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

A Feasible Region Detection Method Based on Machine Learning

A technology of area detection and machine learning, which is applied to instruments, computer parts, image data processing, etc., can solve the problem of detection of feasible areas of unmanned platforms

Active Publication Date: 2018-04-17
ZHEJIANG UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The technical problem to be solved by the present invention is to provide a feasible region detection method based on machine learning, which mainly solves the problem of feasible region detection of unmanned platforms

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
  • A Feasible Region Detection Method Based on Machine Learning
  • A Feasible Region Detection Method Based on Machine Learning
  • A Feasible Region Detection Method Based on Machine Learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0049] The present invention uses the unmanned platform provided with camera equipment as a carrier, and implements through the following steps:

[0050] A method for detecting a feasible region based on machine learning, wherein the method uses an unmanned platform provided with camera equipment as a carrier to detect a feasible region, comprising the steps of:

[0051] Step 1. Define the value w, which represents the image width value in the image, and the number of matrix columns in the matrix; define the value h, represents the image height in the image, and represents the number of matrix rows in the matrix; set the image width to w, and the image height For h, construct an image coordinate system; determine five point coordinates according to the height and angle of the unmanned platform, including (x1, y1), (x2, 0), (w, y3), (x4, y4), (x5, h ); define FM as a matrix with h rows and w columns, and each element value is only +1, 0 or -1, and each element of the matrix is ...

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 technical problem to be solved by the present invention is to provide a feasible region detection method based on machine learning, including: determining parameters, constructing features, classifying features, updating training data sets and updating classifiers. This method iterates cyclically between the above steps, and can extract the feasible region of the entire image by continuously expanding a small feasible region. This method is effective in real time in complex changing scenarios. Based on this method, the outdoor unmanned platform can extract the feasible area in the current field of view, so as to ensure its own safe and stable operation.

Description

technical field [0001] The invention relates to a method for extracting a feasible region of an unmanned platform, in particular to a method for realizing adaptive detection of a feasible region through a visual sensor and a machine learning method. Background technique [0002] Vision-based feasible region detection has made great progress with the improvement of computing unit performance. However, the current method has disadvantages such as poor scene adaptability, poor light adaptability and weak stability, which bring hidden dangers to the stable and safe operation of the unmanned platform. For example, some feasible region extraction methods first require that the feasible region to be detected has a vanishing point, but the situation that there is no vanishing point in the feasible region often occurs when the robot is actually running, for example, in the intersection scene. Even on the straight road, the vanishing point may not appear in the field of view due to t...

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): G06K9/00G06K9/62G06T7/11
Inventor 刘勇张清泉薛睿
Owner ZHEJIANG UNIV