Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Real time binocular vision guidance method facing to underwater carrying vehicle

An underwater vehicle and binocular vision technology, which is applied to instruments, line-of-sight measurement, surveying and navigation, etc., can solve problems such as inaccurate distance measurement and increased matching error

Inactive Publication Date: 2009-10-28
HARBIN ENG UNIV
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

When there is a local low-texture area, the matching error in this area increases, resulting in inaccurate ranging

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Real time binocular vision guidance method facing to underwater carrying vehicle
  • Real time binocular vision guidance method facing to underwater carrying vehicle
  • Real time binocular vision guidance method facing to underwater carrying vehicle

Examples

Experimental program
Comparison scheme
Effect test

example 1

[0080]Experiments were performed using the widely used Tsukuba synthetic stereo image pairs. The Tsukuba stereo image pair has been rectified and has a size of 384×288. Its real disparity map is shown in Figure 6(b). In the experiment, the pyramid series k=4, the zoom ratio is r=2, the matching window is 15×15, the NCC threshold is 0.5, the parallax range is (0, 40), and the texture threshold is 0.1. On the Pentium 42.40GHz industrial computer, the stereo matching is calculated The parallax process took an average of 164.5ms. The NCC value can be used to measure the matching accuracy. The obtained disparity map and NCC map are shown in Fig. 7 .

example 2

[0082] Experiments are carried out in a simulated autonomous underwater vehicle (AUV for short) motion environment. The resolution of the two industrial digital CCD cameras is 1280×1024. On the basis of accurate calibration, first correct the captured image pair, and then adopt the above method, set the pyramid level k=4, the scaling ratio is r=2, the matching window is 21×21, the NCC threshold is 0.85, and the parallax range is (- 230, -200), Texture Threshold 15. The average time spent on obtaining the disparity map by running the program on the same industrial computer is only 258ms, and the accuracy of the obtained depth map can reach millimeter level. See Figure 8 for the corrected image pair, and Figure 9 for the disparity map and NCC map. It can be seen from Figure 9 that the pyramid NCC algorithm based on texture control effectively extracts the depth of obstacles from the simulated ocean environment. The calculation time of 258ms proves that this method has practica...

example 3

[0084] Using this binocular vision guidance system, a simulation test of AUV space obstacle avoidance was carried out in the laboratory. The experiment was carried out on a four-degree-of-freedom AUV simulated motion platform. The platform is driven by four high-precision stepping motors, which can well simulate the advance and retreat, lateral movement, heave and head turn of the AUV in the marine environment. The entire obstacle avoidance simulation system is composed of three parts: binocular guidance system, obstacle avoidance control system and executive mechanism - AUV simulation motion platform and peripheral equipment.

[0085] as attached Figure 10 As shown, the binocular camera is fixed on the front end of the AUV motion model, and a spherical obstacle and a square obstacle with a flat surface are placed on its walking path. The spherical obstacle is located to the left of the AUV's initial position, and the square obstacle is located to the right of the AUV. Whe...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a real-time binocular vision guidance method for underwater vehicles. Aiming at the characteristics of the marine environment where the underwater vehicle moves, a pyramid normalized covariance binocular vision algorithm based on texture control is proposed, which achieves real-time performance at the application level, and the accuracy can be stabilized at the centimeter level. For the final representation of environmental information, a virtual sonar model is proposed to represent the 2.5-dimensional environmental information as an obstacle (or target) matrix based on the virtual sonar model, which includes the depth and depth of the obstacle (or target) orientation information.

Description

(1) Technical field [0001] The invention relates to a method capable of providing azimuth and distance information of targets or obstacles for underwater vehicles in real time. (2) Background technology [0002] With the development of various underwater vehicle technologies, sonar has become the most widely used sensing technology in underwater depth detection and obstacle detection. However, sonar has certain limitations in short-range applications, including low accuracy and blind spots. The superior performance of visual sensing technology at close range can just make up for this deficiency. It can provide high-resolution and high-precision (especially in short range) depth information for underwater vehicles. In recent years, the application of vision technology in tasks such as obstacle avoidance, target recognition, dock docking, cable maintenance, and submarine terrain modeling has become a research hotspot. In 1999, Negahdaripour and others at the University of M...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G01C3/00G01C11/00G06K9/64
Inventor 施小成王晓娟边信黔唐照东刘和祥
Owner HARBIN ENG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Eureka Blog
Learn More
PatSnap group products