Unlock instant, AI-driven research and patent intelligence for your innovation.

Mine obstacle recognition method, system and equipment based on image processing

An obstacle recognition and image processing technology, applied in image data processing, image enhancement, scene recognition and other directions, can solve the problems of large memory occupation, large parameters, large model calculation amount, etc., to ensure real-time performance, low power consumption, low cost effect

Active Publication Date: 2021-10-01
安徽海博智能科技有限责任公司
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In the prior art, when deep learning is used for obstacle detection, the model has a large amount of calculation, large parameters, and large memory usage. However, embedded computing resources are limited, resulting in poor real-time performance of obstacle recognition during unmanned driving control of mining vehicles.

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
  • Mine obstacle recognition method, system and equipment based on image processing

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0026] A preferred embodiment of the present invention will be described in detail below with reference to the accompanying drawings.

[0027] like figure 1 As shown, a mine obstacle recognition method based on image processing, the steps are as follows:

[0028] S1: Mark the historical images containing obstacles to generate a training set, and input the training set into the deep learning model to generate a recognition model.

[0029] Specifically, the obstacles include road potholes, roadside guardrails, ponding water, stones, unloading guardrails, unloading warehouses, communication poles, reflective ground, gravels, excavators, mounds, pedestrians, and mine cars; Objects can be divided into dynamic obstacles and static obstacles.

[0030] Collect about 100,000 image samples in the mining area for model training. The GPU configuration of the deep learning server is 2*RTX2080Super, mainly using Pytorch1.3.1 as the framework.

[0031] S2: Input the real-time image contai...

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 mine obstacle recognition method based on image processing. The steps are as follows: mark historical images containing obstacles to generate a training set, input the training set into a deep learning model to generate a recognition model; Input the real-time image into the recognition model to obtain obstacle information, and generate obstacle features through obstacle information; compare the obstacle features of the same obstacle at different times, obtain the real-time position of the obstacle and generate the obstacle's running track.

Description

technical field [0001] The invention relates to the field of unmanned driving of mining vehicles, in particular to a mine obstacle recognition method, system and equipment based on image processing. Background technique [0002] The mine road conditions are complex, but the number of personnel is small, which is suitable for the implementation of unmanned driving technology. [0003] In the existing technology, deep learning has a large amount of model calculation, large parameters, and large memory usage when doing obstacle detection. However, embedded computing resources are limited, resulting in poor real-time performance of obstacle recognition during unmanned driving control of mining vehicles. Contents of the invention [0004] In order to solve the above technical problems, the present invention provides a mine obstacle recognition method, system and equipment based on image processing. [0005] In order to solve the problems of the technologies described above, th...

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): G06K9/00G06K9/32G06K9/62G06T7/246G06N3/04
CPCG06T7/246G06T2207/20081G06T2207/20084G06T2207/30241G06T2207/30252G06V20/58G06V10/25G06N3/045G06F18/22G06F18/25G06F18/214
Inventor 郑恩涛吴旭宾秦晓驹王亚飞
Owner 安徽海博智能科技有限责任公司