House vectorization method for multi-element remote sensing data

A remote sensing data and vectorization technology, applied in the field of image processing, can solve the problems of lack of house post-processing method, low degree of automation, unsatisfactory house extraction results, etc., to achieve fast detection results, avoid trouble and adverse effects, and calculate the amount of less effect
CN106096497AActive Publication Date: 2016-11-09安徽省(水利部淮河水利委员会)水利科学研究院(安徽省水利工程质量检测中心站)
6 Cites 7 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
安徽省(水利部淮河水利委员会)水利科学研究院(安徽省水利工程质量检测中心站)
Publication Date
2016-11-09

Smart Images

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

Abstract

The invention discloses a house vectorization method for multi-element remote sensing data. The house vectorization method comprises the following steps of acquiring information which can represent house position and shape and is needed to be processed; obtaining an ordered edge point by eight neighborhood edge tracing, and simultaneously generating a binary image of a house edge, wherein the edge point is 1, and a non-edge point is 0; obtaining two main directions orthogonal to each other of a house by Radon conversion combined with main shaft analysis; partitioning an edge line of the house to three types of edge line segment, namely, a first main direction type, a second main direction type and a non-directional type; accurately positioning edge line segments of the first main direction type and the second main direction type; calculating a corner edge of adjacent edge line segments; and generating a vectorization house boundary line. By the house vectorization method, the troubles of initial point selection and sequence processing and unfavorable influence during the regularization process of the house can be prevented, the influence of a house extraction result error can also be reduced, and house profile information which is relatively consistent with the shape of the house and is effective and regulated can be acquired.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention relates to a building vectorization method for multivariate remote sensing data, belonging to the technical field of image processing. Background technique

[0002] As a major geospatial research information, houses (also known as buildings) are of great significance to the update of two-dimensional maps, the construction of smart cities, and the supervision of illegal buildings. The continuous development of remote sensing technology makes it possible to automatically vectorize and extract house information from remote sensing data.

[0003] Multivariate remote sensing data include: image data, LiDAR data, DEM, DSM and other data. Due to the interference of various factors, the house patches (edge ​​lines) currently extracted from multivariate remote sensing data often do not match the actual shape of the house, and it needs to be post-processed - denoising, regularization, etc.

[0004] The building outline regularization method discl...

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