Search

CN-122024062-A - Street elevation micro-update identification method based on grid network

CN122024062ACN 122024062 ACN122024062 ACN 122024062ACN-122024062-A

Abstract

The invention discloses a street elevation micro-updating identification method based on a grid network, which comprises the steps of determining an acquisition range, grading a road network, arranging sampling points, acquiring street view images, carrying out gridding treatment on a building elevation, identifying new construction or dismantling of the building through two granularity, identifying materials of the building elevation, carrying out micro-updating judgment and the like. According to the invention, the material change of the building elevation is automatically identified through the street view image, so that the urban micro-update is efficiently and accurately monitored, and the method is suitable for the fields of urban planning, update evaluation and the like.

Inventors

  • ZHANG JUNXUE
  • ZHU TAO
  • CAO JUN
  • ZHAI XIAOTING

Assignees

  • 江苏科技大学

Dates

Publication Date
20260512
Application Date
20260202

Claims (10)

  1. 1. The street elevation micro-update identification method based on the grid network is characterized by comprising the following steps of: Step 1, determining a target area, and acquiring road network data and street view image data of two different years of the area; step 2, grading the obtained road network, and arranging sampling points; step 3, respectively acquiring street view images of each sampling point at two years and multiple view angles, and identifying and associating building elevation; step 4, rasterizing the continuous building elevation image with different granularities; Step 5, judging new construction and dismantling based on the storage loss of the building elevation; And 6, identifying the grid material, comparing the grid material with the change, and judging micro-update according to the change degree.
  2. 2. The grid network street facade based micro-update identification method according to claim 1, wherein obtaining road network data comprises extracting road vector data of two years of target areas from open street map data, and performing simplification processing to exclude expressways, bridges and non-public traffic roads.
  3. 3. The grid network-based street elevation micro-update identification method according to claim 1, wherein in the step 2, the road network is classified into a main road and a secondary main road, sampling points are distributed on the main road at intervals of 50 meters, and sampling points are distributed on the secondary main road at intervals of 30 meters.
  4. 4. The grid network street elevation micro-update identification method according to claim 1, wherein in the step 3, street view images of each sampling point in two years and multiple views are obtained, the street view images of each sampling point in two years and multiple views comprise front, rear, left and right directions along the road direction, and building elevation is identified and associated.
  5. 5. The grid network street facade micro-update identification method according to claim 1, wherein building facades are identified by using a pre-trained semantic segmentation model and associated marks are carried out.
  6. 6. The method for identifying the street facade micro-update based on the grid network according to claim 1, wherein the specific step of performing two kinds of granularity rasterization processing on the continuous building facade image in the step 4 comprises the following steps: step 4-1, dividing the building elevation image into rectangular grid units which are identical in size and are regularly arranged for subsequent analysis and comparison of material levels; step 4-2, dividing a rectangular grid unit of the building elevation image into two layers of a large-granularity grid with the granularity of 5 multiplied by 5m and a small-granularity grid with the granularity of 1 multiplied by 1 m; And 4-3, identifying new construction and dismantling of the large-granularity grids and identifying materials of the vertical surfaces of the small-granularity grids.
  7. 7. The grid network street elevation based micro-update identification method according to claim 1, wherein the step 5 of determining new construction and demolition based on the storage loss of the building elevation comprises the following specific steps: Step 5-1, judging that an image grid with the same sampling point and view angle is newly built if no effective building elevation is identified in the previous year image and is identified in the subsequent year image; And 5-2, judging that the image grid with the same sampling point and view angle is dismantled if the effective building elevation is identified in the previous year image and is not identified in the later year image.
  8. 8. The method for identifying micro-updates of street facades based on grid network according to claim 1, wherein the step 6 of identifying grid materials and comparing the changes thereof, the specific step of determining micro-updates according to the degree of the changes comprises the following steps: The step 6-1, wherein the materials comprise glass, concrete, bricks, stones and others; And 6-2, judging the degree of change by calculating the grid proportion of the material change, and judging that micro-update occurs when the proportion exceeds 30%, wherein the specific expression is as follows: Wherein, the method comprises the steps of, The material change ratio is represented, N represents the total number of grids of the vertical face in the reference year, Indicating whether the ith grid material is changed or not, if so =1, Otherwise 0; And 6-3, summarizing the update states of all the sampling points, and generating a city micro-update recognition result distribution map of the target area.
  9. 9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of a grid network street facade micro update identification method according to claim 1.
  10. 10. A computer readable storage medium, having stored thereon a computer program, characterized in that the computer program when executed by a processor implements the steps of a grid network street facade micro update based identification method according to claim 1.

Description

Street elevation micro-update identification method based on grid network Technical Field The invention relates to the technical field of urban updating monitoring, in particular to a street elevation micro-updating identification method based on a grid network. Background Along with the acceleration of the urban process and the transformation of the urban development mode, the urban updating gradually changes from a large-scale disassembly and large-scale construction to a fine treatment mode mainly comprising micro updating. Micro-updates usually involve subtle changes in building facade modifications, material replacement, local repairs, etc., which, although small in scale, have significant impact on urban landscape, neighborhood vitality and space quality. Therefore, continuous and accurate monitoring of micro-updating of urban building facades has become an important task in the fields of urban planning, historical protection, community management and the like. At present, urban updating monitoring mainly depends on technical means such as manual inspection, remote sensing image interpretation and oblique photogrammetry, is difficult to realize large-scale, systematic and dynamic monitoring, has limited spatial resolution, is difficult to identify detail changes such as building elevation materials, colors and components, and particularly in urban high-density built-up areas, the extraction and analysis of elevation information still face a plurality of challenges. In recent years, street view images are becoming an important data source for urban space perception and analysis due to the advantages of wide coverage, frequent updating, real viewing angle, high resolution and the like. Therefore, how to combine street view images and road network data to realize progressive recognition from building storage loss judgment to elevation material change provides reliable technical support for urban updating and monitoring, and is a technical problem to be solved currently urgently. Disclosure of Invention In order to solve the defects existing in the prior art, the application provides a street elevation micro-update identification method based on a grid network, which combines street view images and road network data, the method can realize progressive identification from building storage loss judgment to elevation material change, and realize automatic identification and monitoring of micro-update of the urban building elevation. The technical scheme adopted by the invention is as follows: a street elevation micro-update identification method based on a grid network comprises the following steps: Step 1, determining a target area, and acquiring road network data and street view image data of two different years of the area; step 2, grading the obtained road network, and arranging sampling points; step 3, respectively acquiring street view images of each sampling point at two years and multiple view angles, and identifying and associating building elevation; step 4, rasterizing the continuous building elevation image with different granularities; Step 5, judging new construction and dismantling based on the storage loss of the building elevation; And 6, identifying the grid material, comparing the grid material with the change, and judging micro-update according to the change degree. Further, acquiring road network data includes extracting road vector data of two year target areas from open street map data, and performing simplification processing to exclude expressways, bridges, and non-public traffic roads. In step 2, the road network is classified into a main road and a secondary road, sampling points are distributed on the main road at intervals of 50 meters, and sampling points are distributed on the secondary road at intervals of 30 meters. Further, in the step 3, street view images of each sampling point in two years and multiple view angles are obtained, and the street view images of each sampling point in two years and multiple view angles comprise front, back, left and right directions along the road direction, and the building elevation is identified and related. Further, building facades are identified by utilizing the pre-trained semantic segmentation model, and association marks are carried out. Further, the specific step of performing two kinds of granularity rasterization processing on the continuous building elevation image in the step 4 includes: step 4-1, dividing the building elevation image into rectangular grid units which are identical in size and are regularly arranged for subsequent analysis and comparison of material levels; step 4-2, dividing a rectangular grid unit of the building elevation image into two layers of a large-granularity grid with the granularity of 5 multiplied by 5m and a small-granularity grid with the granularity of 1 multiplied by 1 m; And 4-3, identifying new construction and dismantling of the large-granularity grids and identifying materials of the vertical