Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Deep-sea seabed obstacle measurement system and identification method based on deep learning

An obstacle recognition and deep learning technology, applied in neural learning methods, measurement devices, character and pattern recognition, etc., can solve problems such as inability to recognize obstacles, difficult to measure the actual feature information of obstacles, etc., to enhance environmental perception Capability, broad engineering application value, the effect of improving flexibility

Active Publication Date: 2021-08-31
HUNAN UNIV OF SCI & TECH
View PDF10 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the shapes and surface textures of deep-sea submarine obstacles are different, and the conventional pattern recognition method cannot realize the identification of obstacles, and it is also difficult to measure the actual characteristic information of obstacles.

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
  • Deep-sea seabed obstacle measurement system and identification method based on deep learning
  • Deep-sea seabed obstacle measurement system and identification method based on deep learning
  • Deep-sea seabed obstacle measurement system and identification method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0045] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0046] Such as figure 1 As shown, a deep-sea submarine obstacle measurement system based on deep learning, including a calibration module, a data preprocessing module, a data acquisition module, a data processing module, a data fusion module and a measurement result module;

[0047] The data acquisition module includes a camera and a laser ranging sensor. The camera is used to collect seabed topographic images, and the laser ranging sensor is used to measure the distance between the camera and obstacles;

[0048] The calibration module is used for camera calibration to obtain internal and external parameters of the camera;

[0049] The data preprocessing module collects the image data collected by the camera and the distance data measured by the laser ranging sensor and performs preprocessing, trains the model offline, evaluates the trained model, adju...

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 discloses a deep-sea seabed obstacle identification method based on deep learning, and the method comprises the following steps: calibrating a camera, and obtaining the internal and external parameters of the camera; collecting and preprocessing an image data set, and carrying out mold offline training and evaluating; collecting a submarine topographic image, and measuring the distance from the camera to an obstacle; processing the collected image based on a deep learning semantic segmentation technology, and obtaining feature parameters of the obstacle; fusing the internal and external parameters obtained through calibration and the measured distance parameters through a data fusion module, converting the characteristic parameters of the obstacle into actual parameters of the obstacle, and therefore basic information of the obstacle is obtained. According to the identification method, the deep learning technology, camera calibration, laser ranging and other methods are fused to obtain the actual characteristic parameters of the obstacle, so that the mobile deep sea sampling equipment can carry out driving trafficability analysis, automatic obstacle avoidance and path planning research.

Description

technical field [0001] The invention relates to a deep-sea submarine obstacle measurement system and identification method based on deep learning. Background technique [0002] The ocean contains extremely rich mineral resources such as polymetallic nodules, cobalt-rich crusts, and polymetallic sulfides. According to current preliminary surveys, 15% of the deep sea area contains polymetallic nodule resources, with a total reserve of about 3 trillion tons. It mainly occurs on the surface of the seabed at a water depth of 3000m to 6000m; the seabed of about 6.35 million square kilometers (accounting for 1.7% of the seabed area) is covered by cobalt-rich crusts. Based on this calculation, the total amount of cobalt is about 1 billion tons, mainly Distributed on the seabed at a water depth of about 400m to 4000m, with a maximum thickness of about 24cm; polymetallic sulfides (also known as hydrothermal sulfides) are heavy metal mineral resources in the seabed that have attracted ...

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 Applications(China)
IPC IPC(8): G06T7/80G06K9/32G06K9/62G06N3/04G06N3/08G01B11/02
CPCG06T7/85G06N3/08G01B11/02G06T2207/20081G06T2207/20084G06T2207/20104G06T2207/30244G06V10/25G06N3/045G06F18/214
Inventor 宁宇金永平彭佑多何术东颜健
Owner HUNAN UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products