WO2023115661A1 - 区域的智能划分方法及装置 - Google Patents

区域的智能划分方法及装置 Download PDF

Info

Publication number
WO2023115661A1
WO2023115661A1 PCT/CN2022/070839 CN2022070839W WO2023115661A1 WO 2023115661 A1 WO2023115661 A1 WO 2023115661A1 CN 2022070839 W CN2022070839 W CN 2022070839W WO 2023115661 A1 WO2023115661 A1 WO 2023115661A1
Authority
WO
WIPO (PCT)
Prior art keywords
area
item
target
initial
sub
Prior art date
Application number
PCT/CN2022/070839
Other languages
English (en)
French (fr)
Inventor
陈小平
罗韬
杨旭
王云华
Original Assignee
广东栗子科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 广东栗子科技有限公司 filed Critical 广东栗子科技有限公司
Publication of WO2023115661A1 publication Critical patent/WO2023115661A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/02—Control of position or course in two dimensions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/587—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using geographical or spatial information, e.g. location
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes

Definitions

  • the invention relates to the field of smart devices, in particular to a method and device for intelligently dividing areas.
  • the existing technology mainly sets markers or reference objects on the boundary of the area manually, such as setting markers on the door frame of the room artificially, and recognizes the boundary of the area according to the manually set markers or reference objects, and then realizes the division of the area .
  • the existing technology relies on manual marking, the operation is complicated and the efficiency is low; at the same time, when there is no pre-set marker on the area boundary in the area, the boundary of each area cannot be accurately identified, and thus the accurate area division result cannot be obtained . Therefore, how to improve the accuracy of the result of intelligent area division is a technical problem to be solved in this field.
  • the technical problem to be solved by the present invention is how to improve the accuracy of the results of intelligent area division, and provide an intelligent area division method and device, which can realize the intelligent area division simply and efficiently by identifying items in the area.
  • the first aspect of the present invention discloses a method for intelligently dividing regions, the method comprising:
  • the target scene includes a plurality of divided areas
  • the initial area map includes an initial area submap corresponding to each of the divided areas
  • the initial area map of the target scene is updated according to the target area sub-map corresponding to the target divided area.
  • the item information of the item includes the type information of the item and the first location information of the item, wherein the first location information of the item The location information of the item in the initial area submap corresponding to the target division area;
  • the updating of the initial area submap corresponding to the target division area based on the item information of the item includes:
  • the type information of the item determine whether the item is a boundary type item
  • the distribution sub-area where the item is located in the initial area sub-map corresponding to the target division area is determined based on the first position information of the item, and according to the The distribution sub-area updates the initial area sub-map corresponding to the target divided area, wherein the distribution sub-area where the item is located is used to represent the distribution of the item in the target divided area;
  • the initial boundary position information is the initial area sub-map corresponding to the target divided area The initial boundary position information corresponding to the item in ;
  • the boundary location information updates the initial region sub-map corresponding to the target divided region.
  • the item information of the item further includes size information of the item, wherein the size information of the item is the size information of the item in the target division area The size information in the corresponding initial area submap;
  • the initial boundary position information is updated based on the first position information of the item to obtain target boundary position information, and according to the Before updating the target boundary position information to the initial area submap corresponding to the target divided area, the method further includes:
  • the item edge position point of the item in the initial area submap corresponding to the target divided area, wherein the item edge position point of the item It is used to indicate the edge position of the item, and there are multiple item edge position points of the item;
  • the cleaning device controlling the cleaning device to move within the area to be tested corresponding to the item, and recording the movement track of the cleaning device in the initial area submap corresponding to the target divided area, wherein the area to be tested corresponding to the item is the area containing all item edge position points of said item;
  • the moving track of the cleaning device is a closed track, and the moving track of the cleaning device includes all edge position points of the item, then it is determined based on the first position information of the item that the target division area corresponds to The distribution sub-area where the item is located in the initial area sub-map, and update the initial area sub-map corresponding to the target division area according to the distribution sub-area;
  • the moving track of the cleaning device is not a closed track, or the moving track of the cleaning device does not include all edge position points of the item, then performing the updating the Initial boundary position information, obtaining target boundary position information, and updating the initial area submap corresponding to the target divided area according to the target boundary position information.
  • the item information of the item further includes size information of the item, wherein the size information of the item is the size information of the item in the target division area The size information in the corresponding initial area submap;
  • the first location information of the item is used to determine the distribution sub-area where the item is located in the initial area sub-map corresponding to the target division area, and update the distribution sub-area corresponding to the target division area according to the distribution sub-area.
  • Initial area submap including:
  • update the initial distribution sub-area corresponding to the item based on the current distribution sub-area where the item is located obtain the target distribution sub-area of the item, and update the item according to the target distribution sub-area of the item.
  • the method before updating the initial area submap corresponding to the target division area based on the item information of the item, the method further includes:
  • the determining the initial boundary position information corresponding to the item in the initial area submap corresponding to the target divided area includes:
  • the method further includes:
  • all the items in the target division area include items whose category information is the restricted category of the area, then obtain the target item from all the items in the target division area, and determine the area attribute corresponding to the target item An attribute for dividing the target area, wherein the target item is an item whose category information is a category restricted by the area;
  • the area attributes corresponding to each of the items in the target divided area determine the intersection area attribute of all items in the target divided area, and determine the intersection area attribute as the attribute of the target divided area;
  • the attribute of the target divided area is determined as the attribute of the target area submap corresponding to the target divided area.
  • the method after determining the intersection area attributes of all items in the target divided area, the determining the intersection area attribute as the target divided area Before the attribute, the method also includes:
  • the attribute of the target intersection area with the highest priority among the attributes of the intersection area is obtained according to the priority of the preset area attributes, and the attribute of the target intersection area is determined as the attribute of the target division area.
  • the method further includes:
  • the item recognition probability of the item is less than a preset threshold, adjust the shooting angle of the cleaning device, and perform the acquisition of the target image of the target division area based on the cleaning device, and the analysis of the target image to obtain the Items contained in the target image, and item information of the items.
  • the second aspect of the present invention discloses a device for intelligently dividing regions, the device comprising:
  • An acquisition module configured to acquire an initial area map for any target scene, where the target scene includes a plurality of divided areas, and the initial area map includes an initial area submap corresponding to each of the divided areas;
  • a collection module configured to collect a target image of a target divided area, where the target divided area is any one of all the divided areas;
  • An analysis module configured to analyze the target image to obtain items contained in the target image and item information of the items
  • a first update module configured to update the initial area submap corresponding to the target divided area based on the item information of the item, to obtain the target area submap corresponding to the target divided area;
  • the second updating module is configured to update the initial area map of the target scene according to the target area sub-map corresponding to the target divided area.
  • the item information of the item includes the type information of the item and the first location information of the item, wherein the first location information of the item The location information of the item in the initial area submap corresponding to the target division area;
  • the first update module includes:
  • the first judging submodule is used to judge whether the item is a boundary type item according to the type information of the item;
  • the distribution update submodule is configured to determine that the item is in the initial area submap corresponding to the target division area based on the first position information of the item if the first determination submodule determines that the item is a non-boundary type item The distribution sub-area where the item is located, and update the initial area sub-map corresponding to the target division area according to the distribution sub-area, wherein the distribution sub-area where the item is located is used to indicate that the item is in the The distribution of the target division area;
  • the second judging submodule is used to judge whether the first position information of the item matches the initial boundary position information if the first judging submodule judges that the item is a boundary type item, wherein the initial The boundary position information is the initial boundary position information corresponding to the item in the initial area submap corresponding to the target divided area;
  • a boundary update submodule configured to update the initial boundary based on the first position information of the item if the second judging submodule determines that the first position information of the item does not match the initial boundary position information
  • the location information is to obtain the target boundary location information, and update the initial area submap corresponding to the target divided area according to the target boundary location information.
  • the item information of the item further includes size information of the item, wherein the size information of the item is the size information of the item in the target division area The size information in the corresponding initial area submap;
  • the first update module also includes:
  • a determining submodule configured to, after the second judging submodule judges that the first location information of the item does not match the initial boundary location information, the boundary update submodule based on the first location information of the item Updating the initial boundary position information to obtain the target boundary position information, and before updating the initial area sub-map corresponding to the target divided area according to the target boundary position information, based on the size information of the item and the first Position information, determining the item edge position point of the item in the initial area submap corresponding to the target divided area, wherein the item edge position point of the item is used to represent the edge position of the item, and the item edge position point of the item There are multiple points on the edge of the item;
  • the control submodule is used to control the movement of the cleaning equipment in the area to be measured corresponding to the item, and record the movement track of the cleaning equipment in the initial area sub-map corresponding to the target division area, wherein the item corresponds to The area to be tested is the area containing all edge position points of the item;
  • the third judging sub-module is used to judge whether the moving track of the cleaning device is a closed track, and whether the moving track of the cleaning device includes all edge positions of the items;
  • the distribution updating sub-module is further configured to determine that the moving track of the cleaning device is a closed track by the third judging sub-module, and the moving track of the cleaning device includes all edge position points of the item , then determine the distribution sub-area where the item is located in the initial area sub-map corresponding to the target division area based on the first position information of the item, and update the distribution sub-area corresponding to the target division area according to the distribution sub-area initial region submap;
  • the boundary update submodule is further configured to determine that the movement track of the cleaning device is not a closed track, or the movement track of the cleaning device does not include all the object edges of the item if the third judgment submodule determines that position point, update the initial boundary position information based on the first position information of the item to obtain target boundary position information, and update the initial area submap corresponding to the target divided area according to the target boundary position information.
  • the item information of the item further includes size information of the item, wherein the size information of the item is the size information of the item in the target division area The size information in the corresponding initial area submap;
  • the distribution update submodule determines the distribution sub-area where the item is located in the initial area sub-map corresponding to the target division area based on the first position information of the item, and updates the target according to the distribution sub-area
  • the methods of dividing the initial region submap corresponding to the region include:
  • update the initial distribution sub-area corresponding to the item based on the current distribution sub-area where the item is located obtain the target distribution sub-area of the item, and update the item according to the target distribution sub-area of the item.
  • the device further includes:
  • the first determining module is used for determining the item corresponding to the item in the initial area submap corresponding to the target divided area before updating the initial area submap corresponding to the target divided area based on the item information of the item Initial boundary position information;
  • the manner in which the first determination module determines the initial boundary position information corresponding to the item in the initial area submap corresponding to the target division area includes:
  • the device further includes: a second determining module, configured to:
  • all the items in the target division area include items whose category information is the restricted category of the area, then obtain the target item from all the items in the target division area, and determine the area attribute corresponding to the target item An attribute for dividing the target area, wherein the target item is an item whose category information is a category restricted by the area;
  • the area attributes corresponding to each of the items in the target divided area determine the intersection area attribute of all items in the target divided area, and determine the intersection area attribute as the attribute of the target divided area;
  • the attribute of the target divided area is determined as the attribute of the target area sub-map corresponding to the target divided area.
  • the second determination module is further configured to: after determining the intersection area attributes of all items in the target divided area, the Before the attribute of the intersection area is determined as the attribute of the target division area, it is judged whether the attribute of the intersection area is unique;
  • the attribute of the target intersection area with the highest priority among the attributes of the intersection area is obtained according to the priority of the preset area attributes, and the attribute of the target intersection area is determined as the attribute of the target division area.
  • the analysis module is also used for:
  • the first update module is further configured to execute the item information based on the item and update the corresponding target division area if the analysis module determines that the item recognition probability of the item is greater than or equal to a preset threshold. The operation of obtaining the target area submap corresponding to the target divided area of the initial area submap;
  • the acquisition module is further configured to adjust the shooting angle of the cleaning equipment if the analysis module judges that the item identification probability of the item is less than a preset threshold, and execute the target of dividing the area based on the acquisition target of the cleaning equipment. image, and the operation of analyzing the target image to obtain items contained in the target image and item information of the items.
  • the third aspect of the present invention discloses another device for intelligent division of regions, which includes:
  • a processor coupled to the memory
  • the processor invokes the executable program code stored in the memory to execute the method for intelligently dividing regions disclosed in the first aspect of the present invention.
  • the fourth aspect of the present invention discloses a cleaning device, which is used to implement the method for intelligently dividing areas disclosed in the first aspect of the present invention.
  • the fifth aspect of the present invention discloses a computer-storage medium, the computer storage medium stores computer instructions, and when the computer instructions are invoked, it is used to execute the method for intelligently dividing regions disclosed in the first aspect of the present invention.
  • the initial area map for any target scene is acquired, the target scene includes a plurality of divided areas, and the initial area map includes the initial area sub-map corresponding to each divided area; the target image of the target divided area is collected based on the cleaning equipment , the target divided area is any area in all divided areas; analyze the target image to obtain the items contained in the target image and the item information of the item; based on the item information of the item, update the initial area submap corresponding to the target divided area, and get The target area sub-map corresponding to the target division area, and then realize the update of the initial area map of the target scene.
  • the implementation of the present invention can realize the intelligent division of the area based on the items in the target area, and can improve the accuracy of the intelligent division of the area while avoiding manual setting of markers.
  • Fig. 1 is a schematic flow chart of an intelligent region division method disclosed in an embodiment of the present invention
  • Fig. 2 is a schematic flowchart of an initial area sub-map corresponding to an item-based item information update target division area disclosed in an embodiment of the present invention
  • Fig. 3 is a schematic flow diagram of updating the initial area sub-map corresponding to the target division area based on the distribution sub-area where the item is located according to the embodiment of the present invention
  • Fig. 4 is a schematic flowchart of another method for intelligently dividing regions disclosed in an embodiment of the present invention.
  • Fig. 5 is a schematic flowchart of determining the attributes of the target area sub-map corresponding to the target division area disclosed in the embodiment of the present invention
  • Fig. 6 is a schematic structural diagram of an intelligent region division device disclosed in an embodiment of the present invention.
  • Fig. 7 is a schematic structural diagram of a first update module disclosed in an embodiment of the present invention.
  • Fig. 8 is a schematic structural diagram of another intelligent region division device disclosed in an embodiment of the present invention.
  • Fig. 9 is a schematic structural diagram of another intelligent region division device disclosed in an embodiment of the present invention.
  • the invention discloses an area intelligent division method and device, and cleaning equipment, which can realize area intelligent division based on objects in a target area, and improve the accuracy of area intelligent division results. Each will be described in detail below.
  • FIG. 1 is a schematic flowchart of a method for intelligently dividing regions disclosed in an embodiment of the present invention.
  • the intelligent area division method described in FIG. 1 may be applied to an intelligent cleaning device or to a server, which is not limited in this embodiment of the present invention.
  • the intelligent division method of this area may include the following operations:
  • the target scene includes multiple divided areas, and the initial area map includes an initial area submap corresponding to each divided area.
  • the initial area map of the target scene can be obtained by scanning the target scene with lidar, it can be obtained based on manually set boundary markers, and it can also be obtained by downloading the initial map of the scene pre-stored in the database.
  • the target scene is specifically a scene requiring area division, which may be an indoor scene including multiple divided areas, or other scenes requiring area division.
  • the acquired target scene includes multiple divided areas, and correspondingly, the acquired initial area map of the target scene includes an initial area sub-map corresponding to each divided area.
  • the target image of the target divided area is collected, where the target divided area is any one of all the divided areas.
  • the cleaning device includes a collection device.
  • the collection device included in the cleaning device may be a single-eye camera device, a multi-eye camera device, or a camera device including a pan-tilt.
  • This embodiment of the invention does not limited.
  • the target image of any one of the multiple divided areas of the target scene can be collected by the cleaning device.
  • the collected target image can be a single image of the target divided area, multiple images of different directions and different angles of the target divided area, One or more of the panoramic images of the target divided area.
  • the target image collected by the cleaning device includes objects in the target divided area. There may be one or more items in the target divided area included in the target image.
  • the items included in the target images and the item information of the items can be identified.
  • the item information of the item may include type information of the item and first location information of the item, and may also include size information of the item.
  • the type information of the item can be a boundary type or a non-boundary type. For example, if the target image is analyzed to include a door or door frame, a coffee table or a dining table, the type information of the door or door frame can be obtained as a boundary type, and the type information of the coffee table or dining table is a non-boundary type.
  • the first position information of the item is the position information of the item in the initial area sub-map corresponding to the target divided area, and the item can be located in the initial area sub-map through the first position information of the item.
  • the size information of the item is the size information of the item in the initial area sub-map corresponding to the target division area, and the size of the spatial area occupied by the item can be marked in the initial area sub-map through the size information of the item.
  • the initial area submap corresponding to the target divided area is updated to obtain the target area submap corresponding to the target divided area.
  • the area boundary or area distribution of the initial area sub-map can be updated based on the item information of the analyzed items, and then the target area sub-map corresponding to the intelligently divided target area can be obtained.
  • the initial area submap corresponding to the target division area is updated through the item information of the item, and after the target area submap corresponding to the target division area is obtained, the area map of the target scene can be correspondingly updated.
  • the implementation of the intelligent division of regions as described in Figure 1 can realize the intelligent division of regions based on the objects in the target region, and improve the accuracy of the results of intelligent region division while avoiding manual setting of markers.
  • FIG. 2 is a schematic flowchart of an initial area sub-map corresponding to an item-based item information update target division area disclosed in an embodiment of the present invention.
  • Step S4 is based on an item-based item information update.
  • the initial area submap corresponding to the target division area may include the following steps:
  • step S401 According to the type information of the item, determine whether the item is a boundary type item. If it is determined that the item is a non-boundary type item, perform step S402; if it is determined that the item is a boundary type item, perform steps S403-S404.
  • step S401 it is judged whether the item is a boundary type item according to the type information of the item identified in step S3.
  • the boundary of the initial region submap can be corrected according to the boundary type item; Items, but can be used as items to modify the distribution area of the initial area submap.
  • S402. Determine the distribution sub-area where the item is located in the initial area sub-map corresponding to the target division area based on the first location information of the item, and update the initial area sub-map corresponding to the target division area according to the distribution sub-area, wherein the item is located
  • the distribution sub-area of is used to represent the distribution of items in the target division area.
  • the distribution area of the initial area submap may be corrected according to the item. Specifically, the distribution of the items in the target divided area may be determined according to the distribution sub-area where the item is located in the initial area sub-map corresponding to the target divided area.
  • step S403. Determine whether the first position information of the item matches the initial boundary position information, wherein the initial boundary position information is the initial boundary position information corresponding to the item in the initial area submap corresponding to the target divided area. If it is determined that the first position information of the item does not match the initial boundary position information, step S404 is performed.
  • the initial boundary position information corresponding to the item in the initial area sub-map corresponding to the target divided area is accurate, and there is no need to pass The location information of the item is corrected.
  • the initial boundary position information corresponding to the item in the initial area submap corresponding to the target division area is inaccurate, and it is necessary to pass the item information of the identified item, Correct the initial boundary position information.
  • the initial boundary position information needs to be corrected through the item information of the identified item. Specifically, modify the initial boundary position information corresponding to the item in the initial area submap corresponding to the target division area according to the first position information of the item, obtain the corrected target boundary position information, and then update the target division area according to the corrected target boundary position information The corresponding initial region submap.
  • this optional embodiment can determine the content to be updated of the initial area submap according to the type information of the item, and further modify or update the boundary position/ distribution area, and then obtain a more accurate regional sub-map of the target division area.
  • step S404 updates the initial boundary position information based on the first position information of the item, obtains the target boundary position information, and updates according to the target boundary position information
  • the method further includes:
  • the item information of the item includes the size information of the item, and through the size information of the item and the first position information, multiple items of the item in the initial area sub-map corresponding to the target division area can be determined Edge location points.
  • the edge position point of the item may be the vertex of the item, or any point on the edge of the item, which is not limited in this embodiment of the present invention.
  • the area to be tested including all the edge position points of the item can be determined according to all the edge position points of the item.
  • the area to be tested is a virtual area in the initial area map and is not affected by the initial The limit of the initial boundary in the area map. By controlling the cleaning equipment to move in the area to be tested, it can be determined whether the item is placed at the boundary of the area.
  • step S407. Determine whether the moving track of the cleaning device is a closed track, and whether the moving track of the cleaning device includes all edge position points of the item. If the movement trajectory of the cleaning equipment is a closed trajectory, and the movement trajectory of the cleaning equipment contains all the object edge position points of the item, then perform step S402; if the movement trajectory of the cleaning equipment is not a closed trajectory, or there is no If all edge positions of the item are included, step S404 is executed.
  • the moving trajectory of the cleaning device is a closed trajectory
  • the closed trajectory contains all the edge position points of the item, that is, the item is not placed on the boundary position
  • the item can be used as the corrected initial Items in the distribution area of the area sub-map, and then perform step S402 to determine the distribution sub-area where the item is located in the initial area sub-map corresponding to the target division area based on the first position information of the item, and update the distribution sub-area corresponding to the target division area according to the distribution sub-area Operations on the initial region submap.
  • step S404 is performed to update the initial boundary position information based on the first position information of the item to obtain the target boundary position information, and update the initial area submap corresponding to the target divided area according to the target boundary position information.
  • FIG. 3 is a schematic flowchart of updating the initial area sub-map corresponding to the target division area based on the distribution sub-area where the item is located according to the embodiment of the present invention.
  • Step S402 based on the first position information of the item, determine the distribution sub-area where the item is located in the initial area sub-map corresponding to the target division area, and update the initial area sub-map corresponding to the target division area according to the distribution sub-area, including the following steps:
  • the item information of the item includes the size information of the item. Through the size information and the first position information of any item, it can be determined that the item is located in the initial area sub-map corresponding to the target division area.
  • the current distribution sub-area where the item is located is the current distribution of the item in the target division area.
  • step S4024 Obtain the initial distribution sub-area corresponding to the item, and judge whether the initial distribution sub-area corresponding to the item matches the current distribution sub-area where the item is located. If not, execute step S4025.
  • the target distribution sub-area of the item is obtained after updating the initial distribution sub-area corresponding to the item in the current distribution sub-area where the item is located in the initial area sub-map, and the position of the item in the target division area can be determined. distribution, and then obtain the initial area sub-map corresponding to the updated target division area.
  • the distribution of items in the target division area can be determined according to the current distribution sub-area where the item is located, and further the corresponding distribution of the target division area can be updated through the current distribution sub-area where the item The distribution area of the initial area sub-map, and then obtain a more accurate area sub-map of the target division area.
  • the method before updating the initial area submap corresponding to the target division area based on the item information of the item, the method further includes:
  • determining the initial boundary position information corresponding to the item in the initial area submap corresponding to the target divided area includes the following steps:
  • the initial boundary position information corresponding to the item in the initial area sub-map corresponding to the target divided area can be determined through the collection direction when the cleaning device collects the target image of the target divided area.
  • determining the collection direction when the cleaning device collects the target image of the target divided area may be determined by a direction sensor, or may be determined by an image recognition technology.
  • the distribution of items in the target division area can be determined according to the current distribution sub-area where the item is located, and further the corresponding distribution of the target division area can be updated through the current distribution sub-area where the item The distribution area of the initial area sub-map, and then obtain a more accurate area sub-map of the target division area.
  • the method further includes the following steps:
  • the item identification probability of the item is less than the preset threshold, then adjust the shooting angle of the cleaning device, and execute the target image based on the cleaning device to collect the target division area, and analyze the target image to obtain the items contained in the target image and the object's Operation of item information.
  • the item recognition probability of the item included in the target image is determined.
  • the item recognition probability is used to indicate the accuracy of the recognition result.
  • the item recognition probability can be determined by the degree of matching between the current item and the standard item. When the item identification probability exceeds the preset threshold, it indicates that the current identification result is accurate, and subsequent operations can be performed on the identified item and item information. When the item recognition probability does not exceed the preset threshold, it indicates that the current recognition result is inaccurate, and it is necessary to adjust the shooting angle of the acquisition device, re-collect the image of the target division area, and re-identify to obtain an accurate recognition result.
  • the accuracy of the recognition result can be judged, and an inaccurate recognition result can be adaptively changed for re-recognition to prevent misjudgment. While ensuring the accuracy of item identification results, it also ensures the accuracy of subsequent intelligent division of areas.
  • FIG. 4 is a schematic flowchart of another method for intelligently dividing regions disclosed in an embodiment of the present invention.
  • the intelligent division method of this area may include the following operations:
  • the target scene includes multiple divided areas, and the initial area map includes an initial area submap corresponding to each divided area.
  • the target image of the target divided area is collected, where the target divided area is any one of all the divided areas.
  • the initial area submap corresponding to the target divided area is updated to obtain the target area submap corresponding to the target divided area.
  • step S6 may be performed before step S5, or may be performed after step S5, which is not limited in this embodiment of the present invention.
  • step S6 determining the attribute of the target area sub-map corresponding to the target divided area includes the following steps:
  • the target divided area includes one or more items, and all items included in the target divided area can be acquired through a single, multiple or panoramic target images. And the category information of each item included in the target division area is identified through the image recognition technology. If the target division area includes item 1, item 2, and item 3, it is recognized that the category information of item 1 is a dining table, the category information of item 2 is coffee table, and the category information of item 3 is sofa.
  • S602. Determine whether all the items in the target division area include items whose category information is an area-restricted category.
  • the area-restricted category is used to indicate that the probability that the corresponding item belongs to a determined area is greater than or equal to a preset probability threshold. If all the items in the target division area include items whose category information is the area-restricted category, then perform steps S603 and S606; S604-step S606.
  • the region-restricted category is used to indicate that the probability that the corresponding item belongs to a determined region is greater than or equal to the preset probability threshold, which can be set according to the needs of the actual application scenario.
  • the category information of the item 4 is a range hood
  • the range hood category information may be set as an area-restricted category, which means that the probability that the item 4 belongs to the kitchen is greater than or equal to the preset probability threshold.
  • the acquired category information is the items of the area-restricted category, for example, all items in the target division area include item 4 (the category information is range hood), the item 4 is taken as the target item, and the area attribute (kitchen) corresponding to the item 4 is determined as the attribute (kitchen) of the target divided area.
  • the attribute of the target division area needs to be determined through the area attributes of multiple items in the target division area.
  • the area attributes of item 1 in the target division area are dining room and living room
  • the area attributes of item 2 are living room
  • the area attributes of item 3 are living room and bedroom.
  • the intersection area attribute set of the target division area can be obtained, and the area attributes in the intersection area attribute set can be used as candidate attributes of the target division area. For example, if the attribute of the intersection area of all items in the target division area is the living room, then the living room is determined as the attribute of the target division area.
  • step S605 determines the intersection area attribute of all items in the target divided area, before determining the intersection area attribute as the attribute of the target divided area, the method further includes:
  • the operation of determining the attribute of the intersection area as the attribute of the target division area is performed.
  • the attribute of the target intersection area with the highest priority among the attributes of the intersection area is obtained according to the priority of the preset area attributes, and the attribute of the target intersection area is determined as the attribute of the target division area.
  • the intersection area attribute may be determined as the attribute of the target division area. If there are multiple intersection area attributes in the intersection area attribute set, the target intersection area attribute with the highest priority can be determined by presetting the area attribute priority, and the target intersection area attribute with the highest priority can be determined as the attribute of the target division area.
  • the region attribute priority can be set differently according to different target scenarios.
  • other intersection area attributes in the intersection area attribute set except the target intersection area attribute are used as candidate attributes of the target division area for selection by the operator.
  • the attribute of the target divided area is determined as the attribute of the target area sub-map corresponding to the target divided area, and displayed in the target area sub-map of the target divided area.
  • implementing the intelligent division of regions as described in FIG. 4 can determine the attributes of the target division region based on the items in the target region, and can accurately obtain the result of intelligent region division.
  • the method further includes :
  • the area attribute set of the target divided area is obtained according to the area attribute corresponding to each item in the target divided area.
  • the number of items including the area attribute is counted to obtain the frequency of the area attribute.
  • the attributes of the target area sub-map corresponding to the target divided area can be determined through the common area attributes of multiple items in the target divided area, and the area intelligent division result can be accurately obtained.
  • FIG. 6 is a schematic structural diagram of an apparatus for an intelligent division method of an area disclosed in an embodiment of the present invention.
  • the device for intelligently dividing the area includes: an acquisition module 601 , a collection module 602 , an analysis module 603 , a first update module 604 , and a second update module 605 . in:
  • the acquiring module 601 is configured to acquire an initial area map for any target scene, the target scene includes multiple divided areas, and the initial area map includes an initial area sub-map corresponding to each divided area.
  • the initial area map of the target scene can be obtained by scanning the target scene with lidar, it can be obtained based on manually set boundary markers, and it can also be obtained by downloading the initial map of the scene pre-stored in the database.
  • the target scene is specifically a scene requiring area division, which may be an indoor scene including multiple divided areas, or other scenes requiring area division.
  • the acquired target scene includes multiple divided areas, and correspondingly, the acquired initial area map of the target scene includes an initial area sub-map corresponding to each divided area.
  • the collection module 602 is configured to collect a target image of a target divided area, where the target divided area is any area in all divided areas.
  • the acquisition module can be integrated into the cleaning equipment.
  • the acquisition module included in the cleaning equipment can be a single-eye camera device, a multi-eye camera device, or a camera device including a pan-tilt, which is not limited in this embodiment of the invention.
  • the target image of any one of the multiple divided areas of the target scene can be collected through the acquisition module, and the collected target image can be a single image of the target divided area, multiple images of different directions and different angles of the target divided area, One or more of the panoramic images of the target divided area.
  • the analysis module 603 is configured to analyze the target image to obtain items contained in the target image and item information of the items.
  • the target image collected by the cleaning device includes objects in the target divided area. There may be one or more items in the target divided area included in the target image.
  • the items included in the target images and the item information of the items can be identified.
  • the item information of the item may include type information of the item and first location information of the item, and may also include size information of the item.
  • the type information of the item can be a boundary type or a non-boundary type. For example, if the target image is analyzed to include a door or door frame, a coffee table or a dining table, the type information of the door or door frame can be obtained as a boundary type, and the type information of the coffee table or dining table is a non-boundary type.
  • the first position information of the item is the position information of the item in the initial area sub-map corresponding to the target divided area, and the item can be located in the initial area sub-map through the first position information of the item.
  • the size information of the item is the size information of the item in the initial area sub-map corresponding to the target division area, and the size of the spatial area occupied by the item can be marked in the initial area sub-map through the size information of the item.
  • the first updating module 604 is configured to update the initial area submap corresponding to the target divided area based on the item information of the item, and obtain the target area submap corresponding to the target divided area.
  • the area boundary or area distribution of the initial area sub-map can be updated based on the item information of the analyzed items, and then the target area sub-map corresponding to the intelligently divided target area can be obtained.
  • the second updating module 605 is configured to update the initial area map of the target scene according to the target area sub-map corresponding to the target divided area.
  • the initial area submap corresponding to the target division area is updated through the item information of the item, and after the target area submap corresponding to the target division area is obtained, the area map of the target scene can be correspondingly updated.
  • FIG. 7 is a schematic structural diagram of a first update module disclosed in an embodiment of the present invention.
  • the first updating module 604 includes a first judging submodule 6041 , a distribution updating submodule 6042 , a second judging submodule 6043 , and a boundary updating submodule 6044 .
  • the first judging sub-module 6041 is used for judging whether the item is a boundary type item according to the type information of the item.
  • the boundary of the initial region submap can be corrected according to the boundary type item;
  • the first judging sub-module 6041 judges that the item is a non-boundary type item then the The item cannot be used as an item to modify the boundary of the submap of the initial area, but can be used as an item to modify the distribution area of the submap of the initial area.
  • the distribution update sub-module 6042 is used to determine the distribution sub-module where the item is located in the initial area sub-map corresponding to the target division area based on the first position information of the item if the first judgment sub-module 6041 determines that the item is a non-boundary type item. area, and update the initial area submap corresponding to the target division area according to the distribution sub-area, wherein the distribution sub-area where the item is located is used to represent the distribution of the item in the target division area.
  • the distribution area of the initial area submap may be corrected according to the item. Specifically, the distribution of the items in the target divided area may be determined according to the distribution sub-area where the item is located in the initial area sub-map corresponding to the target divided area.
  • the second judging sub-module 6043 is used to judge whether the first position information of the item matches the initial boundary position information if the first judging sub-module 6041 determines that the item is a boundary type item, wherein the initial boundary position information is the target division The initial boundary position information corresponding to the item in the initial area submap corresponding to the area.
  • the initial boundary position corresponding to the item in the initial area submap corresponding to the target divided area The information is accurate and does not need to be corrected by the location information of the item.
  • the second judging sub-module 6043 judges that the first position information of the item does not match the initial boundary position information, the initial boundary position information corresponding to the item in the initial area sub-map corresponding to the target divided area is inaccurate, and needs to be identified by The item information of the item, and the initial boundary position information is corrected.
  • the boundary updating sub-module 6044 is used to update the initial boundary position information based on the first position information of the item to obtain the target boundary position information if the second judging sub-module 6043 judges that the first position information of the item does not match the initial boundary position information , and update the initial area submap corresponding to the target division area according to the target boundary position information.
  • the initial boundary position information needs to be corrected through the item information of the identified item. Specifically, modify the initial boundary position information corresponding to the item in the initial area submap corresponding to the target division area according to the first position information of the item, obtain the corrected target boundary position information, and then update the target division area according to the corrected target boundary position information The corresponding initial region submap.
  • this optional embodiment can determine the content to be updated of the initial area submap according to the type information of the item, and further modify or update the boundary position/ distribution area, and then obtain a more accurate regional sub-map of the target division area.
  • the first update module 604 further includes: a determination submodule 6045 , a control submodule 6046 , and a third judgment submodule 6047 .
  • the determination sub-module 6045 is used to update the initial boundary position information based on the first position information of the item after the second judgment sub-module 6043 judges that the first position information of the item does not match the initial boundary position information, and obtain the target Boundary position information, and before updating the initial area sub-map corresponding to the target division area according to the target boundary position information, determine the item of the item in the initial area sub-map corresponding to the target division area based on the size information of the item and the first position information of the item Edge position points, wherein the item edge position points of the item are used to indicate the edge position of the item, and there are multiple item edge position points of the item.
  • the item information of the item includes the size information of the item. Through the size information of the item and the first position information, a plurality of edge position points of the item in the initial area sub-map corresponding to the target divided area can be determined.
  • the edge position point of the item may be the vertex of the item, or any point on the edge of the item, which is not limited in this embodiment of the present invention.
  • the control sub-module 6046 is used to control the movement of the cleaning equipment in the area to be tested corresponding to the item, and record the movement track of the cleaning equipment in the initial area sub-map corresponding to the target division area, wherein the area to be tested corresponding to the item contains the item The area of all item edge position points.
  • the area to be tested including all the edge position points of the item can be determined according to all the edge position points of the item.
  • the area to be tested is a virtual area in the initial area map and is not affected by the initial The limit of the initial boundary in the area map. By controlling the cleaning device to move within the area to be tested, it can be determined whether the item is placed at the boundary of the area.
  • the third judging sub-module 6047 is used for judging whether the moving track of the cleaning device is a closed track, and whether the moving track of the cleaning device includes all edge positions of the item.
  • the distribution update sub-module 6042 is also used for if the third judging sub-module judges that the moving track of the cleaning device is a closed track, and the moving track of the cleaning device includes all edge position points of the item, then based on the first position information of the item Determine the distribution sub-area where the item is located in the initial area sub-map corresponding to the target division area, and update the initial area sub-map corresponding to the target division area according to the distribution sub-area.
  • the boundary update sub-module 6044 is also used for if the third judging sub-module judges that the moving track of the cleaning device is not a closed track, or the moving track of the cleaning device does not contain all the edge position points of the item, then based on the first
  • the position information updates the initial boundary position information to obtain the target boundary position information, and updates the initial area sub-map corresponding to the target divided area according to the target boundary position information.
  • the third judging sub-module 6047 judges that the moving track of the cleaning equipment is a closed track, and at the same time, the closed track contains all the edge position points of the item, that is, the item is not placed on the boundary position, then the item can be used as the corrected initial area Items in the sub-map distribution area, and then through the distribution update sub-module 6042, the distribution sub-area where the item is located in the initial area sub-map corresponding to the target division area is determined based on the first position information of the item, and the target division is updated according to the distribution sub-area The initial region submap corresponding to the region.
  • the item can be used as Correct the item at the boundary position of the initial area submap, and then update the initial boundary position information based on the first position information of the item through the boundary update sub-module 6044, obtain the target boundary position information, and update the initial boundary position corresponding to the target division area according to the target boundary position information.
  • Region submap
  • the distribution update submodule 6042 determines the distribution sub-area where the item is located in the initial area sub-map corresponding to the target division area based on the first position information of the item, and The methods of updating the initial region submap corresponding to the target division region include:
  • the current distribution sub-area where the item is located in the initial area sub-map corresponding to the target division area is determined.
  • the initial distribution sub-area corresponding to the item is obtained, and it is judged whether the initial distribution sub-area corresponding to the item matches the current distribution sub-area where the item is located.
  • update the initial distribution sub-area corresponding to the item based on the current distribution sub-area where the item is located obtain the target distribution sub-area of the item, and update the initial area sub-map corresponding to the target division area according to the target distribution sub-area of the item.
  • the item information of the item includes the size information of the item. Through the size information and the first position information of any item, it can be determined that the item is located in the initial area sub-map corresponding to the target division area.
  • the current distribution sub-area where the item is located is the current distribution of the item in the target division area.
  • the initial area sub-map corresponding to the target division area it is judged whether there is an initial distribution sub-area corresponding to the item. If there is no initial distribution sub-area corresponding to the item in the initial area sub-map corresponding to the target division area, it means that the item is a newly added item, and the current distribution sub-area of the item can be added to the initial area sub-map. If there is an initial distribution sub-area corresponding to the item in the initial area sub-map corresponding to the target division area, it means that the item is not a new item, and you can judge whether it is necessary by comparing the matching between the initial distribution sub-area and the current distribution sub-area Update the initial region submap.
  • the initial area sub-map needs to be updated. If the initial distribution subregion matches the current distribution subregion, the subregion the item is in in the initial region submap does not need to be corrected. If the initial distribution sub-area does not match the current distribution sub-area, the item's sub-area in the initial area sub-map has changed and needs to be updated by correction.
  • the current distribution sub-area where the item is located in the initial area sub-map updates the initial distribution sub-area corresponding to the item to obtain the target distribution sub-area of the item, which can determine the distribution of the item in the target division area, and then obtain the updated target The initial region submap corresponding to the divided region.
  • the distribution of items in the target division area can be determined according to the current distribution sub-area where the item is located, and further the corresponding distribution of the target division area can be updated through the current distribution sub-area where the item The distribution area of the initial area sub-map, and then obtain a more accurate area sub-map of the target division area.
  • FIG. 8 is a schematic structural diagram of another intelligent region division device disclosed in an embodiment of the present invention.
  • the device also includes:
  • the first determining module 606 is configured to determine the initial boundary position information corresponding to the item in the initial area submap corresponding to the target divided area based on the item information of the item before updating the initial area submap corresponding to the target divided area.
  • the first determination module 606 determines the initial boundary position information corresponding to the item in the initial area sub-map corresponding to the target division area includes:
  • the initial boundary position information corresponding to the item in the initial area sub-map corresponding to the target divided area can be determined through the collection direction when the cleaning device collects the target image of the target divided area.
  • determining the collection direction when the cleaning device collects the target image of the target divided area may be determined by a direction sensor, or may be determined by an image recognition technology.
  • the distribution of items in the target division area can be determined according to the current distribution sub-area where the item is located, and further the corresponding distribution of the target division area can be updated through the current distribution sub-area where the item The distribution area of the initial area sub-map, and then obtain a more accurate area sub-map of the target division area.
  • the analyzing module 603 is also used for:
  • the first update module 604 is also used to update the initial area submap corresponding to the target division area based on the item information of the item if the analysis module 603 judges that the item recognition probability of the item is greater than or equal to the preset threshold value, and obtain the target The operation of the sub-map of the target area corresponding to the divided area;
  • the acquisition module 602 is further configured to adjust the shooting angle of the cleaning equipment if the analysis module 603 judges that the item recognition probability of the item is less than the preset threshold, and perform acquisition of target images based on the target division area of the cleaning equipment, and analyze the target images , the operation of obtaining the items contained in the target image and the item information of the items.
  • the item recognition probability is used to indicate the accuracy of the recognition result.
  • the item recognition probability can be determined by the degree of matching between the current item and the standard item.
  • the item identification probability exceeds the preset threshold, it indicates that the current identification result is accurate, and subsequent operations can be performed on the identified item and item information.
  • the item recognition probability does not exceed the preset threshold, it indicates that the current recognition result is inaccurate, and it is necessary to adjust the shooting angle of the acquisition device, re-collect the image of the target division area, and re-identify to obtain an accurate recognition result.
  • the accuracy of the recognition result can be judged, and an inaccurate recognition result can be adaptively changed for re-recognition to prevent misjudgment. While ensuring the accuracy of item identification results, it also ensures the accuracy of subsequent intelligent division of areas.
  • the device further includes: a second determining module 607, configured to:
  • the target divided area includes one or more items, and all items included in the target divided area can be acquired through a single, multiple or panoramic target images. And the category information of each item included in the target division area is identified through the image recognition technology. If the target division area includes item 1, item 2, and item 3, it is recognized that the category information of item 1 is a dining table, the category information of item 2 is coffee table, and the category information of item 3 is sofa.
  • the target division area It is judged whether all the items in the target division area include items whose category information is the area-restricted category, and the area-restricted category is used to indicate that the probability that the corresponding item belongs to a determined area is greater than or equal to the preset probability threshold.
  • the region-restricted category is used to indicate that the probability that the corresponding item belongs to a certain determined region is greater than or equal to a preset probability threshold, which can be set according to the needs of actual application scenarios.
  • a preset probability threshold which can be set according to the needs of actual application scenarios.
  • the category information of the item 4 is a range hood
  • the range hood category information may be set as an area-restricted category, which means that the probability that the item 4 belongs to the kitchen is greater than or equal to the preset probability threshold.
  • all the items in the target division area include items whose category information is the area-restricted category, then obtain the target item from all the items in the target division area, and determine the area attribute corresponding to the target item as the attribute of the target division area, where , the target item is an item whose category information is a regionally restricted category.
  • all the items in the target division area include items whose category information is the area-restricted category
  • the items whose category information is the area-restricted category are acquired, for example, all items in the target division area include item 4( If the category information is range hood), item 4 is taken as the target item, and the area attribute (kitchen) corresponding to item 4 is determined as the attribute (kitchen) of the target divided area.
  • the intersection area attribute of all items in the target divided area is determined, and the intersection area attribute is determined as the attribute of the target divided area.
  • the attribute of the target division area needs to be determined through the area attributes of multiple items in the target division area.
  • the area attributes of item 1 in the target division area are dining room and living room
  • the area attributes of item 2 are living room
  • the area attributes of item 3 are living room and bedroom.
  • the intersection area attributes of all items in the target division area the intersection area attribute set of the target division area can be obtained, and the area attributes in the intersection area attribute set can be used as candidate attributes of the target division area. For example, if the attribute of the intersection area of all items in the target division area is the living room, then the living room is determined as the attribute of the target division area.
  • the attribute of the target divided area is determined as the attribute of the target area submap corresponding to the target divided area.
  • the attribute of the target divided area is determined as the attribute of the target area submap corresponding to the target divided area, and displayed in the target area submap of the target divided area.
  • the attribute of the target division area can be determined based on the items in the target area, and the intelligent area division result can be accurately obtained.
  • the second determination module 607 is further configured to: after determining the intersection area attributes of all items in the target divided area, before determining the intersection area attribute as the attribute of the target divided area :
  • the attribute of the target intersection area with the highest priority among the attributes of the intersection area is obtained according to the priority of the preset area attributes, and the attribute of the target intersection area is determined as the attribute of the target division area.
  • the intersection area attribute may be determined as the attribute of the target division area. If there are multiple intersection area attributes in the intersection area attribute set, the target intersection area attribute with the highest priority can be determined by presetting the area attribute priority, and the target intersection area attribute with the highest priority can be determined as the attribute of the target division area.
  • the region attribute priority can be set differently according to different target scenarios.
  • other intersection area attributes in the intersection area attribute set except the target intersection area attribute are used as candidate attributes of the target division area for selection by the operator.
  • the second determining module 607 is further configured to: if it is determined that all items in the target division area do not include items whose category information is a category restricted category, obtain the items in the target division area. After the area attributes corresponding to each item:
  • the area attribute set of the target divided area is obtained according to the area attribute corresponding to each item in the target divided area.
  • the number of items including the area attribute is counted to obtain the frequency of the area attribute.
  • the attributes of the target area sub-map corresponding to the target area can be determined through the common area attributes of multiple items in the target area, and the result of area intelligent division can be accurately obtained.
  • FIG. 9 is a schematic structural diagram of another device for intelligently dividing regions disclosed in an embodiment of the present invention. As shown in Figure 9, the intelligent division device in this area may include:
  • the memory 901 stores executable program codes.
  • Processor 902 coupled with memory 901.
  • the processor 902 invokes the executable program code stored in the memory 901 to execute the steps in the method for intelligently dividing areas described in Embodiment 1 or Embodiment 2 of the present invention.
  • the embodiment of the present invention discloses a cleaning device, which is used for implementing the method for intelligently dividing areas described in the first embodiment of the present invention or the second embodiment of the present invention.
  • the embodiment of the present invention discloses a computer storage medium, the computer storage medium stores computer instructions, and when the computer instructions are called, it is used to perform the intelligent division of the regions described in the first embodiment of the present invention or the second embodiment of the present invention steps in the method.
  • the embodiment of the present invention discloses a computer program product.
  • the computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable the computer to execute the computer program described in the first or second embodiment. The steps in the method for intelligent partitioning of regions are described.
  • the device embodiments described above are only illustrative, and the modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in One place, or it can be distributed to multiple network modules. Part or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. It can be understood and implemented by those skilled in the art without any creative efforts.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Library & Information Science (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computing Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Automation & Control Theory (AREA)
  • Remote Sensing (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Image Analysis (AREA)

Abstract

一种区域的智能划分方法及装置，包括：获取针对任一目标场景的初始区域地图，目标场景包括多个划分区域，初始区域地图包括每个划分区域对应的初始区域子地图（S1）；基于清扫设备采集目标划分区域的目标图像，目标划分区域为所有划分区域中的任一区域（S2）；分析目标图像，得到目标图像中包含的物品、以及物品的物品信息（S3）；基于物品的物品信息，更新目标划分区域对应的初始区域子地图，得到目标划分区域对应的目标区域子地图（S4），进而实现目标场景的初始区域地图的更新。可见，该方法能够基于目标区域内的物品实现区域的智能划分，在避免人工设置标记物的同时能够提高区域智能划分结果的准确性。

Description

区域的智能划分方法及装置 技术领域
本发明涉及智能设备领域，尤其涉及一种区域的智能划分方法及装置。
背景技术
为了实现清扫设备在各个区域的移动及清扫，通常需要对区域进行智能的划分并构建相应的区域地图。目前，现有技术主要通过人工在区域边界上设置标记物或参照物，如在房间门框上人为的设置标记物，根据人工设置的标记物或参照物识别出区域的边界，进而实现区域的划分。可见，现有技术依赖于人工标记，操作繁杂，效率低；同时，当区域中没有预先在区域边界上设置的标记物时，无法准确的识别各个区域的边界，进而无法得到准确的区域划分结果。因此，如何提高区域智能划分结果的准确性，是本领域尚待解决的技术问题。
发明内容
本发明所要解决的技术问题在于如何提高区域智能划分结果的准确性，提供一种区域的智能划分方法及装置，能够通过识别区域内的物品，简单高效的实现区域的智能划分。
为了解决上述技术问题，本发明第一方面公开了一种区域的智能划分方法，所述方法包括：
获取针对任一目标场景的初始区域地图，所述目标场景包括多个划分区域，所述初始区域地图包括每个所述划分区域对应的初始区域子地图；
基于清扫设备采集目标划分区域的目标图像，所述目标划分区域为所有所述划分区域中的任一区域；
分析所述目标图像，得到所述目标图像中包含的物品、以及所述物品的物品信息；
基于所述物品的物品信息，更新所述目标划分区域对应的初始区域子地图，得到所述目标划分区域对应的目标区域子地图；
根据所述目标划分区域对应的目标区域子地图，更新所述目标场景的初始区域地图。
作为一种可选的实施方式，在本发明第一方面中，所述物品的物品信息包括所述物品的类型信息以及所述物品的第一位置信息，其中，所述物品的第一位置信息为所述物品在所述目标划分区域对应的初始区域子地图中的位置信息；
所述基于所述物品的物品信息，更新所述目标划分区域对应的初始区域子地图，包括：
根据所述物品的类型信息，判断所述物品是否为边界类型物品；
若判断出所述物品为非边界类型物品，则基于所述物品的第一位置信息确 定在所述目标划分区域对应的初始区域子地图中所述物品所处的分布子区域，并根据所述分布子区域更新所述目标划分区域对应的初始区域子地图，其中，所述物品所处的分布子区域用于表示所述物品在所述目标划分区域的分布情况；
若判断出所述物品为边界类型物品，则判断所述物品的第一位置信息是否与初始边界位置信息相匹配，其中，所述初始边界位置信息为所述目标划分区域对应的初始区域子地图中与所述物品对应的初始边界位置信息；
若判断出所述物品的第一位置信息与所述初始边界位置信息不匹配，则基于所述物品的第一位置信息更新所述初始边界位置信息，得到目标边界位置信息，并根据所述目标边界位置信息更新所述目标划分区域对应的初始区域子地图。
作为一种可选的实施方式，在本发明第一方面中，所述物品的物品信息还包括所述物品的尺寸信息，其中，所述物品的尺寸信息为所述物品在所述目标划分区域对应的初始区域子地图中的尺寸信息；
在判断出所述物品的第一位置信息与所述初始边界位置信息不匹配之后，所述基于所述物品的第一位置信息更新所述初始边界位置信息，得到目标边界位置信息，并根据所述目标边界位置信息更新所述目标划分区域对应的初始区域子地图之前，所述方法还包括：
基于所述物品的尺寸信息及所述物品的第一位置信息，确定在所述目标划分区域对应的初始区域子地图中所述物品的物品边缘位置点，其中，所述物品的物品边缘位置点用于表示所述物品的边缘位置，所述物品的物品边缘位置点为多个；
控制所述清扫设备在所述物品对应的待测区域内移动，并在所述目标划分区域对应的初始区域子地图中记录所述清扫设备的移动轨迹，其中，所述物品对应的待测区域为包含所述物品的所有物品边缘位置点的区域；
判断所述清扫设备的移动轨迹是否为封闭轨迹，以及所述清扫设备的移动轨迹中是否包含所述物品的所有物品边缘位置点；
若所述清扫设备的移动轨迹为封闭轨迹，且所述清扫设备的移动轨迹中包含所述物品的所有物品边缘位置点，则基于所述物品的第一位置信息确定在所述目标划分区域对应的初始区域子地图中所述物品所处的分布子区域，并根据所述分布子区域更新所述目标划分区域对应的初始区域子地图；
若所述清扫设备的移动轨迹不为封闭轨迹，或所述清扫设备的移动轨迹中不包含所述物品的所有物品边缘位置点，则执行所述基于所述物品的第一位置信息更新所述初始边界位置信息，得到目标边界位置信息，并根据所述目标边界位置信息更新所述目标划分区域对应的初始区域子地图的操作。
作为一种可选的实施方式，在本发明第一方面中，所述物品的物品信息还包括所述物品的尺寸信息，其中，所述物品的尺寸信息为所述物品在所述目标划分区域对应的初始区域子地图中的尺寸信息；
所述基于所述物品的第一位置信息确定在所述目标划分区域对应的初始区 域子地图中所述物品所处的分布子区域，并根据所述分布子区域更新所述目标划分区域对应的初始区域子地图，包括：
根据所述物品的第一位置信息以及所述物品的尺寸信息，确定在所述目标划分区域对应的初始区域子地图中所述物品所处的当前分布子区域；
判断在所述目标划分区域对应的初始区域子地图中是否存在所述物品对应的初始分布子区域；
若不存在，则根据所述物品所处的当前分布子区域更新所述目标划分区域对应的初始区域子地图；
若存在，则获取所述物品对应的初始分布子区域，并判断所述物品对应的初始分布子区域与所述物品所处的当前分布子区域是否匹配；
若不匹配，则基于所述物品所处的当前分布子区域更新所述物品对应的初始分布子区域，得到所述物品的目标分布子区域，并根据所述物品的目标分布子区域更新所述目标划分区域对应的初始区域子地图。
作为一种可选的实施方式，在本发明第一方面中，所述基于所述物品的物品信息，更新所述目标划分区域对应的初始区域子地图之前，所述方法还包括：
确定所述目标划分区域对应的初始区域子地图中与所述物品对应的初始边界位置信息；
其中，所述确定所述目标划分区域对应的初始区域子地图中与所述物品对应的初始边界位置信息，包括：
确定所述清扫设备采集所述目标划分区域的目标图像时的采集方向；
获取所述目标划分区域对应的初始区域子地图中与所述清扫设备采集所述目标划分区域的目标图像时的采集方向相匹配的初始边界，将所述初始边界对应的位置信息确定为所述目标划分区域对应的初始区域子地图中与所述物品对应的初始边界位置信息。
作为一种可选的实施方式，在本发明第一方面中，所述方法还包括：
获取所述目标划分区域中的所有物品以及每个所述物品所属的品类信息；
判断所述目标划分区域中的所有物品中是否包括品类信息为区域限制品类的物品，所述区域限制品类用于表示对应的物品属于某一确定出的区域的概率大于等于预设概率阈值；
若所述目标划分区域中的所有物品中包括品类信息为所述区域限制品类的物品，则从所述目标划分区域中的所有物品中获取目标物品，并将所述目标物品对应的区域属性确定为所述目标划分区域的属性，其中，所述目标物品为品类信息为所述区域限制品类的物品；
若所述目标划分区域中的所有物品中不包括品类信息为所述区域限制品类的物品，则获取所述目标划分区域中每个所述物品对应的区域属性；
根据所述目标划分区域中每个所述物品对应的区域属性，确定所述目标划分区域中的所有物品的交集区域属性，将所述交集区域属性确定为所述目标划分区域的属性；
将所述目标划分区域的属性确定为所述目标划分区域对应的目标区域子地 图的属性。
作为一种可选的实施方式，在本发明第一方面中，所述确定所述目标划分区域中的所有物品的交集区域属性之后，所述将所述交集区域属性确定为所述目标划分区域的属性之前，所述方法还包括：
判断所述交集区域属性是否唯一；
若所述交集区域属性唯一，则执行所述将所述交集区域属性确定为所述目标划分区域的属性的操作；
若所述交集区域属性不唯一，则根据预设区域属性优先级获取所述交集区域属性中优先级最高的目标交集区域属性，将所述目标交集区域属性确定为所述目标划分区域的属性。
作为一种可选的实施方式，在本发明第一方面中，所述分析所述目标图像，得到所述目标图像中包含的物品、以及所述物品的物品信息之后，所述基于所述物品的物品信息，更新所述目标划分区域对应的初始区域子地图，得到所述目标划分区域对应的目标区域子地图之前，所述方法还包括：
确定所述物品的物品识别概率；
判断所述物品的物品识别概率是否大于或等于预设阈值；
若所述物品的物品识别概率大于或等于预设阈值，则执行所述基于所述物品的物品信息，更新所述目标划分区域对应的初始区域子地图，得到所述目标划分区域对应的目标区域子地图的操作；
若所述物品的物品识别概率小于预设阈值，则调整所述清扫设备的拍摄角度，并执行所述基于清扫设备采集目标划分区域的目标图像，以及所述分析所述目标图像，得到所述目标图像中包含的物品、以及所述物品的物品信息的操作。
本发明第二方面公开了一种区域的智能划分装置，所述装置包括：
获取模块，用于获取针对任一目标场景的初始区域地图，所述目标场景包括多个划分区域，所述初始区域地图包括每个所述划分区域对应的初始区域子地图；
采集模块，用于采集目标划分区域的目标图像，所述目标划分区域为所有所述划分区域中的任一区域；
分析模块，用于分析所述目标图像，得到所述目标图像中包含的物品、以及所述物品的物品信息；
第一更新模块，用于基于所述物品的物品信息，更新所述目标划分区域对应的初始区域子地图，得到所述目标划分区域对应的目标区域子地图；
第二更新模块，用于根据所述目标划分区域对应的目标区域子地图，更新所述目标场景的初始区域地图。
作为一种可选的实施方式，在本发明第二方面中，所述物品的物品信息包括所述物品的类型信息以及所述物品的第一位置信息，其中，所述物品的第一位置信息为所述物品在所述目标划分区域对应的初始区域子地图中的位置信息；
所述第一更新模块包括：
第一判断子模块，用于根据所述物品的类型信息，判断所述物品是否为边界类型物品；
分布更新子模块，用于若所述第一判断子模块判断出所述物品为非边界类型物品，则基于所述物品的第一位置信息确定在所述目标划分区域对应的初始区域子地图中所述物品所处的分布子区域，并根据所述分布子区域更新所述目标划分区域对应的初始区域子地图，其中，所述物品所处的分布子区域用于表示所述物品在所述目标划分区域的分布情况；
第二判断子模块，用于若所述第一判断子模块判断出所述物品为边界类型物品，则判断所述物品的第一位置信息是否与初始边界位置信息相匹配，其中，所述初始边界位置信息为所述目标划分区域对应的初始区域子地图中与所述物品对应的初始边界位置信息；
边界更新子模块，用于若所述第二判断子模块判断出所述物品的第一位置信息与所述初始边界位置信息不匹配，则基于所述物品的第一位置信息更新所述初始边界位置信息，得到目标边界位置信息，并根据所述目标边界位置信息更新所述目标划分区域对应的初始区域子地图。
作为一种可选的实施方式，在本发明第二方面中，所述物品的物品信息还包括所述物品的尺寸信息，其中，所述物品的尺寸信息为所述物品在所述目标划分区域对应的初始区域子地图中的尺寸信息；
所述第一更新模块还包括：
确定子模块，用于在所述第二判断子模块判断出所述物品的第一位置信息与所述初始边界位置信息不匹配之后，所述边界更新子模块基于所述物品的第一位置信息更新所述初始边界位置信息，得到目标边界位置信息，并根据所述目标边界位置信息更新所述目标划分区域对应的初始区域子地图之前，基于所述物品的尺寸信息及所述物品的第一位置信息，确定在所述目标划分区域对应的初始区域子地图中所述物品的物品边缘位置点，其中，所述物品的物品边缘位置点用于表示所述物品的边缘位置，所述物品的物品边缘位置点为多个；
控制子模块，用于控制清扫设备在所述物品对应的待测区域内移动，并在所述目标划分区域对应的初始区域子地图中记录所述清扫设备的移动轨迹，其中，所述物品对应的待测区域为包含所述物品的所有物品边缘位置点的区域；
第三判断子模块，用于判断所述清扫设备的移动轨迹是否为封闭轨迹，以及所述清扫设备的移动轨迹中是否包含所述物品的所有物品边缘位置点；
所述分布更新子模块，还用于若所述第三判断子模块判断出所述清扫设备的移动轨迹为封闭轨迹，且所述清扫设备的移动轨迹中包含所述物品的所有物品边缘位置点，则基于所述物品的第一位置信息确定在所述目标划分区域对应的初始区域子地图中所述物品所处的分布子区域，并根据所述分布子区域更新所述目标划分区域对应的初始区域子地图；
所述边界更新子模块，还用于若所述第三判断子模块判断出所述清扫设备的移动轨迹不为封闭轨迹，或所述清扫设备的移动轨迹中不包含所述物品的所 有物品边缘位置点，则基于所述物品的第一位置信息更新所述初始边界位置信息，得到目标边界位置信息，并根据所述目标边界位置信息更新所述目标划分区域对应的初始区域子地图。
作为一种可选的实施方式，在本发明第二方面中，所述物品的物品信息还包括所述物品的尺寸信息，其中，所述物品的尺寸信息为所述物品在所述目标划分区域对应的初始区域子地图中的尺寸信息；
所述分布更新子模块基于所述物品的第一位置信息确定在所述目标划分区域对应的初始区域子地图中所述物品所处的分布子区域，并根据所述分布子区域更新所述目标划分区域对应的初始区域子地图的方式包括：
根据所述物品的第一位置信息以及所述物品的尺寸信息，确定在所述目标划分区域对应的初始区域子地图中所述物品所处的当前分布子区域；
判断在所述目标划分区域对应的初始区域子地图中是否存在所述物品对应的初始分布子区域；
若不存在，则根据所述物品所处的当前分布子区域更新所述目标划分区域对应的初始区域子地图；
若存在，则获取所述物品对应的初始分布子区域，并判断所述物品对应的初始分布子区域与所述物品所处的当前分布子区域是否匹配；
若不匹配，则基于所述物品所处的当前分布子区域更新所述物品对应的初始分布子区域，得到所述物品的目标分布子区域，并根据所述物品的目标分布子区域更新所述目标划分区域对应的初始区域子地图。
作为一种可选的实施方式，在本发明第二方面中，所述装置还包括：
第一确定模块，用于所述基于所述物品的物品信息，更新所述目标划分区域对应的初始区域子地图之前，确定所述目标划分区域对应的初始区域子地图中与所述物品对应的初始边界位置信息；
其中，所述第一确定模块确定所述目标划分区域对应的初始区域子地图中与所述物品对应的初始边界位置信息的方式包括：
确定所述清扫设备采集所述目标划分区域的目标图像时的采集方向；
获取所述目标划分区域对应的初始区域子地图中与所述清扫设备采集所述目标划分区域的目标图像时的采集方向相匹配的初始边界，将所述初始边界对应的位置信息确定为所述目标划分区域对应的初始区域子地图中与所述物品对应的初始边界位置信息。
作为一种可选的实施方式，在本发明第二方面中，所述装置还包括：第二确定模块，用于：
获取所述目标划分区域中的所有物品以及每个所述物品所属的品类信息；
判断所述目标划分区域中的所有物品中是否包括品类信息为区域限制品类的物品，所述区域限制品类用于表示对应的物品属于某一确定出的区域的概率大于等于预设概率阈值；
若所述目标划分区域中的所有物品中包括品类信息为所述区域限制品类的物品，则从所述目标划分区域中的所有物品中获取目标物品，并将所述目标物 品对应的区域属性确定为所述目标划分区域的属性，其中，所述目标物品为品类信息为所述区域限制品类的物品；
若所述目标划分区域中的所有物品中不包括品类信息为所述区域限制品类的物品，则获取所述目标划分区域中每个所述物品对应的区域属性；
根据所述目标划分区域中每个所述物品对应的区域属性，确定所述目标划分区域中的所有物品的交集区域属性，将所述交集区域属性确定为所述目标划分区域的属性；
将所述目标划分区域的属性确定为所述目标划分区域对应的目标区域子地图的属性。
作为一种可选的实施方式，在本发明第二方面中，所述第二确定模块还用于：在所述确定所述目标划分区域中的所有物品的交集区域属性之后，所述将所述交集区域属性确定为所述目标划分区域的属性之前，判断所述交集区域属性是否唯一；
若所述交集区域属性唯一，则执行所述将所述交集区域属性确定为所述目标划分区域的属性的操作；
若所述交集区域属性不唯一，则根据预设区域属性优先级获取所述交集区域属性中优先级最高的目标交集区域属性，将所述目标交集区域属性确定为所述目标划分区域的属性。
作为一种可选的实施方式，在本发明第二方面中，所述分析模块还用于：
所述分析所述目标图像，得到所述目标图像中包含的物品、以及所述物品的物品信息之后，确定所述物品的物品识别概率；判断所述物品的物品识别概率是否大于或等于预设阈值；
所述第一更新模块，还用于若所述分析模块判断出所述物品的物品识别概率大于或等于预设阈值，则执行所述基于所述物品的物品信息，更新所述目标划分区域对应的初始区域子地图，得到所述目标划分区域对应的目标区域子地图的操作；
所述采集模块，还用于若所述分析模块判断出所述物品的物品识别概率小于预设阈值，则调整所述清扫设备的拍摄角度，并执行所述基于清扫设备采集目标划分区域的目标图像，以及所述分析所述目标图像，得到所述目标图像中包含的物品、以及所述物品的物品信息的操作。
本发明第三方面公开了另一种区域的智能划分装置，所述装置包括：
存储有可执行程序代码的存储器；
与所述存储器耦合的处理器；
所述处理器调用所述存储器中存储的所述可执行程序代码，执行本发明第一方面公开的区域的智能划分方法。
本发明第四方面公开了一种清扫设备，所述清扫设备用于执行本发明第一方面公开的区域的智能划分方法。
本发明第五方面公开了一种计算机可存储介质，所述计算机存储介质存储有计算机指令，所述计算机指令被调用时，用于执行本发明第一方面公开的区 域的智能划分方法。
与现有技术相比，本发明实施例具有以下有益效果：
本发明实施例中，获取针对任一目标场景的初始区域地图，目标场景包括多个划分区域，初始区域地图包括每个划分区域对应的初始区域子地图；基于清扫设备采集目标划分区域的目标图像，目标划分区域为所有划分区域中的任一区域；分析目标图像，得到目标图像中包含的物品、以及物品的物品信息；基于物品的物品信息，更新目标划分区域对应的初始区域子地图，得到目标划分区域对应的目标区域子地图，进而实现目标场景的初始区域地图的更新。可见，实施本发明能够基于目标区域内的物品实现区域的智能划分，在避免人工设置标记物的同时能够提高区域智能划分结果的准确性。
附图说明
为了更清楚地说明本发明实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本发明实施例公开的一种区域的智能划分方法的流程示意图；
图2是本发明实施例公开的基于物品的物品信息更新目标划分区域对应的初始区域子地图的流程示意图；
图3是本发明实施例公开的基于物品所处的分布子区域更新目标划分区域对应的初始区域子地图的流程示意图；
图4是本发明实施例公开的又一种区域的智能划分方法的流程示意图；
图5是本发明实施例公开的确定目标划分区域对应的目标区域子地图的属性的流程示意图；
图6是本发明实施例公开的一种区域的智能划分装置的结构示意图；
图7是本发明实施例公开的第一更新模块的结构示意图；
图8是本发明实施例公开的又一种区域的智能划分装置的结构示意图；
图9是本发明实施例公开的又一种区域的智能划分装置的结构示意图。
具体实施方式
为了使本技术领域的人员更好地理解本发明方案，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
本发明的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别不同对象，而不是用于描述特定顺序。此外，术语“包括”和“具有”以及它们任何变形，意图在于覆盖不排他的包含。例如包含了一系列步骤或单元的过程、方法、装置、产品或端没有限定于已列出的步骤或单元，而是可选地还包括没 有列出的步骤或单元，或可选地还包括对于这些过程、方法、产品或端固有的其他步骤或单元。
在本文中提及“实施例”意味着，结合实施例描述的特定特征、结构或特性可以包含在本发明的至少一个实施例中。在说明书中的各个位置出现该短语并不一定均是指相同的实施例，也不是与其它实施例互斥的独立的或备选的实施例。本领域技术人员显式地和隐式地理解的是，本文所描述的实施例可以与其它实施例相结合。
本发明公开了一种区域的智能划分方法及装置、清扫设备，能够基于目标区域内的物品实现区域的智能划分，提高区域智能划分结果的准确性。以下分别进行详细说明。
实施例一
请参阅图1，图1是本发明实施例公开的一种区域的智能划分方法的流程示意图。其中，图1所描述的区域的智能划分方法可以应用于智能清扫设备中，也可以应用于服务器，本发明实施例不做限定。如图1所示，该区域的智能划分方法可以包括以下操作：
S1、获取针对任一目标场景的初始区域地图，目标场景包括多个划分区域，初始区域地图包括每个划分区域对应的初始区域子地图。
本发明实施例中，执行区域智能划分之前，需要获取目标场景的初始区域地图作为基准区域地图进行区域边界的智能划分以及区域分布的智能划分。可选的，目标场景的初始区域地图可以通过激光雷达扫描目标场景得到，可以基于人工设置的边界标记物得到，也可以通过下载预先存储在数据库中的场景初始地图得到，本发明实施例不做限定。目标场景具体为有区域划分需求的场景，可以是包括多个划分区域的室内场景，也可以是有区域划分需求的其他场景。获取的目标场景包括多个划分区域，相应的，获取的目标场景的初始区域地图中包括每个划分区域对应的初始区域子地图。
S2、基于清扫设备采集目标划分区域的目标图像，目标划分区域为所有划分区域中的任一区域。
本发明实施例中，清扫设备包括采集装置，可选的，该清扫设备包括采集装置可以是单目摄像装置、多目摄像装置，也可以是包括云台的摄像装置，本发明实施例不做限定。通过清扫设备可以采集目标场景的多个划分区域中的任意一个目标划分区域的目标图像，采集到的目标图像可以为目标划分区域的单张图像、目标划分区域不同方向不同角度的多张图像、目标划分区域的全景图像中的一种或多种。
S3、分析目标图像，得到目标图像中包含的物品、以及物品的物品信息。
本发明实施例中，清扫设备采集到的目标图像中包括目标划分区域中的物品。目标图像中包括的目标划分区域中的物品可以为一个或多个。通过分析采集到的目标图像可以识别出目标图像中包括的物品，以及物品的物品信息。可选的，物品的物品信息可以包括物品的类型信息以及物品的第一位置信息，还可以包括物品的尺寸信息。物品的类型信息可以是边界类型或非边界类型，示 例性的，如分析出目标图像包括门或门框、茶几或餐桌，进一步可以得到门或门框的类型信息为边界类型，茶几或餐桌的类型信息为非边界类型。同时，物品的第一位置信息为物品在目标划分区域对应的初始区域子地图中的位置信息，通过物品的第一位置信息可以在初始区域子地图中定位到该物品。另外，物品的尺寸信息为物品在目标划分区域对应的初始区域子地图中的尺寸信息，通过物品的尺寸信息可以在初始区域子地图中标注出物品所占的空间区域尺寸。
S4、基于物品的物品信息，更新目标划分区域对应的初始区域子地图，得到目标划分区域对应的目标区域子地图。
本发明实施例中，基于分析得到的物品的物品信息可以对初始区域子地图的区域边界或区域分布进行更新，进而得到智能划分后的目标划分区域对应的目标区域子地图。
S5、根据目标划分区域对应的目标区域子地图，更新目标场景的初始区域地图。
本发明实施例中，通过物品的物品信息更新目标划分区域对应的初始区域子地图，得到目标划分区域对应的目标区域子地图之后，相应的可以实现目标场景的区域地图的更新。
可见，实施如图1所描述的区域的智能划分能够基于目标区域内的物品实现区域的智能划分，在避免人工设置标记物的同时提高区域智能划分结果的准确性。
在一个可选的实施例中，请参阅图2，图2是本发明实施例公开的基于物品的物品信息更新目标划分区域对应的初始区域子地图的流程示意图，步骤S4基于物品的物品信息更新目标划分区域对应的初始区域子地图可以包括以下步骤：
S401、根据物品的类型信息，判断物品是否为边界类型物品。若判断出物品为非边界类型物品，则执行步骤S402；若判断出物品为边界类型物品，则执行步骤S403-S404。
在该可选的实施例中，根据步骤S3识别得到的物品的类型信息判断物品是否为边界类型物品。当步骤S401判断出物品是边界类型物品时，则可以根据该边界类型物品修正初始区域子地图边界；当步骤S401判断出物品是非边界类型物品时，则该物品不能作为修正初始区域子地图边界的物品，而可以作为修正初始区域子地图分布区域的物品。
S402、基于物品的第一位置信息确定在目标划分区域对应的初始区域子地图中物品所处的分布子区域，并根据分布子区域更新目标划分区域对应的初始区域子地图，其中，物品所处的分布子区域用于表示物品在目标划分区域的分布情况。
在该可选的实施例中，物品是非边界类型物品时，可以根据该物品修正初始区域子地图分布区域。具体的，可以根据物品在目标划分区域对应的初始区域子地图中物品所处的分布子区域，确定物品在目标划分区域的分布情况。
S403、判断物品的第一位置信息是否与初始边界位置信息相匹配，其中，初始边界位置信息为目标划分区域对应的初始区域子地图中与物品对应的初始边界位置信息。若判断出物品的第一位置信息与初始边界位置信息不匹配，则执行步骤S404。
在该可选的实施例中，当判断出物品的第一位置信息与初始边界位置信息匹配时，则目标划分区域对应的初始区域子地图中与物品对应的初始边界位置信息准确，不需要通过物品的位置信息进行修正。当判断出物品的第一位置信息与初始边界位置信息不匹配时，则目标划分区域对应的初始区域子地图中与物品对应的初始边界位置信息不准确，需要通过识别出的物品的物品信息，对初始边界位置信息进行修正。
S404、基于物品的第一位置信息更新初始边界位置信息，得到目标边界位置信息，并根据目标边界位置信息更新目标划分区域对应的初始区域子地图。
在该可选的实施例中，需要通过识别出的物品的物品信息，对初始边界位置信息进行修正。具体根据物品的第一位置信息修正目标划分区域对应的初始区域子地图中与物品对应的初始边界位置信息，得到修正后的目标边界位置信息，进而根据修正后的目标边界位置信息更新目标划分区域对应的初始区域子地图。
可见，实施该可选的实施例，能够根据物品的类型信息确定初始区域子地图的待更新内容，进一步的可以通过物品的位置信息修正或更新目标划分区域对应的初始区域子地图的边界位置/分布区域，进而得到目标划分区域更准确的区域子地图。
在该可选的实施例中，进一步的，目标场景中需要进一步验证物品类型为边界类型的物品是否被放置于边界位置。也就是在步骤S403判断出物品的第一位置信息与初始边界位置信息不匹配之后，步骤S404基于物品的第一位置信息更新初始边界位置信息，得到目标边界位置信息，并根据目标边界位置信息更新目标划分区域对应的初始区域子地图之前，该方法还包括：
S405、基于物品的尺寸信息及物品的第一位置信息，确定在目标划分区域对应的初始区域子地图中物品的物品边缘位置点，其中，物品的物品边缘位置点用于表示物品的边缘位置，物品的物品边缘位置点为多个。
在该可选的实施例中，物品的物品信息包括物品的尺寸信息，通过物品的尺寸信息及第一位置信息，能够确定出在目标划分区域对应的初始区域子地图中该物品的多个物品边缘位置点。物品的边缘位置点可以为物品的顶点，也可以为物品的边上任意一点，本发明实施例不做限定。
S406、控制清扫设备在物品对应的待测区域内移动，并在目标划分区域对应的初始区域子地图中记录清扫设备的移动轨迹，其中，物品对应的待测区域为包含物品的所有物品边缘位置点的区域。
在该可选的实施例中，根据物品的所有物品边缘位置点可以确定出包含物品的所有物品边缘位置点的待测区域，该待测区域为初始区域地图中的一虚拟区域，不受到初始区域地图中的初始边界的限制。通过控制清扫设备在该待测 区域内移动，可以确定物品是否被放置于区域边界位置。
S407、判断清扫设备的移动轨迹是否为封闭轨迹，以及清扫设备的移动轨迹中是否包含物品的所有物品边缘位置点。若清扫设备的移动轨迹为封闭轨迹，且清扫设备的移动轨迹中包含物品的所有物品边缘位置点，则执行步骤S402；若清扫设备的移动轨迹不为封闭轨迹，或清扫设备的移动轨迹中不包含物品的所有物品边缘位置点，则执行步骤S404。
在该可选的实施例中，当清扫设备的移动轨迹为封闭轨迹，同时该封闭轨迹内包含物品的所有边缘位置点，即该物品未被放置在边界位置上，则该物品可以作为修正初始区域子地图分布区域的物品，进而执行步骤S402基于物品的第一位置信息确定在目标划分区域对应的初始区域子地图中物品所处的分布子区域，并根据分布子区域更新目标划分区域对应的初始区域子地图的操作。
当清扫设备的移动轨迹无法形成封闭轨迹，或清扫设备的移动轨迹内没有包含物品的所有边缘位置点，即该物品被放置在边界位置上，则该物品可以作为修正初始区域子地图边界位置的物品，进而执行步骤S404基于物品的第一位置信息更新初始边界位置信息，得到目标边界位置信息，并根据目标边界位置信息更新目标划分区域对应的初始区域子地图的操作。
可见，实施该可选的实施例，能够通过控制清扫设备的移动验证物品类型为边界类型的物品是否被放置于边界位置，再根据验证结果执行相应的更新操作，进而得到更准确的区域划分结果。
在该可选的实施例中，进一步可选的，请参阅图3，图3是本发明实施例公开的基于物品所处的分布子区域更新目标划分区域对应的初始区域子地图的流程示意图。步骤S402、基于物品的第一位置信息确定在目标划分区域对应的初始区域子地图中物品所处的分布子区域，并根据分布子区域更新目标划分区域对应的初始区域子地图，包括以下步骤：
S4021、根据物品的第一位置信息以及物品的尺寸信息，确定在目标划分区域对应的初始区域子地图中物品所处的当前分布子区域。
在该可选的实施例中，物品的物品信息包括物品的尺寸信息，通过任一物品的尺寸信息及第一位置信息，能够确定出在目标划分区域对应的初始区域子地图中该物品所处的当前分布子区域。物品所处的当前分布子区域是物品在目标划分区域当前的分布情况。
S4022、判断在目标划分区域对应的初始区域子地图中是否存在物品对应的初始分布子区域。若不存在，则执行步骤S4023；若存在，则执行步骤S4024-步骤S4025。
在该可选的实施例中，在目标划分区域对应的初始区域子地图中判断是否存在与该物品对应的初始分布子区域。如果目标划分区域对应的初始区域子地图中不存在与该物品对应的初始分布子区域，则说明该物品为新增物品，可以将该物品的当前分布子区域添加至初始区域子地图中。如果目标划分区域对应的初始区域子地图中存在与该物品对应的初始分布子区域，则说明该物品不为新增物品，可以通过比较初始分布子区域与当前分布子区域的匹配情况判断是 否需要更新初始区域子地图。
S4023、根据物品所处的当前分布子区域更新目标划分区域对应的初始区域子地图。
S4024、获取物品对应的初始分布子区域，并判断物品对应的初始分布子区域与物品所处的当前分布子区域是否匹配。若不匹配，则执行步骤S4025。
在该可选的实施例中，通过比较初始分布子区域与当前分布子区域的匹配情况判断是否需要更新初始区域子地图。如果初始分布子区域与当前分布子区域相匹配，则该物品在初始区域子地图中所处的子区域不需要修正。如果初始分布子区域与当前分布子区域不匹配，则该物品在初始区域子地图中所处的子区域有变动，需要通过修正进行更新。
S4025、基于物品所处的当前分布子区域更新物品对应的初始分布子区域，得到物品的目标分布子区域，并根据物品的目标分布子区域更新目标划分区域对应的初始区域子地图。
在该可选的实施例中，该物品在初始区域子地图中所处的当前分布子区域更新物品对应的初始分布子区域后得到物品的目标分布子区域，可以确定出物品在目标划分区域的分布情况，进而得到更新后的目标划分区域对应的初始区域子地图。
可见，实施该可选的实施例，能够根据物品的所处的当前分布子区域确定确定出物品在目标划分区域的分布情况，进一步的可以通过物品所处的当前分布子区域更新目标划分区域对应的初始区域子地图的分布区域，进而得到目标划分区域更准确的区域子地图。
在又一个可选的实施例中，基于物品的物品信息，更新目标划分区域对应的初始区域子地图之前，该方法还包括：
确定目标划分区域对应的初始区域子地图中与物品对应的初始边界位置信息。
该可选的实施例中，确定目标划分区域对应的初始区域子地图中与物品对应的初始边界位置信息，包括以下步骤：
确定清扫设备采集目标划分区域的目标图像时的采集方向。
获取目标划分区域对应的初始区域子地图中与清扫设备采集目标划分区域的目标图像时的采集方向相匹配的初始边界，将初始边界对应的位置信息确定为目标划分区域对应的初始区域子地图中与物品对应的初始边界位置信息。
在该可选的实施例中，通过清扫设备采集目标划分区域的目标图像时的采集方向，可以确定出目标划分区域对应的初始区域子地图中与物品对应的初始边界位置信息。其中，确定清扫设备采集目标划分区域的目标图像时的采集方向可以通过方向传感器确定，也可以通过图像识别技术确定。
可见，实施该可选的实施例，能够根据物品的所处的当前分布子区域确定确定出物品在目标划分区域的分布情况，进一步的可以通过物品所处的当前分布子区域更新目标划分区域对应的初始区域子地图的分布区域，进而得到目标划分区域更准确的区域子地图。
在又一个可选的实施例中，分析目标图像，得到目标图像中包含的物品、以及物品的物品信息之后，基于所述物品的物品信息，更新目标划分区域对应的初始区域子地图，得到目标划分区域对应的目标区域子地图之前，该方法还包括一下步骤：
确定物品的物品识别概率；
判断物品的物品识别概率是否大于或等于预设阈值；
若物品的物品识别概率大于或等于预设阈值，则执行基于所述物品的物品信息，更新目标划分区域对应的初始区域子地图，得到目标划分区域对应的目标区域子地图的操作；
若物品的物品识别概率小于预设阈值，则调整清扫设备的拍摄角度，并执行基于清扫设备采集目标划分区域的目标图像，以及分析所述目标图像，得到目标图像中包含的物品、以及物品的物品信息的操作。
在该可选的实施例中，识别出目标图像中包括的物品，以及物品的物品信息之后，确定目标图像中包括的物品的物品识别概率。物品识别概率用于表示识别结果的准确程度，可选的，物品识别概率可以通过当前物品与标准物品的匹配程度来确定。当物品识别概率超过预设阈值时，表明当前识别结果准确，可以对识别出的物品以及物品信息进行后续操作。当物品识别概率未超过预设阈值时，表明当前识别结果不准确，需要对采集设备的拍摄角度进行调整，重新对目标划分区域进行图像采集，再次识别，进而得到准确的识别结果。
可见，实施该可选的实施例，能够对识别结果的准确性进行判断，并对不准确的识别结果进行适应性的改变，重新识别，防止误判。保证物品识别结果准确的同时，确保后续区域智能划分的准确性。
实施例二
如图4所示，图4是本发明实施例公开的又一种区域的智能划分方法的流程示意图。该区域的智能划分方法可以包括以下操作：
S1、获取针对任一目标场景的初始区域地图，目标场景包括多个划分区域，初始区域地图包括每个划分区域对应的初始区域子地图。
S2、基于清扫设备采集目标划分区域的目标图像，目标划分区域为所有划分区域中的任一区域。
S3、分析目标图像，得到目标图像中包含的物品、以及物品的物品信息。
S4、基于物品的物品信息，更新目标划分区域对应的初始区域子地图，得到目标划分区域对应的目标区域子地图。
S5、根据目标划分区域对应的目标区域子地图，更新目标场景的初始区域地图。
S6、确定目标划分区域对应的目标区域子地图的属性。
本发明实施例中，针对步骤S1-步骤S5的其它详细描述，请参照实施例一中针对步骤S1-步骤S5的详细描述，本发明实施例不再赘述。需要说明的是，步骤S6可以在步骤S5之前执行，也可以在步骤S5之后执行，本发明实施例不做限定。
本发明实施例中，步骤S6、确定目标划分区域对应的目标区域子地图的属性，包括以下步骤：
S601、获取目标划分区域中的所有物品以及每个物品所属的品类信息。
本发明实施例中，目标划分区域包括一个或多个物品，通过单张、多张或全景目标图像可以获取到目标划分区域中包括的所有物品。并通过图像识别技术识别出目标划分区域中包括的每个物品所属的品类信息。如目标划分区域包括物品1、物品2、物品3，识别出物品1所属的品类信息为餐桌，物品2所属的品类信息为茶几，物品3所属的品类信息为沙发。
S602、判断目标划分区域中的所有物品中是否包括品类信息为区域限制品类的物品，区域限制品类用于表示对应的物品属于某一确定出的区域的概率大于等于预设概率阈值。若目标划分区域中的所有物品中包括品类信息为区域限制品类的物品，则执行步骤S603以及步骤S606；若目标划分区域中的所有物品中不包括品类信息为区域限制品类的物品，则执行步骤S604-步骤S606。
本发明实施例中，区域限制品类用于表示对应的物品属于某一确定出的区域的概率大于等于预设概率阈值，预设概率阈值可以根据实际应用场景的需要进行设置。示例性的，若物品4的品类信息为油烟机，油烟机品类信息可以被设置为区域限制品类，即表示物品4属于厨房的概率大于等于预设概率阈值。
S603、从目标划分区域中的所有物品中获取目标物品，并将目标物品对应的区域属性确定为目标划分区域的属性，其中，目标物品为品类信息为区域限制品类的物品。
本发明实施例中，当目标划分区域的所有物品中包括品类信息为区域限制品类的物品，则获取品类信息为区域限制品类的物品，例如目标划分区域的所有物品中包括物品4(品类信息为油烟机)，则将物品4作为目标物品，并将物品4对应的区域属性(厨房)确定为目标划分区域的属性(厨房)。
S604、获取目标划分区域中每个物品对应的区域属性。
本发明实施例中，当目标划分区域的所有物品中不包括品类信息为区域限制品类的物品，则需通过目标划分区域中的多个物品的区域属性确定目标划分区域的属性。例如，目标划分区域中物品1的区域属性为餐厅、客厅，物品2的区域属性为客厅，物品3所属的区域属性为客厅、卧室。
S605、根据目标划分区域中每个物品对应的区域属性，确定目标划分区域中的所有物品的交集区域属性，将交集区域属性确定为目标划分区域的属性。
本发明实施例中，根据目标划分区域中所有物品的交集区域属性，可以得到目标划分区域的交集区域属性集合，交集区域属性集合中的区域属性可以作为目标划分区域的候选属性。例如，目标划分区域中所有物品的交集区域属性为客厅，则将客厅确定为目标划分区域的属性。
可选的，步骤S605确定目标划分区域中的所有物品的交集区域属性之后，将交集区域属性确定为目标划分区域的属性之前，该方法还包括：
判断交集区域属性是否唯一。
若交集区域属性唯一，则执行将交集区域属性确定为目标划分区域的属性 的操作。
若交集区域属性不唯一，则根据预设区域属性优先级获取交集区域属性中优先级最高的目标交集区域属性，将目标交集区域属性确定为目标划分区域的属性。
本发明可选的实施例中，若交集区域属性集合中仅有一个交集区域属性时，可以将该交集区域属性确定为目标划分区域的属性。若交集区域属性集合中有多个交集区域属性时，可以通过预设区域属性优先级确定出优先级最高的目标交集区域属性，将优先级最高的目标交集区域属性确定为目标划分区域的属性。区域属性优先级可以根据不同的目标场景进行不同的设置。可选的，将交集区域属性集合中除目标交集区域属性外的其他交集区域属性作为目标划分区域的候选属性供操作人员选择。
S606、将目标划分区域的属性确定为目标划分区域对应的目标区域子地图的属性。
本发明实施例中，将目标划分区域的属性确定为目标划分区域对应的目标区域子地图的属性，并在目标划分区域的目标区域子地图中显示。
可见，实施如图4所描述的区域的智能划分能够基于目标区域内的物品确定目标划分区域的属性，能够准确的得到区域智能划分结果。
在一个可选的实施例中，若判断出目标划分区域中的所有物品中不包括品类信息为区域限制品类的物品，获取目标划分区域中每个物品对应的区域属性之后，所述方法还包括：
根据目标划分区域中每个物品对应的区域属性得到目标划分区域的区域属性集合。
对于目标划分区域的区域属性集合中的每个区域属性，统计包括该区域属性的物品的个数，得到该区域属性的频率。
获取目标划分区域的区域属性集合中所有区域属性的频率，并按照频率由高至低的顺序对目标划分区域的区域属性集合中所有区域属性进行排序，将排序最靠前的区域属性确定为目标划分区域的属性，进而得到目标划分区域对应的目标区域子地图的属性。
可见，实施该可选实施例，可以通过目标划分区域中多个物品的共有区域属性将确定目标划分区域对应的目标区域子地图的属性，能够准确的得到区域智能划分结果。
实施例三
参阅图6，图6是本发明实施例公开的一种区域的智能划分方法装置的结构示意图。如图6所示，该区域的智能划分装置包括：获取模块601、采集模块602、分析模块603、第一更新模块604、第二更新模块605。其中：
获取模块601，用于获取针对任一目标场景的初始区域地图，目标场景包括多个划分区域，初始区域地图包括每个划分区域对应的初始区域子地图。
本发明实施例中，执行区域智能划分之前，需要获取目标场景的初始区域地图作为基准区域地图进行区域边界的智能划分以及区域分布的智能划分。可 选的，目标场景的初始区域地图可以通过激光雷达扫描目标场景得到，可以基于人工设置的边界标记物得到，也可以通过下载预先存储在数据库中的场景初始地图得到，本发明实施例不做限定。目标场景具体为有区域划分需求的场景，可以是包括多个划分区域的室内场景，也可以是有区域划分需求的其他场景。获取的目标场景包括多个划分区域，相应的，获取的目标场景的初始区域地图中包括每个划分区域对应的初始区域子地图。
采集模块602，用于采集目标划分区域的目标图像，目标划分区域为所有划分区域中的任一区域。
本发明实施例中，采集模块可以集成于清扫设备，该清扫设备包括的采集模块可以是单目摄像装置、多目摄像装置，也可以是包括云台的摄像装置，本发明实施例不做限定。通过采集模块可以采集目标场景的多个划分区域中的任意一个目标划分区域的目标图像，采集到的目标图像可以为目标划分区域的单张图像、目标划分区域不同方向不同角度的多张图像、目标划分区域的全景图像中的一种或多种。
分析模块603，用于分析目标图像，得到目标图像中包含的物品、以及物品的物品信息。
本发明实施例中，清扫设备采集到的目标图像中包括目标划分区域中的物品。目标图像中包括的目标划分区域中的物品可以为一个或多个。通过分析采集到的目标图像可以识别出目标图像中包括的物品，以及物品的物品信息。可选的，物品的物品信息可以包括物品的类型信息以及物品的第一位置信息，还可以包括物品的尺寸信息。物品的类型信息可以是边界类型或非边界类型，示例性的，如分析出目标图像包括门或门框、茶几或餐桌，进一步可以得到门或门框的类型信息为边界类型，茶几或餐桌的类型信息为非边界类型。同时，物品的第一位置信息为物品在目标划分区域对应的初始区域子地图中的位置信息，通过物品的第一位置信息可以在初始区域子地图中定位到该物品。另外，物品的尺寸信息为物品在目标划分区域对应的初始区域子地图中的尺寸信息，通过物品的尺寸信息可以在初始区域子地图中标注出物品所占的空间区域尺寸。
第一更新模块604，用于基于物品的物品信息，更新目标划分区域对应的初始区域子地图，得到目标划分区域对应的目标区域子地图。
本发明实施例中，基于分析得到的物品的物品信息可以对初始区域子地图的区域边界或区域分布进行更新，进而得到智能划分后的目标划分区域对应的目标区域子地图。
第二更新模块605，用于根据目标划分区域对应的目标区域子地图，更新目标场景的初始区域地图。
本发明实施例中，通过物品的物品信息更新目标划分区域对应的初始区域子地图，得到目标划分区域对应的目标区域子地图之后，相应的可以实现目标场景的区域地图的更新。
可见，实施如图6所描述的区域的智能划分能够基于目标区域内的物品实 现区域的智能划分，在避免人工设置标记物的同时提高区域智能划分结果的准确性。
在一个可选的实施例中，请参阅图7，图7是本发明实施例公开的第一更新模块的结构示意图。第一更新模块604包括第一判断子模块6041、分布更新子模块6042、第二判断子模块6043、边界更新子模块6044。
第一判断子模块6041，用于根据物品的类型信息，判断物品是否为边界类型物品。
在该可选的实施例中，根据分析模块603得到的物品的类型信息判断物品是否为边界类型物品。当第一判断子模块6041判断出某一物品是边界类型物品时，则可以根据该边界类型物品修正初始区域子地图边界；当第一判断子模块6041判断出物品是非边界类型物品时，则该物品不能作为修正初始区域子地图边界的物品，而可以作为修正初始区域子地图分布区域的物品。
分布更新子模块6042，用于若第一判断子模块6041判断出物品为非边界类型物品，则基于物品的第一位置信息确定在目标划分区域对应的初始区域子地图中物品所处的分布子区域，并根据分布子区域更新目标划分区域对应的初始区域子地图，其中，物品所处的分布子区域用于表示物品在目标划分区域的分布情况。
在该可选的实施例中，物品是非边界类型物品时，可以根据该物品修正初始区域子地图分布区域。具体的，可以根据物品在目标划分区域对应的初始区域子地图中物品所处的分布子区域，确定物品在目标划分区域的分布情况。
第二判断子模块6043，用于若第一判断子模块6041判断出物品为边界类型物品，则判断物品的第一位置信息是否与初始边界位置信息相匹配，其中，初始边界位置信息为目标划分区域对应的初始区域子地图中与物品对应的初始边界位置信息。
在该可选的实施例中，当第二判断子模块6043判断出物品的第一位置信息与初始边界位置信息匹配时，则目标划分区域对应的初始区域子地图中与物品对应的初始边界位置信息准确，不需要通过物品的位置信息进行修正。当第二判断子模块6043判断出物品的第一位置信息与初始边界位置信息不匹配时，则目标划分区域对应的初始区域子地图中与物品对应的初始边界位置信息不准确，需要通过识别出的物品的物品信息，对初始边界位置信息进行修正。
边界更新子模块6044，用于若第二判断子模块6043判断出物品的第一位置信息与初始边界位置信息不匹配，则基于物品的第一位置信息更新初始边界位置信息，得到目标边界位置信息，并根据目标边界位置信息更新目标划分区域对应的初始区域子地图。
在该可选的实施例中，需要通过识别出的物品的物品信息，对初始边界位置信息进行修正。具体根据物品的第一位置信息修正目标划分区域对应的初始区域子地图中与物品对应的初始边界位置信息，得到修正后的目标边界位置信息，进而根据修正后的目标边界位置信息更新目标划分区域对应的初始区域子地图。
可见，实施该可选的实施例，能够根据物品的类型信息确定初始区域子地图的待更新内容，进一步的可以通过物品的位置信息修正或更新目标划分区域对应的初始区域子地图的边界位置/分布区域，进而得到目标划分区域更准确的区域子地图。
在该可选的实施例中，进一步的，第一更新模块604还包括：确定子模块6045、控制子模块6046、第三判断子模块6047。
确定子模块6045，用于在第二判断子模块6043判断出物品的第一位置信息与初始边界位置信息不匹配之后，边界更新子模块基于物品的第一位置信息更新初始边界位置信息，得到目标边界位置信息，并根据目标边界位置信息更新目标划分区域对应的初始区域子地图之前，基于物品的尺寸信息及物品的第一位置信息，确定在目标划分区域对应的初始区域子地图中物品的物品边缘位置点，其中，物品的物品边缘位置点用于表示物品的边缘位置，物品的物品边缘位置点为多个。
在该可选的实施例中，在目标场景中，需要进一步验证物品类型为边界类型的物品是否被放置于边界位置。物品的物品信息包括物品的尺寸信息，通过物品的尺寸信息及第一位置信息，能够确定出在目标划分区域对应的初始区域子地图中该物品的多个物品边缘位置点。物品的边缘位置点可以为物品的顶点，也可以为物品的边上任意一点，本发明实施例不做限定。
控制子模块6046，用于控制清扫设备在物品对应的待测区域内移动，并在目标划分区域对应的初始区域子地图中记录清扫设备的移动轨迹，其中，物品对应的待测区域为包含物品的所有物品边缘位置点的区域。
在该可选的实施例中，根据物品的所有物品边缘位置点可以确定出包含物品的所有物品边缘位置点的待测区域，该待测区域为初始区域地图中的一虚拟区域，不受到初始区域地图中的初始边界的限制。通过控制清扫设备在该待测区域内移动，可以确定物品是否被放置于区域边界位置。
第三判断子模块6047，用于判断清扫设备的移动轨迹是否为封闭轨迹，以及清扫设备的移动轨迹中是否包含物品的所有物品边缘位置点。
分布更新子模块6042，还用于若第三判断子模块判断出清扫设备的移动轨迹为封闭轨迹，且清扫设备的移动轨迹中包含物品的所有物品边缘位置点，则基于物品的第一位置信息确定在目标划分区域对应的初始区域子地图中物品所处的分布子区域，并根据分布子区域更新目标划分区域对应的初始区域子地图。
边界更新子模块6044，还用于若第三判断子模块判断出清扫设备的移动轨迹不为封闭轨迹，或清扫设备的移动轨迹中不包含物品的所有物品边缘位置点，则基于物品的第一位置信息更新初始边界位置信息，得到目标边界位置信息，并根据目标边界位置信息更新目标划分区域对应的初始区域子地图。
当第三判断子模块6047判断出清扫设备的移动轨迹为封闭轨迹，同时该封闭轨迹内包含物品的所有边缘位置点，即该物品未被放置在边界位置上，则该物品可以作为修正初始区域子地图分布区域的物品，进而通过分布更新子模块6042实现基于物品的第一位置信息确定在目标划分区域对应的初始区域子地图 中物品所处的分布子区域，并根据分布子区域更新目标划分区域对应的初始区域子地图。
当第三判断子模块6047判断出清扫设备的移动轨迹无法形成封闭轨迹，或清扫设备的移动轨迹内没有包含物品的所有边缘位置点，即该物品被放置在边界位置上，则该物品可以作为修正初始区域子地图边界位置的物品，进而通过边界更新子模块6044实现基于物品的第一位置信息更新初始边界位置信息，得到目标边界位置信息，并根据目标边界位置信息更新目标划分区域对应的初始区域子地图。
可见，实施该可选的实施例，能够通过控制清扫设备的移动验证物品类型为边界类型的物品是否被放置于边界位置，再根据验证结果执行相应的更新操作，进而得到更准确的区域划分结果。
在该可选的实施例中，进一步可选的，分布更新子模块6042基于物品的第一位置信息确定在目标划分区域对应的初始区域子地图中物品所处的分布子区域，并根据分布子区域更新目标划分区域对应的初始区域子地图的方式包括：
根据物品的第一位置信息以及物品的尺寸信息，确定在目标划分区域对应的初始区域子地图中物品所处的当前分布子区域。
判断在目标划分区域对应的初始区域子地图中是否存在物品对应的初始分布子区域。
若不存在，则根据物品所处的当前分布子区域更新目标划分区域对应的初始区域子地图。
若存在，则获取物品对应的初始分布子区域，并判断物品对应的初始分布子区域与物品所处的当前分布子区域是否匹配。
若不匹配，则基于物品所处的当前分布子区域更新物品对应的初始分布子区域，得到物品的目标分布子区域，并根据物品的目标分布子区域更新目标划分区域对应的初始区域子地图。
在该可选的实施例中，物品的物品信息包括物品的尺寸信息，通过任一物品的尺寸信息及第一位置信息，能够确定出在目标划分区域对应的初始区域子地图中该物品所处的当前分布子区域。物品所处的当前分布子区域是物品在目标划分区域当前的分布情况。
在目标划分区域对应的初始区域子地图中判断是否存在与该物品对应的初始分布子区域。如果目标划分区域对应的初始区域子地图中不存在与该物品对应的初始分布子区域，则说明该物品为新增物品，可以将该物品的当前分布子区域添加至初始区域子地图中。如果目标划分区域对应的初始区域子地图中存在与该物品对应的初始分布子区域，则说明该物品不为新增物品，可以通过比较初始分布子区域与当前分布子区域的匹配情况判断是否需要更新初始区域子地图。
通过比较初始分布子区域与当前分布子区域的匹配情况判断是否需要更新初始区域子地图。如果初始分布子区域与当前分布子区域相匹配，则该物品在初始区域子地图中所处的子区域不需要修正。如果初始分布子区域与当前分布 子区域不匹配，则该物品在初始区域子地图中所处的子区域有变动，需要通过修正进行更新。
该物品在初始区域子地图中所处的当前分布子区域更新物品对应的初始分布子区域后得到物品的目标分布子区域，可以确定出物品在目标划分区域的分布情况，进而得到更新后的目标划分区域对应的初始区域子地图。
可见，实施该可选的实施例，能够根据物品的所处的当前分布子区域确定确定出物品在目标划分区域的分布情况，进一步的可以通过物品所处的当前分布子区域更新目标划分区域对应的初始区域子地图的分布区域，进而得到目标划分区域更准确的区域子地图。
在又一个可选的实施例中，参阅图8，图8是本发明实施例公开的又一种区域的智能划分装置的结构示意图，该装置还包括：
第一确定模块606，用于基于物品的物品信息，更新目标划分区域对应的初始区域子地图之前，确定目标划分区域对应的初始区域子地图中与物品对应的初始边界位置信息。
其中，第一确定模块606确定目标划分区域对应的初始区域子地图中与物品对应的初始边界位置信息的方式包括：
确定清扫设备采集目标划分区域的目标图像时的采集方向。
获取目标划分区域对应的初始区域子地图中与清扫设备采集目标划分区域的目标图像时的采集方向相匹配的初始边界，将初始边界对应的位置信息确定为目标划分区域对应的初始区域子地图中与物品对应的初始边界位置信息。
在该可选的实施例中，通过清扫设备采集目标划分区域的目标图像时的采集方向，可以确定出目标划分区域对应的初始区域子地图中与物品对应的初始边界位置信息。其中，确定清扫设备采集目标划分区域的目标图像时的采集方向可以通过方向传感器确定，也可以通过图像识别技术确定。
可见，实施该可选的实施例，能够根据物品的所处的当前分布子区域确定确定出物品在目标划分区域的分布情况，进一步的可以通过物品所处的当前分布子区域更新目标划分区域对应的初始区域子地图的分布区域，进而得到目标划分区域更准确的区域子地图。
在又一个可选的实施例中，分析模块603还用于：
分析目标图像，得到目标图像中包含的物品、以及物品的物品信息之后，确定物品的物品识别概率；判断物品的物品识别概率是否大于或等于预设阈值；
第一更新模块604，还用于若分析模块603判断出物品的物品识别概率大于或等于预设阈值，则执行基于所述物品的物品信息，更新目标划分区域对应的初始区域子地图，得到目标划分区域对应的目标区域子地图的操作；
采集模块602，还用于若分析模块603判断出物品的物品识别概率小于预设阈值，则调整清扫设备的拍摄角度，并执行基于清扫设备采集目标划分区域的目标图像，以及分析所述目标图像，得到目标图像中包含的物品、以及物品的物品信息的操作。
在该可选的实施例中，识别出目标图像中包括的物品，以及物品的物品信 息之后，确定目标图像中包括的物品的物品识别概率。物品识别概率用于表示识别结果的准确程度，可选的，物品识别概率可以通过当前物品与标准物品的匹配程度来确定。当物品识别概率超过预设阈值时，表明当前识别结果准确，可以对识别出的物品以及物品信息进行后续操作。当物品识别概率未超过预设阈值时，表明当前识别结果不准确，需要对采集设备的拍摄角度进行调整，重新对目标划分区域进行图像采集，再次识别，进而得到准确的识别结果。
可见，实施该可选的实施例，能够对识别结果的准确性进行判断，并对不准确的识别结果进行适应性的改变，重新识别，防止误判。保证物品识别结果准确的同时，确保后续区域智能划分的准确性。
在另一个可选的实施例中，该装置还包括：第二确定模块607，用于：
获取目标划分区域中的所有物品以及每个物品所属的品类信息。
本发明可选的实施例中，目标划分区域包括一个或多个物品，通过单张、多张或全景目标图像可以获取到目标划分区域中包括的所有物品。并通过图像识别技术识别出目标划分区域中包括的每个物品所属的品类信息。如目标划分区域包括物品1、物品2、物品3，识别出物品1所属的品类信息为餐桌，物品2所属的品类信息为茶几，物品3所属的品类信息为沙发。
判断目标划分区域中的所有物品中是否包括品类信息为区域限制品类的物品，区域限制品类用于表示对应的物品属于某一确定出的区域的概率大于等于预设概率阈值。
本发明可选的实施例中，区域限制品类用于表示对应的物品属于某一确定出的区域的概率大于等于预设概率阈值，预设概率阈值可以根据实际应用场景的需要进行设置。示例性的，若物品4的品类信息为油烟机，油烟机品类信息可以被设置为区域限制品类，即表示物品4属于厨房的概率大于等于预设概率阈值。
若目标划分区域中的所有物品中包括品类信息为区域限制品类的物品，则从目标划分区域中的所有物品中获取目标物品，并将目标物品对应的区域属性确定为目标划分区域的属性，其中，目标物品为品类信息为区域限制品类的物品。
本发明可选的实施例中，当目标划分区域的所有物品中包括品类信息为区域限制品类的物品，则获取品类信息为区域限制品类的物品，例如目标划分区域的所有物品中包括物品4(品类信息为油烟机)，则将物品4作为目标物品，并将物品4对应的区域属性(厨房)确定为目标划分区域的属性(厨房)。
若目标划分区域中的所有物品中不包括品类信息为区域限制品类的物品，则获取目标划分区域中每个物品对应的区域属性。
根据目标划分区域中每个物品对应的区域属性，确定目标划分区域中的所有物品的交集区域属性，将交集区域属性确定为目标划分区域的属性。
本发明可选的实施例中，当目标划分区域的所有物品中不包括品类信息为区域限制品类的物品，则需通过目标划分区域中的多个物品的区域属性确定目标划分区域的属性。例如，目标划分区域中物品1的区域属性为餐厅、客厅， 物品2的区域属性为客厅，物品3所属的区域属性为客厅、卧室。根据目标划分区域中所有物品的交集区域属性，可以得到目标划分区域的交集区域属性集合，交集区域属性集合中的区域属性可以作为目标划分区域的候选属性。例如，目标划分区域中所有物品的交集区域属性为客厅，则将客厅确定为目标划分区域的属性。
将目标划分区域的属性确定为目标划分区域对应的目标区域子地图的属性。
本发明可选的实施例中，将目标划分区域的属性确定为目标划分区域对应的目标区域子地图的属性，并在目标划分区域的目标区域子地图中显示。
可见，实施该可选的实施例，能够基于目标区域内的物品确定目标划分区域的属性，能够准确的得到区域智能划分结果。
本发明可选的实施例中，进一步可选的，第二确定模块607还用于：在确定目标划分区域中的所有物品的交集区域属性之后，将交集区域属性确定为目标划分区域的属性之前：
判断交集区域属性是否唯一。
若交集区域属性唯一，则执行将交集区域属性确定为目标划分区域的属性的操作。
若交集区域属性不唯一，则根据预设区域属性优先级获取交集区域属性中优先级最高的目标交集区域属性，将目标交集区域属性确定为目标划分区域的属性。
本发明可选的实施例中，若交集区域属性集合中仅有一个交集区域属性时，可以将该交集区域属性确定为目标划分区域的属性。若交集区域属性集合中有多个交集区域属性时，可以通过预设区域属性优先级确定出优先级最高的目标交集区域属性，将优先级最高的目标交集区域属性确定为目标划分区域的属性。区域属性优先级可以根据不同的目标场景进行不同的设置。可选的，将交集区域属性集合中除目标交集区域属性外的其他交集区域属性作为目标划分区域的候选属性供操作人员选择。
本发明可选的实施例中，进一步可选的，第二确定模块607还用于：若判断出目标划分区域中的所有物品中不包括品类信息为区域限制品类的物品，获取目标划分区域中每个物品对应的区域属性之后：
根据目标划分区域中每个物品对应的区域属性得到目标划分区域的区域属性集合。
对于目标划分区域的区域属性集合中的每个区域属性，统计包括该区域属性的物品的个数，得到该区域属性的频率。
获取目标划分区域的区域属性集合中所有区域属性的频率，并按照频率由高至低的顺序对目标划分区域的区域属性集合中所有区域属性进行排序，将排序最靠前的区域属性确定为目标划分区域的属性，进而得到目标划分区域对应的目标区域子地图的属性。
可见，实施该可选实施例，可以通过目标划分区域中多个物品的共有区域 属性将确定目标划分区域对应的目标区域子地图的属性，能够准确的得到区域智能划分结果。
实施例四
请参阅图9，图9是本发明实施例公开的又一种区域的智能划分装置的结构示意图。如图9所示，该区域的智能划分装置可以包括：
存储有可执行程序代码的存储器901。
与存储器901耦合的处理器902。
处理器902调用存储器901中存储的可执行程序代码，执行本发明实施例一或本发明实施例二所描述的区域的智能划分方法中的步骤。
实施例五
本发明实施例公开了一种清扫设备，该清扫设备用于执行本发明实施例一或本发明实施例二所描述的区域的智能划分方法。
实施例六
本发明实施例公开了一种计算机可存储介质，该计算机存储介质存储有计算机指令，该计算机指令被调用时，用于执行本发明实施例一或本发明实施例二所描述的区域的智能划分方法中的步骤。
实施例七
本发明实施例公开了一种计算机程序产品，该计算机程序产品包括存储了计算机程序的非瞬时性计算机可读存储介质，且该计算机程序可操作来使计算机执行实施例一或实施例二中所描述的区域的智能划分的方法中的步骤。
以上所描述的装置实施例仅是示意性的，其中所述作为分离部件说明的模块可以是或者也可以不是物理上分开的，作为模块显示的部件可以是或者也可以不是物理模块，即可以位于一个地方，或者也可以分布到多个网络模块上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。本领域普通技术人员在不付出创造性的劳动的情况下，即可以理解并实施。
通过以上的实施例的具体描述，本领域的技术人员可以清楚地了解到各实施方式可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件。基于这样的理解，上述技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品可以存储在计算机可读存储介质中,存储介质包括只读存储器(Read-Only Memory，ROM)、随机存储器(Random Access Memory，RAM)、可编程只读存储器(Programmable Read-only Memory，PROM)、可擦除可编程只读存储器(Erasable Programmable Read Only Memory，EPROM)、一次可编程只读存储器(One-time Programmable Read-Only Memory，OTPROM)、电子抹除式可复写只读存储器(Electrically-Erasable Programmable Read-Only Memory，EEPROM)、只读光盘(Compact Disc Read-Only Memory，CD-ROM)或其他光盘存储器、磁盘存储器、磁带存储器、或者能够用于携带或存储数据的计算机可读的任何其他介质。
最后应说明的是：本发明实施例公开的一种区域的智能划分方法及装置所揭露的仅为本发明较佳实施例而已，仅用于说明本发明的技术方案，而非对其 限制；尽管参照前述实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解；其依然可以对前述各项实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或替换，并不使相应的技术方案的本质脱离本发明各项实施例技术方案的精神和范围。

Claims (10)

  1. 一种区域的智能划分方法，其特征在于，所述方法包括：
    获取针对任一目标场景的初始区域地图，所述目标场景包括多个划分区域，所述初始区域地图包括每个所述划分区域对应的初始区域子地图；
    基于清扫设备采集目标划分区域的目标图像，所述目标划分区域为所有所述划分区域中的任一区域；
    分析所述目标图像，得到所述目标图像中包含的物品、以及所述物品的物品信息；
    基于所述物品的物品信息，更新所述目标划分区域对应的初始区域子地图，得到所述目标划分区域对应的目标区域子地图；
    根据所述目标划分区域对应的目标区域子地图，更新所述目标场景的初始区域地图。
  2. 根据权利要求1所述的区域的智能划分方法，其特征在于，所述物品的物品信息包括所述物品的类型信息以及所述物品的第一位置信息，其中，所述物品的第一位置信息为所述物品在所述目标划分区域对应的初始区域子地图中的位置信息；
    所述基于所述物品的物品信息，更新所述目标划分区域对应的初始区域子地图，包括：
    根据所述物品的类型信息，判断所述物品是否为边界类型物品；
    若判断出所述物品为非边界类型物品，则基于所述物品的第一位置信息确定在所述目标划分区域对应的初始区域子地图中所述物品所处的分布子区域，并根据所述分布子区域更新所述目标划分区域对应的初始区域子地图，其中，所述物品所处的分布子区域用于表示所述物品在所述目标划分区域的分布情况；
    若判断出所述物品为边界类型物品，则判断所述物品的第一位置信息是否与初始边界位置信息相匹配，其中，所述初始边界位置信息为所述目标划分区域对应的初始区域子地图中与所述物品对应的初始边界位置信息；
    若判断出所述物品的第一位置信息与所述初始边界位置信息不匹配，则基于所述物品的第一位置信息更新所述初始边界位置信息，得到目标边界位置信息，并根据所述目标边界位置信息更新所述目标划分区域对应的初始区域子地图。
  3. 根据权利要求2所述的区域的智能划分方法，其特征在于，所述物品的物品信息还包括所述物品的尺寸信息，其中，所述物品的尺寸信息为所述物品在所述目标划分区域对应的初始区域子地图中的尺寸信息；
    在判断出所述物品的第一位置信息与所述初始边界位置信息不匹配之后，所述基于所述物品的第一位置信息更新所述初始边界位置信息，得到目标边界位置信息，并根据所述目标边界位置信息更新所述目标划分区域对应的初始区 域子地图之前，所述方法还包括：
    基于所述物品的尺寸信息及所述物品的第一位置信息，确定在所述目标划分区域对应的初始区域子地图中所述物品的物品边缘位置点，其中，所述物品的物品边缘位置点用于表示所述物品的边缘位置，所述物品的物品边缘位置点为多个；
    控制所述清扫设备在所述物品对应的待测区域内移动，并在所述目标划分区域对应的初始区域子地图中记录所述清扫设备的移动轨迹，其中，所述物品对应的待测区域为包含所述物品的所有物品边缘位置点的区域；
    判断所述清扫设备的移动轨迹是否为封闭轨迹，以及所述清扫设备的移动轨迹中是否包含所述物品的所有物品边缘位置点；
    若所述清扫设备的移动轨迹为封闭轨迹，且所述清扫设备的移动轨迹中包含所述物品的所有物品边缘位置点，则基于所述物品的第一位置信息确定在所述目标划分区域对应的初始区域子地图中所述物品所处的分布子区域，并根据所述分布子区域更新所述目标划分区域对应的初始区域子地图；
    若所述清扫设备的移动轨迹不为封闭轨迹，或所述清扫设备的移动轨迹中不包含所述物品的所有物品边缘位置点，则执行所述基于所述物品的第一位置信息更新所述初始边界位置信息，得到目标边界位置信息，并根据所述目标边界位置信息更新所述目标划分区域对应的初始区域子地图的操作。
  4. 根据权利要求2所述的区域的智能划分方法，其特征在于，所述物品的物品信息还包括所述物品的尺寸信息，其中，所述物品的尺寸信息为所述物品在所述目标划分区域对应的初始区域子地图中的尺寸信息；
    所述基于所述物品的第一位置信息确定在所述目标划分区域对应的初始区域子地图中所述物品所处的分布子区域，并根据所述分布子区域更新所述目标划分区域对应的初始区域子地图，包括：
    根据所述物品的第一位置信息以及所述物品的尺寸信息，确定在所述目标划分区域对应的初始区域子地图中所述物品所处的当前分布子区域；
    判断在所述目标划分区域对应的初始区域子地图中是否存在所述物品对应的初始分布子区域；
    若不存在，则根据所述物品所处的当前分布子区域更新所述目标划分区域对应的初始区域子地图；
    若存在，则获取所述物品对应的初始分布子区域，并判断所述物品对应的初始分布子区域与所述物品所处的当前分布子区域是否匹配；
    若不匹配，则基于所述物品所处的当前分布子区域更新所述物品对应的初始分布子区域，得到所述物品的目标分布子区域，并根据所述物品的目标分布子区域更新所述目标划分区域对应的初始区域子地图。
  5. 根据权利要求1或2所述的区域的智能划分方法，其特征在于，所述基于所述物品的物品信息，更新所述目标划分区域对应的初始区域子地图之前， 所述方法还包括：
    确定所述目标划分区域对应的初始区域子地图中与所述物品对应的初始边界位置信息；
    其中，所述确定所述目标划分区域对应的初始区域子地图中与所述物品对应的初始边界位置信息，包括：
    确定所述清扫设备采集所述目标划分区域的目标图像时的采集方向；
    获取所述目标划分区域对应的初始区域子地图中与所述清扫设备采集所述目标划分区域的目标图像时的采集方向相匹配的初始边界，将所述初始边界对应的位置信息确定为所述目标划分区域对应的初始区域子地图中与所述物品对应的初始边界位置信息。
  6. 根据权利要求1-4任一项所述的区域的智能划分方法，其特征在于，所述方法还包括：
    获取所述目标划分区域中的所有物品以及每个所述物品所属的品类信息；
    判断所述目标划分区域中的所有物品中是否包括品类信息为区域限制品类的物品，所述区域限制品类用于表示对应的物品属于某一确定出的区域的概率大于等于预设概率阈值；
    若所述目标划分区域中的所有物品中包括品类信息为所述区域限制品类的物品，则从所述目标划分区域中的所有物品中获取目标物品，并将所述目标物品对应的区域属性确定为所述目标划分区域的属性，其中，所述目标物品为品类信息为所述区域限制品类的物品；
    若所述目标划分区域中的所有物品中不包括品类信息为所述区域限制品类的物品，则获取所述目标划分区域中每个所述物品对应的区域属性；
    根据所述目标划分区域中每个所述物品对应的区域属性，确定所述目标划分区域中的所有物品的交集区域属性，将所述交集区域属性确定为所述目标划分区域的属性；
    将所述目标划分区域的属性确定为所述目标划分区域对应的目标区域子地图的属性。
  7. 根据权利要求6所述的区域的智能划分方法，其特征在于，所述确定所述目标划分区域中的所有物品的交集区域属性之后，所述将所述交集区域属性确定为所述目标划分区域的属性之前，所述方法还包括：
    判断所述交集区域属性是否唯一；
    若所述交集区域属性唯一，则执行所述将所述交集区域属性确定为所述目标划分区域的属性的操作；
    若所述交集区域属性不唯一，则根据预设区域属性优先级获取所述交集区域属性中优先级最高的目标交集区域属性，将所述目标交集区域属性确定为所述目标划分区域的属性。
  8. 根据权利要求1所述的区域的智能划分方法，其特征在于，所述分析所述目标图像，得到所述目标图像中包含的物品、以及所述物品的物品信息之后，所述基于所述物品的物品信息，更新所述目标划分区域对应的初始区域子地图，得到所述目标划分区域对应的目标区域子地图之前，所述方法还包括：
    确定所述物品的物品识别概率；
    判断所述物品的物品识别概率是否大于或等于预设阈值；
    若所述物品的物品识别概率大于或等于预设阈值，则执行所述基于所述物品的物品信息，更新所述目标划分区域对应的初始区域子地图，得到所述目标划分区域对应的目标区域子地图的操作；
    若所述物品的物品识别概率小于预设阈值，则调整所述清扫设备的拍摄角度，并执行所述基于清扫设备采集目标划分区域的目标图像，以及所述分析所述目标图像，得到所述目标图像中包含的物品、以及所述物品的物品信息的操作。
  9. 一种区域的智能划分装置，其特征在于，所述装置包括：
    获取模块，用于获取针对任一目标场景的初始区域地图，所述目标场景包括多个划分区域，所述初始区域地图包括每个所述划分区域对应的初始区域子地图；
    采集模块，用于采集目标划分区域的目标图像，所述目标划分区域为所有所述划分区域中的任一区域；
    分析模块，用于分析所述目标图像，得到所述目标图像中包含的物品、以及所述物品的物品信息；
    第一更新模块，用于基于所述物品的物品信息，更新所述目标划分区域对应的初始区域子地图，得到所述目标划分区域对应的目标区域子地图；
    第二更新模块，用于根据所述目标划分区域对应的目标区域子地图，更新所述目标场景的初始区域地图。
  10. 一种清扫设备，其特征在于，所述清扫设备用于执行如权利要求1-8任一项所述的区域的智能划分方法。
PCT/CN2022/070839 2021-12-22 2022-01-07 区域的智能划分方法及装置 WO2023115661A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111582090.6A CN116416519A (zh) 2021-12-22 2021-12-22 区域的智能划分方法及装置
CN202111582090.6 2021-12-22

Publications (1)

Publication Number Publication Date
WO2023115661A1 true WO2023115661A1 (zh) 2023-06-29

Family

ID=86901149

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/070839 WO2023115661A1 (zh) 2021-12-22 2022-01-07 区域的智能划分方法及装置

Country Status (2)

Country Link
CN (1) CN116416519A (zh)
WO (1) WO2023115661A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108507578A (zh) * 2018-04-03 2018-09-07 珠海市微半导体有限公司 一种全局边界地图的构建方法及其导航方法
CN111026136A (zh) * 2020-03-11 2020-04-17 广州赛特智能科技有限公司 基于监控设备的港口无人驾驶清扫车智能调度方法及装置
WO2020077850A1 (zh) * 2018-10-18 2020-04-23 深圳乐动机器人有限公司 一种室内区域划分和识别方法、装置及终端设备
CN111973075A (zh) * 2020-08-21 2020-11-24 苏州三六零机器人科技有限公司 基于户型图的地面清扫方法、装置、扫地机和计算机介质
CN112790669A (zh) * 2020-12-30 2021-05-14 北京奇虎科技有限公司 扫地机清扫方法、设备以及存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108507578A (zh) * 2018-04-03 2018-09-07 珠海市微半导体有限公司 一种全局边界地图的构建方法及其导航方法
WO2020077850A1 (zh) * 2018-10-18 2020-04-23 深圳乐动机器人有限公司 一种室内区域划分和识别方法、装置及终端设备
CN111026136A (zh) * 2020-03-11 2020-04-17 广州赛特智能科技有限公司 基于监控设备的港口无人驾驶清扫车智能调度方法及装置
CN111973075A (zh) * 2020-08-21 2020-11-24 苏州三六零机器人科技有限公司 基于户型图的地面清扫方法、装置、扫地机和计算机介质
CN112790669A (zh) * 2020-12-30 2021-05-14 北京奇虎科技有限公司 扫地机清扫方法、设备以及存储介质

Also Published As

Publication number Publication date
CN116416519A (zh) 2023-07-11

Similar Documents

Publication Publication Date Title
CN109903312B (zh) 一种基于视频多目标跟踪的足球球员跑动距离统计方法
CN109840549B (zh) 一种病虫害识别方法和装置
CN110175518A (zh) 摄像装置的摄像头角度调整方法、装置、设备及系统
CN109308460B (zh) 物品检测方法、系统及计算机可读存储介质
CN112926410B (zh) 目标跟踪方法、装置、存储介质及智能视频系统
CN107292223A (zh) 一种实时手势检测的在线验证方法及系统
JP5027741B2 (ja) 画像監視装置
EP3531340A2 (en) Human body tracing method, apparatus and device, and storage medium
CN110796135A (zh) 目标的定位方法及装置、计算机设备、计算机存储介质
CN111950440A (zh) 一种识别与定位门的方法、装置及存储介质
KR101469099B1 (ko) 사람 객체 추적을 통한 자동 카메라 보정 방법
CN114972490B (zh) 一种数据自动标注方法、装置、设备及存储介质
CN114511820A (zh) 货架商品检测方法、装置、计算机设备及存储介质
WO2023115661A1 (zh) 区域的智能划分方法及装置
CN110992500A (zh) 考勤方法、装置及存储介质、服务器
CN109669199B (zh) 一种定位方法、终端和计算机和存储介质
CN111009000A (zh) 昆虫取食行为分析方法、装置和存储介质
CN106557523A (zh) 代表性图像选择方法和设备以及对象图像检索方法和设备
CN109146973A (zh) 机器人场地特征识别和定位方法、装置、设备和存储介质
CN115019216A (zh) 实时地物检测和定位计数方法、系统及计算机
CN114445415A (zh) 可行驶区域的分割方法以及相关装置
CN107403151A (zh) 通过室内天花定位的方法、装置、设备及可读介质
CN113610890A (zh) 一种基于多路摄像头的实时人员轨迹二维平面显示方法
KR20220022961A (ko) 개선된 작물생육 분석시스템
CN112464747A (zh) 基于图像采集设备的身高检测方法及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22909007

Country of ref document: EP

Kind code of ref document: A1