Automatic Car Make Recognition in Low-Quality Images

Badura, Paweł; Skotnicka, Maria

doi:10.1007/978-3-319-06593-9_21

Cited by 6 publications

(4 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, an attribute could be chosen in more than one subset. All such feature subsets of class i (i.e., H g ij ), are collected in H g i as shown in Equation (11) where the dimensions of H g i are (N i ) × (S ss )…”

Section: ) Motivation To Use Attribute Bagging (Ab)mentioning

confidence: 99%

See 1 more Smart Citation

Real-Time Vehicle Make and Model Recognition Based on a Bag of SURF Features

Siddiqui

Mammeri

Boukerche

2016

IEEE Trans. Intell. Transport. Syst.

View full text Add to dashboard Cite

In this paper, we propose and evaluate unexplored approaches for real-time automated vehicle make and model recognition (VMMR) based on a bag of speeded-up robust features (BoSURF) and demonstrate the suitability of these approaches for vehicle identification systems. The proposed approaches use SURF features of vehicles' front-or rear-facing images and retain the dominant characteristic features (codewords) in a dictionary. Two schemes of dictionary building are evaluated: "single dictionary" and "modular dictionary." Based on the optimized dictionaries, the SURF features of vehicles' front-or rear-face images are embedded into BoSURF histograms, which are used to train multiclass support vector machines (SVMs) for classification. Two real-time VMMR classification schemes are proposed and evaluated: a single multiclass SVM and an ensemble of multiclass SVM based on attribute bagging. The processing speed and accuracy of the VMMR system are affected greatly by the size of the dictionary. The tradeoff between speed and accuracy is studied to determine optimal dictionary sizes for the VMMR problem. The effectiveness of our approaches is demonstrated through cross-validation tests on a recent publicly accessible VMMR data set. The experimental results prove the superiority of our work over the state of the art, in terms of both processing speed and accuracy, making it highly applicable to real-time VMMR systems.

show abstract

Section: ) Motivation To Use Attribute Bagging (Ab)mentioning

confidence: 99%

“…For example, [4] and [11] employed a simple bruteforce matching scheme using raw SIFT features to match query images to the gallery images. The brute-force pattern matching approach is very time consuming, and hence unsuitable for real-time VMMR.…”

Section: Classification Approachesmentioning

confidence: 99%

Real-Time Vehicle Make and Model Recognition Based on a Bag of SURF Features

Siddiqui

Mammeri

Boukerche

2016

IEEE Trans. Intell. Transport. Syst.

View full text Add to dashboard Cite

show abstract

“…From the ROI region, local color features are extracted and classified using a multi-class SVM to recognize the color. Authors in [12] proposed a robust system for car make recognition from car front images even in presence of low contrast and compression based distortions. Car brand region is segmented and SIFT features are extracted from the brand region.…”

Section: Literature Surveymentioning

confidence: 99%

Traffic Adaptive Deep Learning based Fine Grained Vehicle Categorization in Cluttered Traffic Videos

S¹,

Deepu²

2021

IJACSA

View full text Add to dashboard Cite

Smart traffic management is being proposed for better management of traffic infrastructure and regulate traffic in smart cities. With surge of traffic density in many cities, smart traffic management becomes utmost necessity. Vehicle categorization, traffic density estimation and vehicle tracking are some of the important functionalities in smart traffic management. Vehicles must be categorized based on multiple levels like type, speed, direction of travel and vehicle attributes like color etc. for efficient tracking and traffic density estimation. Vehicle categorization becomes very challenging due to occlusions, cluttered backgrounds and traffic density variations. In this work, a traffic adaptive multi-level vehicle categorization using deep learning is proposed. The solution is designed to solve the problems in vehicle categorization in terms of occlusions, cluttered backgrounds.

show abstract

“…This paper is an extension of [26] and [27], where the preliminary studies results for vehicle make and colour detection have been presented. In this study the image processing stage for reliable car brand pattern detection has been revised and improved mainly by the redevelopment of the car brand segmentation and classification algorithms.…”

Section: Figure 1 -Configuration Of Data Acquisition Devicesmentioning

confidence: 99%

System for Detecting Vehicle Features from Low Quality Data

Bugdoł

Badura

Juszczyk

et al. 2018

PROMET

View full text Add to dashboard Cite

The paper presents a system that recognizes the make, colour and type of the vehicle. The classification has been performed using low quality data from real-traffic measurement devices. For detecting vehicles’ specific features three methods have been developed. They employ several image and signal recognition techniques, e.g. Mamdani Fuzzy Inference System for colour recognition or Scale Invariant Features Transform for make identification. The obtained results are very promising, especially because only on-site equipment, not dedicated for such application, has been employed. In case of car type, the proposed system has better performance than commonly used inductive loops. Extensive information about the vehicle can be used in many fields of Intelligent Transport Systems, especially for traffic supervision.

show abstract

Automatic Car Make Recognition in Low-Quality Images

Cited by 6 publications

References 17 publications

Real-Time Vehicle Make and Model Recognition Based on a Bag of SURF Features

Real-Time Vehicle Make and Model Recognition Based on a Bag of SURF Features

Traffic Adaptive Deep Learning based Fine Grained Vehicle Categorization in Cluttered Traffic Videos

System for Detecting Vehicle Features from Low Quality Data

Contact Info

Product

Resources

About