A method for bicycle detection using ellipse approximation

Fujimoto, Yasuyuki; Hayashi, Jun-ichiro

doi:10.1109/fcv.2013.6485499

Cited by 6 publications

(5 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(4) . ( rank accuracy (ha ) H k = argmm hi +rankD i tterence ( hi , F k -1) (5) where ranka ccuracy(h i) is the rank of measure score in (3), (d) rank D i tterence (h i , F k -1) is the difference between hi and the selected part set F k -1, and H k is the kth selected part. By combining the two rank functions, a final detector that takes the advantage of accuracy and difference is acquired to im prove the efficiency.…”

Section: E Part Selectionmentioning

confidence: 99%

See 1 more Smart Citation

Integrating appearance and edge features for on-road bicycle and motorcycle detection in the nighttime

Chen

Lin²,

Wu³

et al. 2014

17th International IEEE Conference on Intelligent Transportation Systems (ITSC)

View full text Add to dashboard Cite

It is critical to detect bicycles and motorcycles on the road because collision of autos with those light vehicles becomes major cause of on-road accidents nowadays especially in the nighttime. Therefore, a vision-based nighttime bicycle and motorcycle detection method relying on use of a camera and near-infrared lighting mounted on an auto vehicle is pro posed in this paper. Generally, the foreground objects in front of the auto, not the far-away background, will reflect near-infrared lighting in the nighttime environments. However, some components of the bicycles and the motorcycles absorb most infrared lighting and thus make the bicycles and motor cycles hardly recognizable. To cope with this problem, the aforementioned detection method is part-based, which com bines the two kinds of features related to the characteristics of bicycles and motorcycles. Also, the information about the geo metric relation among all the parts and the object centroid is learned off-line. Due to high computation load, Adaboost algo rithm is used to select effective parts with better geometric in formation for detection. To validate the proposed results, sev eral experiments are conducted to show that the developed system is reliable in detecting bicycles and motorcycles in the nighttime.

show abstract

Section: E Part Selectionmentioning

confidence: 99%

“…In [5,6], the authors pro posed a bicycle detection method by detecting two wheels of a bicycle. Instead of wheel detection, Chiu et al [7] further used rider's helmets to detect motorcycles.…”

Section: Related Workmentioning

confidence: 99%

Integrating appearance and edge features for on-road bicycle and motorcycle detection in the nighttime

Chen

Lin²,

Wu³

et al. 2014

17th International IEEE Conference on Intelligent Transportation Systems (ITSC)

View full text Add to dashboard Cite

show abstract

“…The vision-based bicycle detection has been a hot research topic in last decade. The feature of the bicycles can be employed in the detection procedure, e.g., ellipse approximation [2], edge features of the bicycle [3], and deformable part model [4]. Alternatively some approaches employ the features of the cyclists, such as MSC-HOG feature [5] and RealAdaBoost [6].…”

Section: B Related Workmentioning

confidence: 99%

AutoBiVision - A Novel Vision-Based Bicycle Warning System

Liu¹,

Zhao²

2016

JTLE

View full text Add to dashboard Cite

This article presents the concept of a novel visionbased bicycle warning system (AutoBiVision). AutoBiVision system warns the driver of the approaching bicycles without offering redundant or disturbing information, which means when the driver is focusing on one specific area, the system will focus on the other areas that the driver is not paying attention to. To this end both the driver and cyclists require to be monitored, which means two camera systems are needed: The inner camera system captures the interior of the vehicle while the outer camera system captures the bicycle lane as well as the cyclists. With the purpose of intelligently detecting and tracking the bicyclist in wider ranges, two rotatable cameras, which are components of the outer camera system, are mounted on the top of the left and right back mirrors of the vehicle, respectively. The angle control of the rotatable cameras depends on the results of bicycle tracking and bicycle lane detection. In this contribution, the main structure of the system and the initial results of a series of novel methods of bicycle lane detection, are presented. A patent is pending.  Index Terms-bicycle warning system; bicycle lane detection; rotatable cameras Jie Zhao was born in Shandong, PR China on May 4, 1989. He obtained the B.Eng degree in School of Information and Science, Shandong University, in Jinan in 2012, majoring in communication engineering. In 2011 he worked as an INTERN at Hisense Group in Qingdao, China. He published several papers in the field of vision-based driver monitoring. From 2012 he became a M.Sc. degree candidate at the University of Bremen, Bremen, Germany, majoring in information and automation engineering. His paper "Robust Profile Face Detection and Rotation Angle Estimation of the Driver's Head in a Novel Dazzling Avoidance System" proposes a novel way to estimate the yaw rotating angle of the driver's head, which was published by IEEE on the 2015 IEEE Intelligent Vehicles Symposium in Seoul, Korea. His previous research focused on wireless communication and currently his research field is computer vision and robotics. Mr. Zhao is a student member of IEEE. He obtained the "Full-Mark Master Project" award at University of Bremen in 2015.

show abstract

“…Ref. [12] converts the bicycle detection problem into a tire detection problem, using a convergent floating threshold method to detect tires.…”

Section: Introductionmentioning

confidence: 99%

An accurate shared bicycle detection network based on faster R‐CNN

Wang

Yang

et al. 2023

IET Image Processing

View full text Add to dashboard Cite

Detecting shared bicycles is an essential and challenging task. Deep learning has been widely used in object detection tasks in urban scenes, such as vehicle detection. However, deep learning algorithms still face many difficulties and challenges in shared bicycle detection. For example, the problem of large deformation of shared bicycles and the problem of small targets because the camera is far away from the shared bicycles. In order to solve these problems, this study introduces the feature fusion module and deformable convolution into the object detection network, which improves the efficiency of shared bicycle detection. This study proposes an enhanced faster R‐CNN network (A classic two‐stage object detection network) for shared bicycle detection and a shared bicycle dataset (SBD) is constructed for model training and testing. Compared with the original faster R‐CNN, the mean average precision (mAP) of the enhanced method on SBD is improved by 13%, which indicates that the method provided in this study is more suitable for detecting shared bicycles. This study also conducts experiments on the Microsoft Common Objects (COCO) dataset, where this method achieves 40.2% of the mAP, which is 5.8% higher than faster R‐CNN before improvement.

show abstract

A method for bicycle detection using ellipse approximation

Abstract: We proposes a method for bicycle detection using ellipse approximation. In this research, we focus on the area of bicycle tire.

Cited by 6 publications

References 1 publication

Integrating appearance and edge features for on-road bicycle and motorcycle detection in the nighttime

Integrating appearance and edge features for on-road bicycle and motorcycle detection in the nighttime

AutoBiVision - A Novel Vision-Based Bicycle Warning System

An accurate shared bicycle detection network based on faster R‐CNN

Contact Info

Product

Resources

About