A Hybrid Vision System for Dynamic Obstacle Detection

2021 ◽

Vol 14 (2) ◽

pp. 239-251

Author(s):

Hualei Zhang ◽

Mohammad Asif Ikbal

Keyword(s):

Feature Fusion ◽

Obstacle Detection ◽

Geometric Features ◽

Recognition Method ◽

Content Type ◽

Detection And Tracking ◽

Spatio Temporal ◽

Dynamic Obstacle ◽

Real Vehicle ◽

Temporal Feature

PurposeIn response to these shortcomings, this paper proposes a dynamic obstacle detection and tracking method based on multi-feature fusion and a dynamic obstacle recognition method based on spatio-temporal feature vectors.Design/methodology/approachThe existing dynamic obstacle detection and tracking methods based on geometric features have a high false detection rate. The recognition methods based on the geometric features and motion status of dynamic obstacles are greatly affected by distance and scanning angle, and cannot meet the requirements of real traffic scene applications.FindingsFirst, based on the geometric features of dynamic obstacles, the obstacles are considered The echo pulse width feature is used to improve the accuracy of obstacle detection and tracking; second, the space-time feature vector is constructed based on the time dimension and space dimension information of the obstacle, and then the support vector machine method is used to realize the recognition of dynamic obstacles to improve the obstacle The accuracy of object recognition. Finally, the accuracy and effectiveness of the proposed method are verified by real vehicle tests.Originality/valueThe paper proposes a dynamic obstacle detection and tracking method based on multi-feature fusion and a dynamic obstacle recognition method based on spatio-temporal feature vectors. The accuracy and effectiveness of the proposed method are verified by real vehicle tests.

Download Full-text

Virtual Reality Based Multi-Modal Teleoperation Using Mixed Autonomy

Volume 3: 28th Computers and Information in Engineering Conference, Parts A and B ◽

10.1115/detc2008-49144 ◽

2008 ◽

Author(s):

Muthukkumar S. Kadavasal ◽

Abhishek Seth ◽

James H. Oliver

Keyword(s):

Virtual Reality ◽

Stereo Vision ◽

Vision System ◽

Tracking System ◽

A Priori ◽

Obstacle Detection ◽

Remotely Operated Vehicle ◽

Autonomous Operation ◽

Control Interface ◽

Video Feed

A multi modal teleoperation interface is introduced featuring an integrated virtual reality based simulation augmented by sensors and image processing capabilities on-board the remotely operated vehicle. The proposed virtual reality interface fuses an existing VR model with live video feed and prediction states, thereby creating a multi modal control interface. Virtual reality addresses the typical limitations of video-based teleoperation caused by signal lag and limited field of view thereby allowing the operator to navigate in a continuous fashion. The vehicle incorporates an on-board computer and a stereo vision system to facilitate obstacle detection. A vehicle adaptation system with a priori risk maps and real state tracking system enables temporary autonomous operation of the vehicle for local navigation around obstacles and automatic re-establishment of the vehicle’s teleoperated state. As both the vehicle and the operator share absolute autonomy in stages, the operation is referred to as mixed autonomous. Finally, the system provides real time update of the virtual environment based on anomalies encountered by the vehicle. The system effectively balances the autonomy between human and on board vehicle intelligence. The stereo vision based obstacle avoidance system is initially implemented on video based teleoperation architecture and experimental results are presented. The VR based multi modal teleoperation interface is expected to be more adaptable and intuitive when compared to other interfaces.

Download Full-text

Vision-Based Moving Obstacle Detection and Tracking in Paddy Field Using Improved Yolov3 and Deep SORT

Sensors ◽

10.3390/s20154082 ◽

2020 ◽

Vol 20 (15) ◽

pp. 4082 ◽

Cited By ~ 2

Author(s):

Zhengjun Qiu ◽

Nan Zhao ◽

Lei Zhou ◽

Mengcen Wang ◽

Liangliang Yang ◽

...

Keyword(s):

Obstacle Avoidance ◽

Processing Speed ◽

Paddy Field ◽

Autonomous Navigation ◽

Vision System ◽

Obstacle Detection ◽

Paddy Fields ◽

Future Research ◽

Moving Obstacles ◽

Agricultural Machines

Using intelligent agricultural machines in paddy fields has received great attention. An obstacle avoidance system is required with the development of agricultural machines. In order to make the machines more intelligent, detecting and tracking obstacles, especially the moving obstacles in paddy fields, is the basis of obstacle avoidance. To achieve this goal, a red, green and blue (RGB) camera and a computer were used to build a machine vision system, mounted on a transplanter. A method that combined the improved You Only Look Once version 3 (Yolov3) and deep Simple Online and Realtime Tracking (deep SORT) was used to detect and track typical moving obstacles, and figure out the center point positions of the obstacles in paddy fields. The improved Yolov3 has 23 residual blocks and upsamples only once, and has new loss calculation functions. Results showed that the improved Yolov3 obtained mean intersection over union (mIoU) score of 0.779 and was 27.3% faster in processing speed than standard Yolov3 on a self-created test dataset of moving obstacles (human and water buffalo) in paddy fields. An acceptable performance for detecting and tracking could be obtained in a real paddy field test with an average processing speed of 5–7 frames per second (FPS), which satisfies actual work demands. In future research, the proposed system could support the intelligent agriculture machines more flexible in autonomous navigation.

Download Full-text

A Real-Time Stereo Vision System for Obstacle Detection and Recognition

10.13031/2013.42208 ◽

2012 ◽

Author(s):

Ta-Te Lin ◽

An-Chih Tsai ◽

Kai-Chiang Chuang ◽

Yu-Chou Chen ◽

Yu-Sung Chen

Keyword(s):

Real Time ◽

Stereo Vision ◽

Vision System ◽

Obstacle Detection ◽

Stereo Vision System ◽

Detection And Recognition

Download Full-text

An Road Obstacle Detection Method Based on Monocular Measurement

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.55-57.539 ◽

2011 ◽

Vol 55-57 ◽

pp. 539-544

Author(s):

Hong Jiao Jin ◽

Shen Lin ◽

Shi Guang Luo

Keyword(s):

Detection Method ◽

Vision System ◽

Measurement Model ◽

Obstacle Detection ◽

Monocular Vision ◽

Projection Model ◽

Vision Navigation ◽

Acceptable Range ◽

World Coordinate System ◽

The World

Obstacle detection in the intelligent vehicle vision navigation system occupies a very important role. The studies for the obstacles detecting, especially Monocular Measurement from the computer vision, simplifying monocular vision system to camera projection model. Getting the conversion relation between image coordinate and the world coordinate system through the geometry derivation to establish the measurement model and achieve the obstacle measurement. The experiment proved that the error of this measurement model selected is within the acceptable range.

Download Full-text