Computer vision based obstacle detection and target tracking for autonomous vehicles

Fang, Ruoyu; Cai, Chengtao

doi:10.1051/matecconf/202133607004

Cited by 13 publications

(3 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Among them, Reference [ 32 ] introduced the AlexNet network to further judge the recognition results of YOLO, which effectively improves the system recognition efficiency. Reference [ 34 ], on the other hand, implements the control of driverless vehicles based on the tracking detection results of the YOLOv3 network for pedestrians. In terms of pavement disease detection, a YOLOX-based pavement disease detection method was proposed in Reference [ 35 ], which effectively solves the problem of slow identification by traditional methods.…”

Section: Introductionmentioning

confidence: 99%

Real-Time Target Detection System for Intelligent Vehicles Based on Multi-Source Data Fusion

Zou

Zheng

Wang

2023

Sensors

View full text Add to dashboard Cite

To improve the identification accuracy of target detection for intelligent vehicles, a real-time target detection system based on the multi-source fusion method is proposed. Based on the ROS melodic software development environment and the NVIDIA Xavier hardware development platform, this system integrates sensing devices such as millimeter-wave radar and camera, and it can realize functions such as real-time target detection and tracking. At first, the image data can be processed by the You Only Look Once v5 network, which can increase the speed and accuracy of identification; secondly, the millimeter-wave radar data are processed to provide a more accurate distance and velocity of the targets. Meanwhile, in order to improve the accuracy of the system, the sensor fusion method is used. The radar point cloud is projected onto the image, then through space-time synchronization, region of interest (ROI) identification, and data association, the target-tracking information is presented. At last, field tests of the system are conducted, the results of which indicate that the system has a more accurate recognition effect and scene adaptation ability in complex scenes.

show abstract

Section: Introductionmentioning

confidence: 99%

Real-Time Target Detection System for Intelligent Vehicles Based on Multi-Source Data Fusion

Zou

Zheng

Wang

2023

Sensors

View full text Add to dashboard Cite

show abstract

“…The evolution of deep learning and computer hardware has helped computer vision applications become reality. Some disciplines that use DL for computer vision tasks are robotics [ 1 ], image quality assessment [ 2 ], biometrics [ 3 ], face recognition [ 4 ], image classification [ 5 ], autonomous vehicles [ 6 ], etc. One of the most important applications in CV is medical image analysis, where usually DL models were trained to diagnose or predict several diseases from numerous modalities such as MRI, CT-scans, X-rays, Histopathology images, etc.…”

Section: Introductionmentioning

confidence: 99%

Digital Watermarking as an Adversarial Attack on Medical Image Analysis with Deep Learning

Apostolidis

Papakostas

2022

J. Imaging

View full text Add to dashboard Cite

In the past years, Deep Neural Networks (DNNs) have become popular in many disciplines such as Computer Vision (CV), and the evolution of hardware has helped researchers to develop many powerful Deep Learning (DL) models to deal with several problems. One of the most important challenges in the CV area is Medical Image Analysis. However, adversarial attacks have proven to be an important threat to vision systems by significantly reducing the performance of the models. This paper brings to light a different side of digital watermarking, as a potential black-box adversarial attack. In this context, apart from proposing a new category of adversarial attacks named watermarking attacks, we highlighted a significant problem, as the massive use of watermarks, for security reasons, seems to pose significant risks to vision systems. For this purpose, a moment-based local image watermarking method is implemented on three modalities, Magnetic Resonance Images (MRI), Computed Tomography (CT-scans), and X-ray images. The introduced methodology was tested on three state-of-the art CV models, DenseNet 201, DenseNet169, and MobileNetV2. The results revealed that the proposed attack achieved over 50% degradation of the model’s performance in terms of accuracy. Additionally, MobileNetV2 was the most vulnerable model and the modality with the biggest reduction was CT-scans.

show abstract

“…Literature [13] introduces the method of moving multitarget detection and tracking in stable scene. Literature [14] realizes target tracking and real-time obstacle detection of autonomous vehicles based on computer vision. Literature [15] develops a new m-sequence target and circular correlation processing technology based on computer vision for real-time displacement measurement.…”

Section: Introductionmentioning

confidence: 99%

Research on Multitarget Recognition and Detection Based on Computer Vision

Zhang

Chen

2022

Journal of Sensors

View full text Add to dashboard Cite

Information is obtained from human eyes for thinking divergence, and further associated with computer equipment, so human beings endow computers with the ability of “vision” to convey and feel information. This field has developed for many years, and many aspects can be in line with other research directions, such as artificial intelligence, which has become popular in recent years, and pattern recognition, which has been applied a lot. In order to sort out the structure and content of multitarget recognition smoothly, this paper starts from the perspective of shallow vision, uses theory and practical experiments, and chooses the core technology with the largest weight from massive computer technologies, so that the recognition algorithm can compare with the recognition algorithm. The research shows that (1) CNN shows its unique feature ability and incomparable detection accuracy from many models, and the error rate can be reduced from 28.07% to 18.40%. (2) The method of candidate region is complex, and the larger the region, the more difficult it is to calculate. The method based on regression is far beyond it in both precision and speed and is more suitable for the research of this subject. (3) When the mAP increases, the speed is forced to slow down. If the image resolution is high with the same model, the mAP will be high (SSD and YOLO models are often used). Experiments show that the recognition effect is obvious. At the end of the article, the advantages and disadvantages of this study are summarized. In the field of computer vision, people need to do more in-depth research. Follow-up can optimize multitarget recognition and detection and strive to improve the accuracy.

show abstract

Computer vision based obstacle detection and target tracking for autonomous vehicles

Cited by 13 publications

References 5 publications

Real-Time Target Detection System for Intelligent Vehicles Based on Multi-Source Data Fusion

Real-Time Target Detection System for Intelligent Vehicles Based on Multi-Source Data Fusion

Digital Watermarking as an Adversarial Attack on Medical Image Analysis with Deep Learning

Research on Multitarget Recognition and Detection Based on Computer Vision

Contact Info

Product

Resources

About