Abstract:Due to the large variations of environment with ever-changing background and vehicles with different shapes, colors and appearances, to implement a real-time on-board vehicle recognition system with high adaptability, efficiency and robustness in complicated environments, remains challenging. This paper introduces a simultaneous detection and tracking framework for robust on-board vehicle recognition based on monocular vision technology. The framework utilizes a novel layered machine learning and particle filter to build a multi-vehicle detection and tracking system. In the vehicle detection stage, a layered machine learning method is presented, which combines coarse-search and fine-search to obtain the target using the AdaBoost-based training algorithm. The pavement segmentation method based on characteristic similarity is proposed to estimate the most likely pavement area. Efficiency and accuracy are enhanced by restricting vehicle detection within the downsized area of pavement. In vehicle tracking stage, a multi-objective tracking algorithm based on target state management and particle filter is proposed. The proposed system is evaluated by roadway video captured in a variety of traffics, illumination, and weather conditions. The evaluating results show that, under conditions of proper illumination and clear vehicle appearance, the proposed system achieves 91.2% detection rate and 2.6% false detection rate. Experiments compared to typical algorithms show that, the presented algorithm reduces the false detection rate nearly by half at the cost of decreasing 2.7%–8.6% detection rate. This paper proposes a multi-vehicle detection and tracking system, which is promising for implementation in an on-board vehicle recognition system with high precision, strong robustness and low computational cost.

A Rapid Multi-Angle Vehicle Detection Method in Complex Scenario

Research On Multi-Feature Front Vehicle Detection Algorithm Based On Video Image

A Multi-View Vehicle Detection Method Based on Deep Neural Networks

Embedded and Real-Time Vehicle Detection System for Challenging On-Road Scenes

Research on video vehicle detection based on AdaBoost classifiers of the ROI

A Multi-view 3D Vehicle Detection Method Based On Novel 3D Proposal Generation Method

Detection of Front-View Vehicle with Occlusions Using AdaBoost

Efficient Vehicle Detection and Optimization in Multi-Graph Mode Considering Multi-Section Tracking Based on Geographic Similarity

Vehicle Detection Algorithm Based on Future Fusion and Improved ObjectNess Binarized Normed Gradients Technology

AdaBoost-Bagging Vehicle Detection Algorithm Based on Multi-Mode Weak Classifier

Vehicle detection method based on deep learning and multi-layer feature fusion

Octave Convolution-Based Vehicle Detection Using Frame-Difference As Network Input

A Vehicle Detection Algorithm In Complex Traffic Scenes

Vehicles Detection In Complex Urban Traffic Scenes Using A Nonparametric Approach With Confidence Measurement

The AdaBoost Algorithm for Vehicle Detection Based on CNN Features

A Multi-Target Corner Pooling-Based Neural Network for Vehicle Detection

Real-time Vehicle Detection under Complex Road Conditions

A Method Based on Multi-Convolution Layers Joint and Generative Adversarial Networks for Vehicle Detection

Fast Vehicle Detection Based on Feature and Real-Time Prediction

Toward Fast and Accurate Vehicle Detection in Aerial Images Using Coupled Region-Based Convolutional Neural Networks

Simultaneous Multi-Vehicle Detection and Tracking Framework with Pavement Constraints Based on Machine Learning and Particle Filter Algorithm