Wavelet neural network-based method of temperature fitting in unmanned-aerial-vehicle remote-sensing temperature-measurement

A technology of wavelet neural network and machine-based remote sensing, applied in neural learning methods, biological neural network models, etc., to achieve high precision, flexible and effective function approximation capabilities, and avoid blindness

Inactive Publication Date: 2018-08-28
大连云海创新科技有限公司
View PDF6 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0013] Aiming at the above-mentioned deficiencies in the prior art, the present invention proposes a method for temperature fitting in UAV remote sensing temperature measurement based on wavelet neural network, which not only makes full use of the time-frequency localization characteristics of wavelet transform but also has a more flexible and effective function Approximation ability and strong fault tolerance can effectively overcome some inherent defects of ordinary artificial neural network models

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
  • Wavelet neural network-based method of temperature fitting in unmanned-aerial-vehicle remote-sensing temperature-measurement
  • Wavelet neural network-based method of temperature fitting in unmanned-aerial-vehicle remote-sensing temperature-measurement
  • Wavelet neural network-based method of temperature fitting in unmanned-aerial-vehicle remote-sensing temperature-measurement

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0053] This embodiment provides a method for temperature fitting in UAV remote sensing temperature measurement based on wavelet neural network, including:

[0054] S1: Initialize the network; randomly initialize the translation parameters, scaling parameters, and network connection weights of the wavelet function and set the network learning rate, and put the initial value into the sample counter;

[0055] S2: Input the training samples and calculate the network output; divide the temperature values ​​measured in the UAV test into training samples and test samples, where the training samples are used to train the network, and the test samples are used to test the prediction accuracy of the network; during training, in The momentum item is added to the weight and threshold correction algorithm, and the correction value obtained in the previous step is used to smooth the learning path, avoid falling into the local minimum, and accelerate the learning speed. In order to avoid the...

Embodiment 2

[0065] In this embodiment, as a further supplement to Embodiment 1, the method for changing the weight value in step S4 is as follows:

[0066] A. Adjust the weight between the hidden layer and the output layer;

[0067]

[0068]

[0069] in Respectively represent the connection weights between the hidden layer node k and the output layer node n before adjustment and after adjustment; is the momentum item; is the expected output of the nth node in the output layer; is the actual output of the network; p is the number of samples;

[0070] B. Adjust the weight between the input layer node and the hidden layer node;

[0071]

[0072] in Respectively, the weights between the input layer node m and the hidden layer node k before adjustment and after adjustment; is the momentum item;

[0073] C. Adjust the scaling factor;

[0074]

[0075] in is the scaling factor before and after adjustment; is the momentum item of the expansion factor;

[0076] D. Ad...

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 provides a wavelet neural network-based method of temperature fitting in unmanned-aerial-vehicle remote-sensing temperature-measurement. The method includes: S1, initializing a network;S2, inputting training samples, and calculating network output; S3, judging whether training of training samples is all carried out, if yes, calculating an error function, and otherwise, jumping to the step S2; S4, judging whether the error function satisfies a requirement, if yes, jumping to S5, and otherwise, changing weights, and jumping to S2; S5, storing the weights; and S6, outputting a result. The time-frequency localization characteristic of wavelet transform is fully utilized, the method also has more flexible and effective function approximation ability and higher fault tolerance ability, and certain inherent defects of common artificial neural network models can be effectively overcome.

Description

technical field [0001] The invention belongs to the technical field of remote sensing temperature measurement of drones, in particular to a method for temperature fitting in remote sensing temperature measurement of drones based on a wavelet neural network. Background technique [0002] UAV remote sensing technology has developed rapidly in the field of measurement, and at the same time provides an effective survey method for temperature measurement in sea areas near nuclear power plants at home and abroad. To make UAV remote sensing technology an ideal means of temperature measurement, there are still several key technologies that need to be resolved, including the post-processing technology of remote sensing data. Because UAVs are affected by factors such as the atmosphere and wind speed in the process of sea surface temperature measurement, geometric and radiometric corrections should be made to the image according to the characteristics of its remote sensing image, calib...

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): G06N3/08
CPCG06N3/08
Inventor 伊晓东谭玥吴立志
Owner 大连云海创新科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap