Optimal selection of underwater gravity adaptation area based on convolution slope eigenvalue support vector machine

Through the support vector machine method based on the convolutional slope eigenvalue, the problem of underwater gravity field distribution and adaptability evaluation is solved, and the distinction between gravity adaptation zone and non-adaptation zone is achieved, and the underwater gravity is improved. Match navigation accuracy and positioning accuracy.

CN114372498BActive Publication Date: 2025-02-07CHINA ACADEMY OF SPACE TECHNOLOGY
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202111143171.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-09-28
Publication Date
2025-02-07
Estimated Expiration
2041-09-28

AI Technical Summary

Technical Problem

The prior art is difficult to accurately evaluate the advantages and disadvantages of underwater gravity field distribution and adaptability, which affects the accuracy of gravity matching navigation.

Method used

Using the support vector machine method based on the convolutional slope eigenvalue, the gravity anomaly graph is convolutional and single-line slope solution, and combined with the characteristic parameters of the convolutional feature graph and the gravity anomaly graph, a classification discriminant function is constructed to distinguish gravity adaptation. and non-adaptive areas.

Benefits of technology

The effective distinction between gravity adaptation zones and non-adaptation zones is achieved, and the accuracy and positioning accuracy of underwater gravity matching navigation are improved, with a matching accuracy of more than 90%.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114372498B_ABST
    Figure CN114372498B_ABST
Patent Text Reader

Abstract

The present invention discloses a method for optimizing underwater gravity adaptation zone based on convolution slope characteristic value support vector machine, comprising: using Sobel operator to perform convolution operation on gravity anomaly map to obtain convolution characteristic map; performing single-line slope solution according to convolution characteristic map and gravity anomaly map to determine convolution slope parameter; determining convolution row-column difference and convolution variance of convolution characteristic map, pooling difference and range of gravity anomaly map; performing standardization processing on convolution slope parameter, convolution row-column difference, convolution variance pooling difference and range to obtain standardization processing result; constructing a classification discriminant function based on support vector machine algorithm according to the standardization processing result; and discriminating the area to be identified according to the classification discriminant function to determine the gravity adaptation zone and non-adaptation zone of the area to be identified. The present invention realizes the effective distinction between gravity adaptation zone and non-adaptation zone.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Image Semantic Segmentation Method Based on Deep Full Convolutional Network and Conditional Random Field

    AU2020103901A4

  • Spatial layering disturbance gravitational field grid model rapid construction method

    CN104834320A