Vehicle Detection and Classification: A Review

Kiran, V. Keerthi; Parida, Priyadarsan; Dash, Sonali

doi:10.1007/978-3-030-49339-4_6

Cited by 12 publications

(6 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The detection and classification of vehicles remain one of the ITS problems not yet resolved because of several problems: including mainly the acquisition conditions related to sensors (cameras, electromagnetic loop, radar, optical fiber…, etc.) and the variable external environment in atmospheric conditions and lighting [28] Researchers in the field of automatic road traffic monitoring have conducted preliminary research on detecting and classifying vehicles from a video stream issued by cameras placed on red highway lights or onboard vehicles in traffic [29]. Most recent research has adopted deep networks, thus allowing systems to learn from the external environment, which is often complex and variable according to climatic conditions.…”

Section: Related Workmentioning

confidence: 99%

Master-Slave Convolutional Deep Architecture for Vehicle Identification and Type Classification

Ennehar,

Samra

2023

View full text Add to dashboard Cite

Video surveillance of road traffic plays an important role in highway safety and is an important application of intelligent transportation systems. One of the basic applications of intelligent transport systems is the detection and classification of vehicle types. The major problems encountered by these systems are the significant similarity between the vehicles, frequent occlusions on the highway, and low resolution of the surveillance cameras. This paper proposes a novel convolutional neural network architecture called master-slave convolutional deep architecture for vehicle detection and type classification. The basic concept of this architecture is twofold: a. The sequential operation of the two networks where the slave network only works if the master network detects vehicles in the road scene will allow a considerable reduction in the search area for vehicles. It will induce a significant reduction in processing time. b. A combination of deep-shallow neural networks allows the system to share the knowledge gained from the vehicles on two networks. The first (master) shallow learns the shape of vehicles while the second (slave) is responsible for learning all the details of vehicles to distinguish the different classes. The experimental results, performed on 3200 images, have shown that the favorable performance of the proposed CNN architecture allowed us to achieve successful detection with TPR of 92% and TNR of 95% and vehicle type classification with a considerable mean average precision of 93.38% where cars classification gives the highest rate (98.63%).

show abstract

Section: Related Workmentioning

confidence: 99%

Master-Slave Convolutional Deep Architecture for Vehicle Identification and Type Classification

Ennehar,

Samra

2023

View full text Add to dashboard Cite

show abstract

“…The frame difference gives good results of vehicle detection in the dynamic background, but the noise appears in the outcomes. The optical flow can provide a high accuracy of 98.60% in vehicle detection at the speed of 0.212 cm/s of the vehicle but fails in front-view vehicle detection [23].…”

Section: Categorymentioning

confidence: 99%

Shadow Extraction and Elimination of Moving Vehicles for Tracking Vehicles

Jadav,

Sorathiya,

El-Shafai

et al. 2023

CMC

View full text Add to dashboard Cite

Shadow extraction and elimination is essential for intelligent transportation systems (ITS) in vehicle tracking application. The shadow is the source of error for vehicle detection, which causes misclassification of vehicles and a high false alarm rate in the research of vehicle counting, vehicle detection, vehicle tracking, and classification. Most of the existing research is on shadow extraction of moving vehicles in high intensity and on standard datasets, but the process of extracting shadows from moving vehicles in low light of real scenes is difficult. The real scenes of vehicles dataset are generated by self on the Vadodara-Mumbai highway during periods of poor illumination for shadow extraction of moving vehicles to address the above problem. This paper offers a robust shadow extraction of moving vehicles and its elimination for vehicle tracking. The method is distributed into two phases: In the first phase, we extract foreground regions using a mixture of Gaussian model, and then in the second phase, with the help of the Gamma correction, intensity ratio, negative transformation, and a combination of Gaussian filters, we locate and remove the shadow region from the foreground areas. Compared to the outcomes proposed method with outcomes of an existing method, the suggested method achieves an average true negative rate of above 90%, a shadow detection rate SDR (η%), and a shadow discrimination rate SDR (ξ %) of 80%. Hence, the suggested method is more appropriate for moving shadow detection in real scenes.

show abstract

“…The kernels are applied separately to the input image to produce gradient components measurements separately The image is then convolved with the filter. After that horizontal and vertical kernels of the operator are convolved with the original image [10]. the gradient calculated by Sobel operator is given below:…”

Section: A Edge Detection Algorithmsmentioning

confidence: 99%

Hardware Synthesize and Performance Analysis of Intelligent Transportation Using Canny Edge Detection Algorithm

Baloch¹,

Memon²,

Memon³

et al. 2021

IJEM

View full text Add to dashboard Cite

The World is moving toward Smart traffic management and monitoring technologies. Vehicle detection and classification are the two important features of intelligent transportation system. Several algorithms for detection of vehicles such as Sobel, Prewitt, and Robert etc. but due to their less accuracy and sensitivity to noise they could not detect vehicles clearly. In this paper, a simple and rapid prototyping approach for vehicle detection and classification using MATLAB Xilinx system generator and Zedboard is presented. The Simulink model of vehicle detection and classification is designed using a complex canny edge detection algorithm for vehicle detection. The canny edge detection algorithm offers 91% accuracy as compared to its counterpart Sobel and Perwitt algorithms that offer 79.4% and 76.1% accuracy. The feature vector approach is used for vehicle classification. The proposed model is simulated and validated in MATLAB. The Canny edge detection and feature vector algorithms for vehicle detection and classification are synthesized through the Xilinx system generator in Zedboard. The proposed design is validated with the existing works. The implementation results reveal that the proposed system for vehicle detection and classification takes only 8 ns of execution time with a 128MHz clock, which is the lowest and optimum calculation period for the smart city.

show abstract

Vehicle Detection and Classification: A Review

Cited by 12 publications

References 29 publications

Master-Slave Convolutional Deep Architecture for Vehicle Identification and Type Classification

Master-Slave Convolutional Deep Architecture for Vehicle Identification and Type Classification

Shadow Extraction and Elimination of Moving Vehicles for Tracking Vehicles

Hardware Synthesize and Performance Analysis of Intelligent Transportation Using Canny Edge Detection Algorithm

Contact Info

Product

Resources

About