# US 2021/0240983 A1 · Method and apparatus for building extraction, and storage medium > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2021/0240983 A1 - Document ID: US20210240983A1 - Status: Published application - Kind: A1 - Application number: 17/236,050 - Filed: 2021-04-21 - Published: 2021-08-05 - Earliest claimed: 2020-09-11 - Assignees: Beijing Baidu Netcom Science Technology Co., Ltd. - Applicants: Beijing Baidu Netcom Science and Technology Co., Ltd. - Inventors: Jidong Peng; Cheng Wan; Cheng Chen; Shilei Wen; Ke Sun; Chengliang Luo - Web page: https://patentgrep.com/patent/US20210240983A1 ## CPC - G06V 20/176: {Urban or other man-made structures} - G06F 18/25: Fusion techniques - G06F 18/251: {of input or preprocessed data} - G06V 10/803: {of input or preprocessed data} - G06V 20/194: {using hyperspectral data, i.e. more or other wavelengths than RGB} - G06V 2201/10: Recognition assisted with metadata - Y02D 10/00: Energy efficient computing, e.g. low power processors, power management or thermal management - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - Y02D: CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE ## Abstract A method and an apparatus for building extraction, and a storage medium are provided in the present disclosure. The method includes: obtaining remote sensing image data and user behavior associated data of a target area, in which the user behavior associated data and the remote sensing image data have a spatiotemporal correlation; generating new channel data according to the user behavior associated data and the remote sensing image data and extracting a building in the target area according to the remote sensing image data and the new channel data. ## Claims ### Claim 1 A method for building exaction, comprising: obtaining remote sensing image data and user behavior associated data of a target area, wherein the user behavior associated data and the remote sensing image data have a spatiotemporal correlation; generating new channel data according to the user behavior associated data and the remote sensing image data; and extracting a building in the target area according to the remote sensing image data and the new channel data. ### Claim 2 (depends on 1) The method of claim 1, wherein generating the new channel data according to the user behavior associated data and the remote sensing image data comprises: mapping the user behavior associated data to pixels of the remote sensing image data according to location information, and generating the new channel data according to the user behavior associated data associated with the respective pixels. ### Claim 3 (depends on 2) The method of claim 2, wherein the user behavior associated data comprise at least one of points of interest (POI) data, location data, and search data. ### Claim 4 (depends on 3) The method of claim 3, wherein when the user behavior associated data comprise the POI data, mapping the user behavior associated data to pixels of the remote sensing image data according to location information, and generating the new channel data according to the user behavior associated data associated with the respective pixels comprise: determining neighboring POIs of the respective pixels in the remote sensing image data according to location information of respective candidate POIs in the POI data; and generating the new channel data according to POI data of the neighboring POIs. ### Claim 5 (depends on 4) The method of claim 4, wherein generating the new channel data according to POI data of the neighboring POIs comprises: for each pixel, generating a POI topic vector according to text data of the neighboring POIs, and generating the new channel data according to POI topic vectors of respective pixels; or, performing statistics on the neighboring POIs of the respective pixels, and generating the new channel data according to a statistical result. ### Claim 6 (depends on 3) The method of claim 3, wherein when the user behavior associated data comprise the location data, mapping the user behavior associated data to pixels of the remote sensing image data according to location information, and generating the new channel data according to the user behavior associated data associated with the respective pixels comprises: determining resident categories of location points associated with the respective pixels according to a distribution condition of the respective location points in the location data, wherein, the resident categories comprise a permanent category and a non-permanent category; and generating the new channel data according to the resident categories associated with the respective pixels. ### Claim 7 (depends on 6) The method of claim 6, wherein generating the new channel data according to the resident categories associated with the respective pixels comprises: performing statistics on the location points associated with the respective pixels according to the resident categories, and generating the new channel data according to a statistical result. ### Claim 8 (depends on 3) The method of claim 3, wherein when the user behavior associated data comprise the location data, mapping the user behavior associated data to pixels of the remote sensing image data according to location information, and generating the new channel data according to the user behavior associated data associated with the respective pixels comprise: determining location data associated with the respective pixels of the remote sensing image data according to the distribution of each location point in the location data; and performing statistics on at least one of the number of location points and the number of location users in the location data associated with the respective pixels, and generating the new channel data according to a statistical result. ### Claim 9 (depends on 3) The method of claim 3, wherein when the user behavior associated data comprise the search data, mapping the user behavior associated data to pixels of the remote sensing image data according to location information, and generating the new channel data according to the user behavior associated data associated with the respective pixels comprises: determining a searching time of keyword combinations associated with the respective pixels of the remote sensing image data according to the location information of the search data; and generating the new channel data according to searching times associated with the respective pixels. ### Claim 10 (depends on 1) The method of claim 1, wherein extracting the building in the target area according to the remote sensing image data and the new channel data comprises: inputting the remote sensing image data and the new channel data to a pre-trained building extraction model, and extracting the building in the target area. ### Claim 11 A building exaction apparatus, comprising: one or more processors; a memory storing instructions executable by the one or more processors; wherein the one or more processors are configured to: obtain remote sensing image data and user behavior associated data of a target area, wherein the user behavior associated data and the remote sensing image data have a spatiotemporal correlation; generate new channel data according to the user behavior associated data and the remote sensing image data; and extract a building in the target area according to the remote sensing image data and the new channel data. ### Claim 12 (depends on 11) The apparatus of claim 11, wherein the one or more processors are configured to: map the user behavior associated data to pixels of the remote sensing image data according to location information, and generate the new channel data according to the user behavior associated data associated with the respective pixels. ### Claim 13 (depends on 12) The apparatus of claim 12, wherein the user behavior associated data comprise at least one of points of interest (POI) data, location data, and search data. ### Claim 14 (depends on 13) The apparatus of claim 13, wherein when the user behavior associated data comprise the POI data, the one or more processors are configured to: determine neighboring POIs of each the respective pixels in the remote sensing image data according to the location information of each respective candidate POIs in the POI data; and generate the new channel data according to POI data of the neighboring POIs. ### Claim 15 (depends on 14) The apparatus of claim 14, wherein the one or more processors are configured to: for each pixel, generate a POI topic vector according to text data of the neighboring POIs, and generate the new channel data according to POI topic vectors of respective pixels; or, perform statistics on the neighboring POIs of the respective pixels, and generating the new channel data according to a statistical result. ### Claim 16 (depends on 13) The apparatus of claim 13, wherein when the user behavior associated data comprise the location data, the one or more processors are configured to: determine resident categories of location points associated with the respective pixels according to a distribution condition of the respective location points in the location data, wherein, the resident categories comprise a permanent category and a non-permanent category; and generate the new channel data according to the resident categories associated with the respective pixels. ### Claim 17 (depends on 16) The apparatus of claim 16, wherein the one or more processors are configured to: perform statistics on the location points associated with the respective pixels according to the resident categories, and generate the new channel data according to a statistical result. ### Claim 18 (depends on 13) The apparatus of claim 13, wherein when the user behavior associated data comprise the location data, the one or more processors are configured to: determine location data associated with the respective pixels of the remote sensing image data according to the distribution of each location point in the location data; and perform statistics on at least one of the number of location points and the number of location users in the location data associated with the respective pixels, and generating the new channel data according to a statistical result. ### Claim 19 (depends on 13) The apparatus of claim 13, wherein when the user behavior associated data comprise search data, the one or more processors are configured to: determine a searching time of keyword combinations associated with the respective pixels of the remote sensing image data according to the location information of the search data; and generate the new channel data according to searching times associated with the respective pixels. ### Claim 20 A non-transitory computer-readable storage medium having computer instructions stored thereon, wherein the computer instructions are configured to execute a method for building extraction, and the method comprises: obtaining remote sensing image data and user behavior associated data of a target area, wherein the user behavior associated data and the remote sensing image data have a spatiotemporal correlation; generating new channel data according to the user behavior associated data and the remote sensing image data; and extracting a building in the target area according to the remote sensing image data and the new channel data.