Three-dimensional collision avoidance sonar obstacle detection method and system based on deep learning
An obstacle detection and deep learning technology, which is applied in neural learning methods, radio wave measurement systems, measurement devices, etc., can solve the problems of inability to extract deep-level features of sonar targets, low signal-to-noise ratio, and poor target detection effects. To achieve the effect of easy implementation, rapid detection and identification, simple detection method and system
Pending Publication Date: 2022-05-10
宜昌测试技术研究所
View PDF0 Cites 2 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, a large number of research and experiment results show that the underwater sonar target image has many beams and low signal-to-noise ratio, and the traditional image processing method or acoustic signal processing method cannot extract the deep-level features of the sonar target, which leads to the incompatibility of underwater sonar targets. The object detection effect in the environment is not good
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0062] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field of the present invention; the terms used in the specification herein are only for the purpose of describing specific embodiments, and are not intended to To limit the invention, for example, the terms "length", "width", "upper", "lower", "left", "right", "front", "rear", "vertical", "horizontal", "top ", "bottom", "inner", "outer" and other indicated orientations or positions are based on the orientations or positions shown in the drawings, which are only for convenience of description and cannot be understood as limitations on the technical solution.
[0063] The terms "include" and "have" in the specification and claims of the present invention and the description of the above drawings, as well as any variations thereof, are intended to cover a non-exclusive inclusion; The terms "first", "second", etc. are used t...
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
Login to View More Abstract
The invention relates to the technical field of unmanned vehicles, and discloses a three-dimensional collision avoidance sonar obstacle detection method and system based on deep learning, and the method comprises the steps: detecting an obstacle through a three-dimensional collision avoidance sonar, and collecting the three-dimensional point cloud data of the obstacle; processing the three-dimensional point cloud data to generate a two-dimensional fan-shaped image, and performing obstacle labeling to obtain a training image sample; training the training image sample through a deep learning model, carrying out image detection and recognition on the obstacle, obtaining the position information of the obstacle, and obtaining a small image of the obstacle; carrying out image segmentation based on a self-adaptive dual-threshold detection algorithm, obtaining obstacle contour information, and obtaining an average gray value of obstacle imaging; performing three-dimensional point cloud data rasterization processing to generate a three-dimensional raster image, and obtaining three-dimensional boundary data of the obstacle; and performing collision avoidance control on the obstacle according to the three-dimensional boundary data. Accurate obstacle three-dimensional boundary information can be obtained, and the method is integrally simple, reliable and easy to implement.
Description
technical field [0001] The present invention relates to the technical field of unmanned aerial vehicle control, and more specifically, to a three-dimensional collision avoidance sonar obstacle detection method and system based on deep learning. Background technique [0002] For the detection and recognition of sonar images, traditional statistical machine learning methods are mostly used for research. However, a large number of research and experiment results show that the underwater sonar target image has many beams and low signal-to-noise ratio, and the traditional image processing method or acoustic signal processing method cannot extract the deep-level features of the sonar target, which leads to the incompatibility of underwater sonar targets. The target detection effect in the environment is not good. [0003] In addition, since underwater target detection is generally used in the field of national defense and security, not only accurate detection of underwater target...
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
Login to View More IPC IPC(8): G06T7/136G06T7/187G06T7/90G06N3/08G06K9/62G01S15/93G01S15/89G01S7/539G06V10/774G06V10/82
CPCG06T7/136G06T7/90G06T7/187G01S15/93G01S15/89G01S7/539G06N3/08G06T2207/10028G06T2207/20081G06T2207/20084G06T2207/20132G06F18/214
Inventor 徐从营杨邦清曾盎
Owner 宜昌测试技术研究所