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

Image-based place recognition method

A technology for identifying methods and locations, applied in the field of robotics, can solve problems such as resource consumption and slow computing speed, and achieve fast training speed, small model size, and guaranteed performance

Inactive Publication Date: 2020-07-17
安徽果力智能科技股份有限公司
View PDF4 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Using the CNN model can significantly improve the efficiency of feature extraction, but the introduction of deep neural networks inevitably leads to problems such as slow operation speed and resource consumption.

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
  • Image-based place recognition method
  • Image-based place recognition method
  • Image-based place recognition method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0034] The present invention will be further described below in conjunction with examples, but the protection scope of the present invention is not limited thereto.

[0035] like figure 1 Shown, the present invention comprises the following steps:

[0036] Step 1. Collect a large number of scene images to obtain a series of images right Perform preliminary feature extraction to get The set of feature vectors, that is, the sample set right Mark and get the corresponding category label in, for N i dimensional row vector, for N o dimensional row vector, N i is the dimension of preliminary feature extraction, N o is the number of place categories, so the category set is if sample x k of the category in the c-th category, then y k The cth element of is 1, and the rest are 0; Represents the real number field, k is a positive integer from 1 to n, n is the total number of samples, N o with N i Also represent the number of neurons in the output layer and in...

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 relates to an image-based place recognition method, which comprises the following steps of: 1) acquiring an image and preliminarily extracting features; 2) initialization is carried out;3) solving a feature extraction weight matrix; 4) transforming the feature space; 5) randomly generating an input weight vector and an input bias of a hidden layer mapping function; 6) generating a hidden layer output function; 7) generating a hidden layer output matrix; 8) initializing an output weight matrix; according to the method, only four layers of artificial neural networks are adopted, compared with a deep neural network, the method has the advantages that the model size is smaller, the training speed is higher. Meanwhile, a feature extraction layer is introduced to guarantee the feature extraction performance, and the classification accuracy is improved.

Description

technical field [0001] The invention belongs to the technical field of robots, in particular to an image-based location recognition method. Background technique [0002] At present, with the development of artificial intelligence, mobile intelligent robots are widely used in industrial, military and service fields, and are playing an increasingly important role. Therefore, higher and higher requirements are put forward for the ability of intelligent robots to recognize the environment. A robot can only move safely and efficiently if it knows its own position and working environment. The vision system can provide the robot with the richest perceptual information, and it is also the closest to the way humans perceive the environment. In recent years, the problem of vision-based robot self-localization has attracted a large number of researchers, and has also achieved fruitful research results. For this kind of problem, it is often called the location classification problem....

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): G06K9/00G06K9/62G06N3/04
CPCG06V20/00G06N3/045G06F18/2431
Inventor 刘阳刘珂
Owner 安徽果力智能科技股份有限公司