After vehicle detection and vehicle type recognition, it is vehicle make and model recognition (VMMR) that has attracted researchers attention in the last decade. Due to the large number of classes and small inner-class distance, this problem is known as a hard classification problem.
In this paper, a comparison between holistic and part-based approaches has been made and most of the previous methods in each category have been reviewed. In addition, a new part-based method is proposed which tries to overcome some of the hard challenges in this area. This method operates on meaningful parts of vehicle like lights, grilles and logo for distinguishing of different classes. The Histogram of Oriented Gradients (HOG) and Support Vector Machine (SVM) have been used for feature extraction and classification tasks respectively. For evaluation purposes, a dataset including 352 images from frontal and rear view of eight different classes of vehicles have been prepared and fully marked based on their parts. The experimental results show the effectiveness of the proposed part-based approach in comparison to the traditional approaches. The proposed method achieved 95.4% accuracy on frontal view and 100% accuracy on rear view images.
Biglari, M., & Soleimani, A. (2021). A Comparison between Holistic and Part-Based Approaches and Proposing a New Approach for Vehicle Make and Model Recognition. Soft Computing Journal, 5(2), 18-27.
MLA
Mohsen Biglari; Ali Soleimani. "A Comparison between Holistic and Part-Based Approaches and Proposing a New Approach for Vehicle Make and Model Recognition". Soft Computing Journal, 5, 2, 2021, 18-27.
HARVARD
Biglari, M., Soleimani, A. (2021). 'A Comparison between Holistic and Part-Based Approaches and Proposing a New Approach for Vehicle Make and Model Recognition', Soft Computing Journal, 5(2), pp. 18-27.
VANCOUVER
Biglari, M., Soleimani, A. A Comparison between Holistic and Part-Based Approaches and Proposing a New Approach for Vehicle Make and Model Recognition. Soft Computing Journal, 2021; 5(2): 18-27.
)
