Tyzx DeepSea High Speed Stereo Vision System

Author(s):  
J.I. Woodfill ◽  
G. Gordon ◽  
R. Buck
2021 ◽  
Author(s):  
Jamin Islam

For the purpose of autonomous satellite grasping, a high-speed, low-cost stereo vision system is required with high accuracy. This type of system must be able to detect an object and estimate its range. Hardware solutions are often chosen over software solutions, which tend to be too slow for high frame-rate applications. Designs utilizing field programmable gate arrays (FPGAs) provide flexibility and are cost effective versus solutions that provide similar performance (i.e., Application Specific Integrated Circuits). This thesis presents the architecture and implementation of a high frame-rate stereo vision system based on an FPGA platform. The system acquires stereo images, performs stereo rectification and generates disparity estimates at frame-rates close to 100 fpSi and on a large-enough FPGA, it can process 200 fps. The implementation presents novelties in performance and in the choice of the algorithm implemented. It achieves superior performance to existing systems that estimate scene depth. Furthermore, it demonstrates equivalent accuracy to software implementations of the dynamic programming maximum likelihood stereo correspondence algorithm.


2011 ◽  
Vol 5 (3) ◽  
pp. 445-451
Author(s):  
Makoto Yamashita ◽  
◽  
Noriaki Maru ◽  

This article presents the development of a low-cost, high-speed stereo vision system for embedded use employing a mass-produced CMOS imager and a commercially available DSP. We use the system to acquire an image of 96×96 pixels in a period of 1 ms. We embed a read region movement control and a target track function into the system and successfully track a target point within a period of 3 ms. The difference between this high-speed stereo vision system and traditional ones lies in that this system is intended to be used even in a small-scale control system using a microcomputer. We limit the capacity of the image memory in which acquired pixel data is temporarily stored and performed real time processing. As a result, we eliminate frame memory and thus simplify its corresponding circuit. This allows the system to be reduced in size, power consumption, and cost, thereby making the system appropriate for embedded devices, batterypowered robots, and the like.


Author(s):  
Sukjune Yoon ◽  
Chun-Kyu Woo ◽  
Hyun Do Choi ◽  
Sung-Kee Park ◽  
Sung-Chul Kang ◽  
...  

The purpose of this project is to develop a mobile robot for hazardous terrain exploration. The exploration of hazardous terrain requires the development of a passive mechanism adaptable to such terrain and a sensing system for obstacle avoidance, as well as a remote control. We designed a new mobile robot, the Ronahz 6-wheel robot, which uses a passive mechanism that can adapt to hazardous terrains and building stairways without any active control. The suggested passive linkage mechanism consists of a simple four-bar linkage mechanism. In addition, we install a stereo vision system for obstacle avoidance, as well as a remote control. Wide dynamic range CCD cameras are used for outdoor navigation. A stereo vision system commonly requires high computational power. Therefore, we use a new high-speed stereo correspondence algorithm, triangulation, and iterative closest point (ICP) registration to reduce computation time. Disparity maps computed by a newly proposed, high-speed method are sent to the operator by a wireless LAN equipment. At the remote control site, a three-dimensional digital map around a mobile robot is built by ICP registration and reconstruction process, and this three-dimensional map is displayed for the operator. This process allows the operator to sense the environment around the robot and to give commands to the mobile robot when the robot is in a remote site.


2019 ◽  
Vol 68 (8) ◽  
pp. 2742-2754 ◽  
Author(s):  
Shaobo Wang ◽  
Yuan Xu ◽  
Yanghao Zheng ◽  
Mingcheng Zhu ◽  
Haodong Yao ◽  
...  

2014 ◽  
Vol 224 ◽  
pp. 145-150
Author(s):  
Piotr Garbacz ◽  
Piotr Czajka ◽  
Bartłomiej Burski

The article presents a method for monitoring the destruction process of materials under mechanical load on a universal testing machine. Observations of the subjects are made using a high-speed camera and a catadioptric stereo-vision system. The camera allows for data acquisition with the capture speed more than a million frames per second (FPS) with reduced resolution. Catadioptric vision systems use mirrors and lenses in order to modify the observation path. In the proposed system four mirrors are required to divide the observation path into two separate paths. This enables monitoring the test specimens from different perspectives, which provides a number of advantages including information redundancy or stereovision. In order to verify the proposed method several metal specimens were put under mechanical load and monitored with the vision system. Test results are enclosed in the article. Selected registered images presenting the moment of the destruction are described in greater detail with data about the capture speed provided. The tests were conducted under front and back lighting in order to assess the best method of illumination.


2021 ◽  
Author(s):  
Jamin Islam

For the purpose of autonomous satellite grasping, a high-speed, low-cost stereo vision system is required with high accuracy. This type of system must be able to detect an object and estimate its range. Hardware solutions are often chosen over software solutions, which tend to be too slow for high frame-rate applications. Designs utilizing field programmable gate arrays (FPGAs) provide flexibility and are cost effective versus solutions that provide similar performance (i.e., Application Specific Integrated Circuits). This thesis presents the architecture and implementation of a high frame-rate stereo vision system based on an FPGA platform. The system acquires stereo images, performs stereo rectification and generates disparity estimates at frame-rates close to 100 fpSi and on a large-enough FPGA, it can process 200 fps. The implementation presents novelties in performance and in the choice of the algorithm implemented. It achieves superior performance to existing systems that estimate scene depth. Furthermore, it demonstrates equivalent accuracy to software implementations of the dynamic programming maximum likelihood stereo correspondence algorithm.


Sign in / Sign up

Export Citation Format

Share Document