Real-Time Vehicle Make and Model Recognition System

Abstract

A Vehicle Make and Model Recognition (VMMR) system can provide great value in terms of vehicle monitoring and identification based on vehicle appearance in addition to the vehicles’ attached license plate typical recognition. A real-time VMMR system is an important component of many applications such as automatic vehicle surveillance, traffic management, driver assistance systems, traffic behavior analysis, and traffic monitoring, etc. A VMMR system has a unique set of challenges and issues. Few of the challenges are image acquisition, variations in illuminations and weather, occlusions, shadows, reflections, large variety of vehicles, inter-class and intra-class similarities, addition/deletion of vehicles’ models over time, etc. In this work, we present a unique and robust real-time VMMR system which can handle the challenges described above and recognize vehicles with high accuracy. We extract image features from vehicle images and create feature vectors to represent the dataset. We use two classification algorithms, Random Forest (RF) and Support Vector Machine (SVM), in our work. We use a realistic dataset to test and evaluate the proposed VMMR system. The vehicles’ images in the dataset reflect real-world situations. The proposed VMMR system recognizes vehicles on the basis of make, model, and generation (manufacturing years) while the existing VMMR systems can only identify the make and model. Comparison with existing VMMR research demonstrates superior performance of the proposed system in terms of recognition accuracy and processing speed.

A Review on Vehicle Classification and Potential Use of Smart Vehicle-Assisted Techniques

“…In the feature extraction step, suitable features for the classification of vehicles are selected. Texture features and shape features, including Scale-Invariant Feature Transform (SIFT) [119,120], Oriented Fast and Rotated Brief (ORB) [121,122], Speeded-Up Robust Features (SURF) [123,124], and Vehicle Make and Model Recognition (VMMR) [25,125], are among the most common features used for VC. Texture features are used to overcome the disadvantages of color and intensity features.…”

Vehicle Make and Model Recognition using Bag of Expressions

“…The work proposed in [42] does not provides satisfactory results on the NTOU-MMR dataset. The performance and computational cost of [25,33] on the NTOU-MMR dataset are acceptable but far behind from the proposed approach in terms of average accuracy and processing speed. Therefore, it can be concluded from Table 6 that the BoE-based VMMR procedure is superior in the context of performance and computational cost.…”

“…Jie Fang et al [31] and Afshin Dehghan et al [32] describe a convolutional neural network (CNN) for the classification of a vehicle's make and model. Random forest and SVM are suggested for recognition of VMM in [33]. Indeed, SVM-based classification schemes have been applied in different VMMR systems such as [3,13,20,25].…”

Gerrymandering the National Security Narrative: A Case Study of the Canadian Security Intelligence Service’s Handling of its Bulk Metadata Exploitation Program

“…This is especially true since 9/11, as the "war on terror" came to unambiguously elevate national security imperatives to the top of Western policy agendas, oftentimes to the detriment of ideals promoted by other, somewhat antithetical, discourses, such as those defending democratic rights and freedoms (Ericson 2007: 36-71;Laurin 2018: 117). In Canada, this national security turn has, in part, meant that intelligence agencies have greatly extended their reach (Bell 2006)that is, they have been targeting and monitoring an ever-growing amount of people, places, social categories, events, and practices in an ever-growing number of ways (e.g., MacMahon 2005;Molnar and Parsons 2016;Walby et al 2016;Bennett et al 2014: 151-168;Manzoor 2018;Lyon 2014). Among these agencies is CSIS.…”