter and initial results are given below [5] .…”

Vehicle Make and Model Recognition as an Open-Set Recognition Problem and New Class Discovery

“…The VMMRdb database [37] (available at https://github.com/faezetta/VMMRdb, accessed on 13 March 2023) was used in this work since it is one of the most cited in the specialized literature [12,38,39]. Only eight classes from the VMMRdb database were used and were manually filtered to retain only unduplicated images showing the rear view of the vehicles (i.e., samples of each class were not balanced).…”