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

Asteroid image synthesis method and system based on generative adversarial network, and computer readable storage medium

A technology for image synthesis and asteroids, applied in computing, image enhancement, image analysis, etc., can solve problems such as the lack of asteroid image data, achieve the effect of increasing data volume and reliability, and strengthening detailed terrain features

Pending Publication Date: 2021-12-31
视彻科技(杭州)有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Aiming at the problem of lack of real asteroid image data during the simulation of asteroid exploration missions, especially during the exercise contact sampling process, the present invention proposes an asteroid image synthesis method, system and computer-readable storage medium based on a generative countermeasure network, to realize Generate realistic synthetic images of asteroids with detailed surface textures under different environmental conditions based on a single real asteroid image

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
  • Asteroid image synthesis method and system based on generative adversarial network, and computer readable storage medium
  • Asteroid image synthesis method and system based on generative adversarial network, and computer readable storage medium
  • Asteroid image synthesis method and system based on generative adversarial network, and computer readable storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 2

[0127] This embodiment describes the implementation process of the present invention in conjunction with a specific asteroid image synthesis process,

[0128] image 3 Shown is a single real image of the Bennu asteroid (101955Bennu) used as an input image to generate a composite image of the asteroid with a custom silhouette shape and arbitrary lighting direction. Figure 9 Shown is a single Bennu surface rubble image used as input to generate a super-resolution texture-enhanced composite image of the asteroid's surface.

[0129] The workflow of this implementation is as follows: Figure 4 As shown, first use image 3 As input, a pyramidal multi-scale image segment generator in the SinGAN network structure is trained, and then the trained pyramidal generator is used to generate asteroid images with different characteristics, including custom silhouette shapes, with arbitrary illumination directions. then use Figure 9 As input, a new overfit pyramidal multiscale image seg...

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 an asteroid image synthesis method and system based on a generative adversarial network, and a computer readable storage medium. The method comprises the following steps of: performing training to obtain a pyramid-shaped multi-scale image fragment generator; specifying the shape of the asteroid, and inputting the contour image of the asteroid with the specified shape into the generator to output an asteroid composite image with the specified shape; rotating the contour image and the noise image of the asteroid with a specified shape by a specific angle, and acquiring an asteroid composite image with different illumination directions by using the generator; and adding reconstruction weight coefficient value training to obtain a new pyramid generator, carrying out repeated up-sampling on the generated image, and continuously injecting the generated image into the topmost generator in the new pyramid generator to obtain a super-resolution image. According to the method, the realistic asteroid composite image can be generated based on the single asteroid image, and detailed data support is provided for subsequent asteroid surface substance sampling simulation.

Description

technical field [0001] The present invention relates to the technical field of machine learning and image generation, and more specifically, relates to an asteroid image synthesis method, system and computer-readable storage medium based on a generative confrontation network. Background technique [0002] In the field of deep space exploration, asteroid sampling tasks are becoming a very important task. The primary concern of asteroid exploration missions is how to land safely. However, places with great research value are often blocked by rocks and craters. surrounded. Therefore, precise and safe landing and sampling must be performed automatically by the robot throughout the exploration mission. In order to deal with various complex situations that may appear on the asteroid, it is necessary to verify the reliability and effectiveness of the automatic control algorithm, autonomous navigation algorithm, and each functional module through simulation and detection exercises....

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): G06T5/50G06T3/40
CPCG06T5/50G06T3/4046G06T3/4053G06T2207/20081G06T2207/20084G06T2207/20221
Inventor 赵真非商慧强郭云东徐震杭
Owner 视彻科技(杭州)有限公司