Optical building target three-dimensional reconstruction method based on deep neural network

A deep neural network and three-dimensional reconstruction technology, applied in neural learning methods, biological neural network models, neural architectures, etc., to improve accuracy and ensure quality
CN113160392AActive Publication Date: 2021-07-23电子科技大学成都学院
4 Cites 3 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
电子科技大学成都学院
Publication Date
2021-07-23

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention provides an optical building target three-dimensional reconstruction method based on a deep neural network, and the method comprises the steps of constructing an optical image building data set through marking, and constructing the deep neural network for extracting an optical building; preprocessing an input image to enhance the quality of an optical image, segmenting the processed optical image by using a local recursion method based on an OTSU criterion and image entropy, and training a data set of the processed optical image to enable a deep neural network to have better extraction capability; using a method for quickly simulating optical target light-target three-dimensional point cloud to establish a data set through SENSOR simulation software, and using a 3D-R2N2 network to perform centralized learning on obtained optical target image three-dimensional data, so that a two-dimensional image is mapped to a three-dimensional model, an instance of an object is obtained in an end-to-end manner, a three-dimensional reconstruction result of the optical image is obtained, and a good value is provided for the development of the building field.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to the technical field of three-dimensional reconstruction, in particular to a method for three-dimensional reconstruction of optical architectural objects based on a deep neural network. Background technique

[0002] With the development of modern technology, architectural 3D reconstruction technology has attracted more and more attention in today's society. People use computers to convert 2D images into 3D data. 3D data refers to the generated 3D images stored in the form of digital models in the computer However, traditional architecture can no longer meet people's needs, and 3D architecture is playing an increasingly important role in neighborhoods or scenes such as urban planning, intelligent buildings, disaster monitoring, and digital cities. Up to now, in-depth research has been carried out at home and abroad on the three-dimensional feature signal detection, feature signal matching, and camera feature calibration of buildi...

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