Intelligent identification and cleaning method and system for container and medium
Through the intelligent identification and cleaning method of containers, visual identification and data scanning technology are used to divide and type judgment, and targeted cleaning is carried out in combination with historical data, which solves the problem of failure to thoroughly clean in the existing technology, and achieves a comprehensive and clean container cleaning effect.
Patent Information
- Application Number
- CN202410955645.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-17
- Publication Date
- 2024-11-15
AI Technical Summary
The existing container cleaning technology cannot be carried out targeted cleaning for different areas, resulting in unclear cleaning of some locations, affecting subsequent disinfection.
Through visual recognition and data scanning technology, the container internal image and point cloud data are obtained, area division and type judgment are carried out, and historical data are compared to achieve targeted cleaning.
Ensure a comprehensive and thorough cleaning of the container, improve cleaning results, and ensure that all areas meet high standards of cleanliness.
Smart Images

Figure CN118967577A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of container cleaning, and in particular to a method, system and medium for intelligently identifying and cleaning a container. Background Art
[0002] At present, containers are often cleaned indiscriminately, that is, the same water gun or spray robot is used for uniform cleaning, and the cleaning method and cleaning intensity are the same; some locations in the container that are extremely prone to dirt and grime cannot be cleaned thoroughly, resulting in incomplete cleaning of some locations, which is not conducive to subsequent disinfection. In response to the above problems, effective technical solutions are urgently needed. Summary of the invention
[0003] The purpose of the embodiments of the present application is to provide a method, system and medium for intelligent identification and cleaning of containers, which can collect and analyze data through visual recognition and data scanning to achieve targeted cleaning of different areas, thereby ensuring comprehensive and thorough cleaning of the entire container and making the cleaning cleaner.
[0004] The embodiment of the present application also provides a method for intelligent identification and cleaning of a container, comprising:
[0005] Acquire image data of the space to be cleaned inside the container, and pre-process the image data to generate a color area image;
[0006] Based on the division strategy, the color region image is divided into regions to generate point regions;
[0007] Obtain point cloud data about the space to be cleaned inside the container, analyze the distribution information of the point cloud data, and determine the type of the point area based on the distribution information of the point cloud data to obtain the area type;
[0008] The point area is compared with the historical image data to obtain the verification point area, and the point area type is compared with the historical point cloud data to obtain the verification type information;
[0009] Clean point areas of different area types according to the verification type information.
[0010] Optionally, in the intelligent identification and cleaning method for a container described in an embodiment of the present application, image data about the space to be cleaned inside the container is obtained, and the image data is preprocessed to generate a color area image, specifically including:
[0011] Collect image data from various perspectives inside the container to obtain multi-perspective image data.
[0012] Perform image stitching on image data from different perspectives to generate a large image stitching of the cleaning space inside the container;
[0013] Preprocess the image stitching image, extract image features, and calculate the grayscale value based on the image features;
[0014] Based on the grayscale value, image features of the same or similar colors are fitted into closed color regions to generate several color region images.
[0015] Optionally, in the intelligent identification and cleaning method for containers described in the embodiment of the present application, the color region image is divided into regions based on a division strategy to generate point regions, specifically including:
[0016] Obtain the brightness value of the image in the same color area, set the brightness value interval, and divide the image in the same color area based on the brightness value interval to obtain a number of point areas;
[0017] Get the area of each point region, compare the area with the set area, and get the area difference;
[0018] Determine whether the area difference is greater than or equal to a set area difference threshold;
[0019] If it is greater than, a secondary division rule is generated, and the point area is divided twice based on the secondary division rule;
[0020] If it is less than, several point areas are generated.
[0021] Optionally, in the intelligent identification and cleaning method for a container described in an embodiment of the present application, point cloud data about the space to be cleaned inside the container is obtained, distribution information of the point cloud data is analyzed, and type judgment is performed on the point area based on the distribution information of the point cloud data to obtain the area type, specifically including:
[0022] Obtain point cloud data of the space to be cleaned inside the container, perform feature extraction on the point cloud data, and extract distribution information about the point cloud data;
[0023] Generate visualization data according to the point cloud data and its corresponding distribution information, and obtain thickness information based on the visualization data;
[0024] Determine whether the thickness information is greater than zero.
[0025] If it is greater than zero, the point area is a convex area, the reflectivity of the convexity is obtained, and the type of the convexity is determined based on the reflectivity of the convexity;
[0026] If it is less than zero, the point area is a groove area.
[0027] Optionally, in the intelligent identification and cleaning method for a container described in the embodiment of the present application, judging the type of the protrusion based on the reflectivity or scattering characteristics of the protrusion specifically includes:
[0028] Obtaining the reflectivity of the protrusion, and comparing the reflectivity of the protrusion with a preset standard reflectivity;
[0029] If the reflectivity of the protrusion is less than the preset standard reflectivity, the protrusion is caused by stains, and the point area is marked as a key cleaning area;
[0030] If the reflectivity of the protrusion is greater than or equal to the preset standard reflectivity, the protrusion is a deformed protrusion, and the point area is marked as a normal cleaning area.
[0031] Optionally, in the intelligent identification and cleaning method for containers described in the embodiment of the present application, the point area is compared with the historical image data to obtain the verification point area, and the point area type is compared with the historical point cloud data to obtain the verification type information, which specifically includes:
[0032] Obtain historical image data, extract the area corresponding to the key cleaning area in the historical image data, and compare the area with the key cleaning area to determine whether the difference between the two images is greater than a set value. If so, the key cleaning area is the check point area;
[0033] Obtain historical point cloud data, analyze the area corresponding to the historical point cloud data and the key area, and compare the area with the key cleaning area to determine whether the difference in reflectivity between the two is greater than the reflectivity difference threshold. If so, the area is the point area of the verification type information.
[0034] In a second aspect, an embodiment of the present application provides an intelligent identification and cleaning system for a container, the system comprising: a memory and a processor, the memory comprising a program of an intelligent identification and cleaning method for a container, and the program of the intelligent identification and cleaning method for a container, when executed by the processor, implements the following steps:
[0035] Acquire image data of the space to be cleaned inside the container, and pre-process the image data to generate a color area image;
[0036] Based on the division strategy, the color region image is divided into regions to generate point regions;
[0037] Obtain point cloud data about the space to be cleaned inside the container, analyze the distribution information of the point cloud data, and determine the type of the point area based on the distribution information of the point cloud data to obtain the area type;
[0038] The point area is compared with the historical image data to obtain the verification point area, and the point area type is compared with the historical point cloud data to obtain the verification type information;
[0039] Clean point areas of different area types according to the verification type information.
[0040] Optionally, in the intelligent identification and cleaning system for containers described in the embodiment of the present application, image data about the space to be cleaned inside the container is obtained, and the image data is preprocessed to generate a color area image, specifically including:
[0041] Collect image data from various perspectives inside the container to obtain multi-perspective image data.
[0042] Perform image stitching on image data from different perspectives to generate a large image stitching of the cleaning space inside the container;
[0043] Preprocess the image stitching image, extract image features, and calculate the grayscale value based on the image features;
[0044] Based on the grayscale value, image features of the same or similar colors are fitted into closed color regions to generate several color region images.
[0045] Optionally, in the intelligent identification and cleaning system for containers described in the embodiment of the present application, the color region image is divided into regions based on a division strategy to generate point regions, specifically including:
[0046] Obtain the brightness value of the image in the same color area, set the brightness value interval, and divide the image in the same color area based on the brightness value interval to obtain a number of point areas;
[0047] Get the area of each point region, compare the area with the set area, and get the area difference;
[0048] Determine whether the area difference is greater than or equal to a set area difference threshold;
[0049] If it is greater than, a secondary division rule is generated, and the point area is divided twice based on the secondary division rule;
[0050] If it is less than, several point areas are generated.
[0051] In a third aspect, an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium includes a program for a method for intelligently identifying and cleaning a container. When the program for intelligently identifying and cleaning a container is executed by a processor, the steps of the method for intelligently identifying and cleaning a container as described in any one of the above items are implemented.
[0052] As can be seen from the above, an intelligent identification and cleaning method, system and medium for a container provided in an embodiment of the present application obtain image data about the space to be cleaned inside the container, pre-process the image data to generate a color area image; divide the color area image into regions based on a division strategy to generate point regions; obtain point cloud data about the space to be cleaned inside the container, analyze the distribution information of the point cloud data, and judge the type of the point region based on the distribution information of the point cloud data to obtain the region type; compare the point region with historical image data to obtain a verification point region, and compare the point region type with historical point cloud data to obtain verification type information; clean point regions of different region types according to the verification type information; perform data collection and analysis through visual recognition and data scanning to achieve targeted cleaning of different point regions, thereby ensuring comprehensive and thorough cleaning of the entire container and making the cleaning cleaner. BRIEF DESCRIPTION OF THE DRAWINGS
[0053] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.
[0054] Figure 1 A flow chart of the intelligent identification and cleaning method for containers provided in an embodiment of the present application;
[0055] Figure 2 A flow chart of a method for generating a color region image of a container intelligent identification and cleaning method provided in an embodiment of the present application;
[0056] Figure 3 A flow chart of a point area generation method for the intelligent identification and cleaning method for containers provided in an embodiment of the present application. DETAILED DESCRIPTION
[0057] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.
[0058] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.
[0059] Please refer to Figure 1 , Figure 1 The flowchart of a method for intelligent identification and cleaning of a container in some embodiments of the present application. The method for intelligent identification and cleaning of a container is used in a terminal device, and the method for intelligent identification and cleaning of a container includes the following steps:
[0060] S101, acquiring image data of the space to be cleaned inside the container, and preprocessing the image data to generate a color area image;
[0061] S102, dividing the color region image into regions based on a division strategy to generate point regions;
[0062] S103, obtaining point cloud data about the space to be cleaned inside the container, analyzing distribution information of the point cloud data, and determining the type of the point area based on the distribution information of the point cloud data to obtain the area type;
[0063] S104, comparing the point area with the historical image data to obtain the verification point area, and comparing the point area type with the historical point cloud data to obtain verification type information;
[0064] S105, cleaning point areas of different area types according to the verification type information.
[0065] It should be noted that the data scanning and acquisition unit acquires point cloud data and extracts distribution information, and the visualization unit can facilitate operation and maintenance personnel to view. The data scanning and acquisition unit is composed of several laser radars, which can acquire corresponding point cloud data, extract relevant features of the point cloud data, and extract the distribution information of the point cloud data; the distribution information of the point cloud data is generally the thickness information, reflectivity and scattering characteristics of the point cloud data.
[0066] Please refer to Figure 2 , Figure 2 This is a flow chart of a method for generating a color region image of a container intelligent identification and cleaning method in some embodiments of the present application. According to an embodiment of the present invention, image data of a space to be cleaned inside a container is obtained, and the image data is preprocessed to generate a color region image, specifically including:
[0067] S201, collecting image data from various perspectives inside the container to obtain multi-perspective image data,
[0068] S202, stitching the image data from different viewing angles to generate a large stitched image of the cleaning space inside the container;
[0069] S203, preprocessing the image stitching large image, extracting image features, and calculating grayscale values based on the image features;
[0070] S204, fitting image features of the same or similar colors into closed color regions based on grayscale values to generate a plurality of color region images.
[0071] It should be noted that the visual acquisition unit collects image data from various perspectives in the container, and stitches these image data to generate a large image stitching of the cleaning space inside the container. The visual acquisition unit generally includes a number of movable cameras, and these cameras can be rotated and adjusted in angle. Preferably, these cameras can rotate 360 degrees to better capture comprehensive images of the interior of the container.
[0072] In addition, the image characteristics of the same color are the situation where the grayscale values are the same, and the image characteristics of similar colors are the situation where the grayscale value difference between the two is within a set range. For example, between the grayscale values 200-255, the grayscale value difference setting range is 0 to 20. If the grayscale values of the two are between 200-255 and the difference between the two is between 0 and 20, then they are image characteristics of similar colors.
[0073] Please refer to Figure 3 , Figure 3 This is a flow chart of a point region generation method of a container intelligent identification and cleaning method in some embodiments of the present application. According to an embodiment of the present invention, the color region image is divided into regions based on a division strategy to generate point regions, specifically including:
[0074] S301, obtaining brightness values of images in the same color region, setting brightness value intervals, and dividing the images in the same color region based on the brightness value intervals to obtain a plurality of point regions;
[0075] S302, obtaining the area of each point region, comparing the area with the set area, and obtaining the area difference;
[0076] S303, determining whether the area difference is greater than or equal to a set area difference threshold;
[0077] S304, if it is greater than, generating a secondary division rule, and performing secondary division on the point area based on the secondary division rule;
[0078] S305: If it is less than, generate several point areas.
[0079] According to an embodiment of the present invention, point cloud data about the space to be cleaned inside the container is obtained, distribution information of the point cloud data is analyzed, and type judgment is performed on the point area based on the distribution information of the point cloud data to obtain the area type, which specifically includes:
[0080] Obtain point cloud data of the space to be cleaned inside the container, perform feature extraction on the point cloud data, and extract distribution information about the point cloud data;
[0081] Generate visualization data according to the point cloud data and its corresponding distribution information, and obtain thickness information based on the visualization data;
[0082] Determine whether the thickness information is greater than zero.
[0083] If it is greater than zero, the point area is a convex area, the reflectivity of the convexity is obtained, and the type of the convexity is determined based on the reflectivity of the convexity;
[0084] If it is less than zero, the point area is a groove area.
[0085] According to an embodiment of the present invention, judging the type of a protrusion based on the reflectivity or scattering characteristics of the protrusion specifically includes:
[0086] Obtaining the reflectivity of the protrusion, and comparing the reflectivity of the protrusion with a preset standard reflectivity;
[0087] If the reflectivity of the protrusion is less than the preset standard reflectivity, the protrusion is caused by stains, and the point area is marked as a key cleaning area;
[0088] If the reflectivity of the protrusion is greater than or equal to the preset standard reflectivity, the protrusion is a deformed protrusion, and the point area is marked as a normal cleaning area.
[0089] It should be noted that the image is divided twice according to the division strategy through the division strategy generation unit. Specifically, the image is divided based on the same color area, and the basis of division is the brightness value. By setting the brightness value to several level intervals, several point areas with different brightness values are obtained according to the level intervals to which they belong. The type judgment unit can judge whether the point area is a protrusion caused by stains according to the distribution information of the point cloud data; the point area where the type of protrusion caused by this stain is located is the key cleaning area, which needs to be cleaned subsequently.
[0090] Furthermore, key cleaning point areas and ordinary cleaning point areas are determined. The key cleaning point areas are point areas where there are a large number of stains, for example, the type of protrusions caused by the above-mentioned stains are in the point area; the execution unit can perform targeted cleaning on the point area, that is, adopt different cleaning methods or cleaning intensities for cleaning.
[0091] According to an embodiment of the present invention, the point area is compared with the historical image data to obtain the verification point area, and the point area type is compared with the historical point cloud data to obtain the verification type information, which specifically includes:
[0092] Obtain historical image data, extract the area corresponding to the key cleaning area in the historical image data, and compare the area with the key cleaning area to determine whether the difference between the two images is greater than a set value. If so, the key cleaning area is the check point area;
[0093] Obtain historical point cloud data, analyze the area corresponding to the historical point cloud data and the key area, and compare the area with the key cleaning area to determine whether the difference in reflectivity between the two is greater than the reflectivity difference threshold. If so, the area is the point area of the verification type information.
[0094] It should be noted that the accuracy of the results generated by the area division module is ensured by comparing and judging the image data and the point cloud data through different comparison units. The image difference refers to the degree of difference in color and brightness of the image. Specifically, the above-mentioned difference can be expressed according to the image similarity; the difference in reflectivity is the reflectivity of the former obtained by subtracting the reflectivity of the two.
[0095] According to an embodiment of the present invention, the method further includes: performing targeted cleaning on the point area through an execution unit, that is, adopting different cleaning methods or cleaning strengths to perform cleaning.
[0096] The execution unit is an intelligent cleaning robot, which is a six-axis cleaning robot. It performs cleaning operations under the control of the control unit, which can adjust the cleaning position, cleaning method, and cleaning strength. After obtaining the verification type information, the feedback verification module sends the verification type information to the control unit. The control unit can select the corresponding cleaning position, cleaning method, and cleaning strength according to the key cleaning point area and the ordinary cleaning point area. After the six-axis cleaning robot moves to the corresponding cleaning position, it performs intelligent cleaning according to the cleaning method and cleaning strength of the control unit.
[0097] For example, a certain area is a key cleaning point area, and the control unit formulates the corresponding cleaning position, cleaning method, and cleaning intensity according to the key cleaning point area. The cleaning position is the position where the six-axis cleaning robot moves, and the cleaning method can be direct spray or spiral, and the specific method can be flexibly selected; the cleaning intensity is divided into multiple levels.
[0098] To summarize, the visual recognition module collects image data about the space to be cleaned inside the container, and obtains a color area image based on the image data; at the same time, the data scanning module can collect point cloud data; then the area division module performs area division and type judgment based on the above image data and point cloud data to obtain several preliminary point areas; the feedback verification module can verify and judge the results of the area division module to further ensure the accuracy of the results; finally, the point area is cleaned according to the feedback verification results, and the cleaning method and cleaning intensity of the cleaning execution module are adjusted according to the specific type characteristics of the point area, so as to carry out targeted cleaning to ensure comprehensive cleaning of the entire container.
[0099] In a second aspect, an embodiment of the present application provides an intelligent identification and cleaning system for a container, the system comprising: a memory and a processor, the memory comprising a program for an intelligent identification and cleaning method for a container, and when the program for the intelligent identification and cleaning method for a container is executed by the processor, the following steps are implemented:
[0100] Acquire image data of the space to be cleaned inside the container, and pre-process the image data to generate a color area image;
[0101] Based on the division strategy, the color region image is divided into regions to generate point regions;
[0102] Obtain point cloud data about the space to be cleaned inside the container, analyze the distribution information of the point cloud data, and determine the type of the point area based on the distribution information of the point cloud data to obtain the area type;
[0103] The point area is compared with the historical image data to obtain the verification point area, and the point area type is compared with the historical point cloud data to obtain the verification type information;
[0104] Clean point areas of different area types according to the verification type information.
[0105] It should be noted that the data scanning and acquisition unit acquires point cloud data and extracts distribution information, and the visualization unit can facilitate operation and maintenance personnel to view. The data scanning and acquisition unit is composed of several laser radars, which can acquire corresponding point cloud data, extract relevant features of the point cloud data, and extract the distribution information of the point cloud data; the distribution information of the point cloud data is generally the thickness information, reflectivity and scattering characteristics of the point cloud data.
[0106] According to an embodiment of the present invention, image data of the space to be cleaned inside the container is obtained, and the image data is preprocessed to generate a color area image, specifically including:
[0107] Collect image data from various perspectives inside the container to obtain multi-perspective image data.
[0108] Perform image stitching on image data from different perspectives to generate a large image stitching of the cleaning space inside the container;
[0109] Preprocess the image stitching image, extract image features, and calculate the grayscale value based on the image features;
[0110] Based on the grayscale value, image features of the same or similar colors are fitted into closed color regions to generate several color region images.
[0111] It should be noted that the visual acquisition unit collects image data from various perspectives in the container, and stitches these image data to generate a large image stitching of the cleaning space inside the container. The visual acquisition unit generally includes a number of movable cameras, and these cameras can be rotated and adjusted in angle. Preferably, these cameras can rotate 360 degrees to better capture comprehensive images of the interior of the container.
[0112] In addition, the image characteristics of the same color are the situation where the grayscale values are the same, and the image characteristics of similar colors are the situation where the grayscale value difference between the two is within a set range. For example, between the grayscale values 200-255, the grayscale value difference setting range is 0 to 20. If the grayscale values of the two are between 200-255 and the difference between the two is between 0 and 20, then they are image characteristics of similar colors.
[0113] According to an embodiment of the present invention, the color region image is divided into regions based on a division strategy to generate point regions, specifically including:
[0114] Obtain the brightness value of the image in the same color area, set the brightness value interval, and divide the image in the same color area based on the brightness value interval to obtain a number of point areas;
[0115] Get the area of each point region, compare the area with the set area, and get the area difference;
[0116] Determine whether the area difference is greater than or equal to the set area difference threshold;
[0117] If it is greater than, a secondary division rule is generated, and the point area is divided twice based on the secondary division rule;
[0118] If it is less than, several point areas are generated.
[0119] According to an embodiment of the present invention, point cloud data about the space to be cleaned inside the container is obtained, distribution information of the point cloud data is analyzed, and type judgment is performed on the point area based on the distribution information of the point cloud data to obtain the area type, which specifically includes:
[0120] Obtain point cloud data of the space to be cleaned inside the container, perform feature extraction on the point cloud data, and extract distribution information about the point cloud data;
[0121] Generate visualization data according to the point cloud data and its corresponding distribution information, and obtain thickness information based on the visualization data;
[0122] Determine whether the thickness information is greater than zero,
[0123] If it is greater than zero, the point area is a convex area, the reflectivity of the convexity is obtained, and the type of the convexity is determined based on the reflectivity of the convexity;
[0124] If it is less than zero, the point area is a groove area.
[0125] According to an embodiment of the present invention, judging the type of a protrusion based on the reflectivity or scattering characteristics of the protrusion specifically includes:
[0126] Obtaining the reflectivity of the protrusion, and comparing the reflectivity of the protrusion with a preset standard reflectivity;
[0127] If the reflectivity of the protrusion is less than the preset standard reflectivity, the protrusion is caused by stains, and the point area is marked as a key cleaning area;
[0128] If the reflectivity of the protrusion is greater than or equal to the preset standard reflectivity, the protrusion is a deformed protrusion, and the point area is marked as a normal cleaning area.
[0129] It should be noted that the image is divided twice according to the division strategy through the division strategy generation unit. Specifically, the image is divided based on the same color area, and the basis of division is the brightness value. By setting the brightness value to several level intervals, several point areas with different brightness values are obtained according to the level intervals to which they belong. The type judgment unit can judge whether the point area is a protrusion caused by stains according to the distribution information of the point cloud data; the point area where the type of protrusion caused by this stain is located is the key cleaning area, which needs to be cleaned subsequently.
[0130] Furthermore, key cleaning point areas and ordinary cleaning point areas are determined. The key cleaning point areas are point areas where there are a large number of stains, for example, the type of protrusions caused by the above-mentioned stains are in the point area; the execution unit can perform targeted cleaning on the point area, that is, adopt different cleaning methods or cleaning intensities for cleaning.
[0131] According to an embodiment of the present invention, the point area is compared with the historical image data to obtain the verification point area, and the point area type is compared with the historical point cloud data to obtain the verification type information, which specifically includes:
[0132] Obtain historical image data, extract the area corresponding to the key cleaning area in the historical image data, and compare the area with the key cleaning area to determine whether the difference between the two images is greater than a set value. If so, the key cleaning area is the check point area;
[0133] Obtain historical point cloud data, analyze the area corresponding to the historical point cloud data and the key area, and compare the area with the key cleaning area to determine whether the difference in reflectivity between the two is greater than the reflectivity difference threshold. If so, the area is the point area of the verification type information.
[0134] It should be noted that the accuracy of the results generated by the area division module is ensured by comparing and judging the image data and the point cloud data through different comparison units. The image difference refers to the degree of difference in color and brightness of the image. Specifically, the above-mentioned difference can be expressed according to the image similarity; the difference in reflectivity is the reflectivity of the former obtained by subtracting the reflectivity of the two.
[0135] According to an embodiment of the present invention, the method further includes: performing targeted cleaning on the point area through an execution unit, that is, adopting different cleaning methods or cleaning strengths to perform cleaning.
[0136] The execution unit is an intelligent cleaning robot, which is a six-axis cleaning robot. It performs cleaning operations under the control of the control unit, which can adjust the cleaning position, cleaning method, and cleaning strength. After obtaining the verification type information, the feedback verification module sends the verification type information to the control unit. The control unit can select the corresponding cleaning position, cleaning method, and cleaning strength according to the key cleaning point area and the ordinary cleaning point area. After the six-axis cleaning robot moves to the corresponding cleaning position, it performs intelligent cleaning according to the cleaning method and cleaning strength of the control unit.
[0137] For example, a certain area is a key cleaning point area, and the control unit formulates the corresponding cleaning position, cleaning method, and cleaning intensity according to the key cleaning point area. The cleaning position is the position where the six-axis cleaning robot moves, and the cleaning method can be direct spray or spiral, and the specific method can be flexibly selected; the cleaning intensity is divided into multiple levels.
[0138] To summarize, the visual recognition module collects image data about the space to be cleaned inside the container, and obtains a color area image based on the image data; at the same time, the data scanning module can collect point cloud data; then the area division module performs area division and type judgment based on the above image data and point cloud data to obtain several preliminary point areas; the feedback verification module can verify and judge the results of the area division module to further ensure the accuracy of the results; finally, the point area is cleaned according to the feedback verification results, and the cleaning method and cleaning intensity of the cleaning execution module are adjusted according to the specific type characteristics of the point area, so as to carry out targeted cleaning to ensure comprehensive cleaning of the entire container.
[0139] A third aspect of the present invention provides a computer-readable storage medium, which includes a program for the intelligent identification and cleaning method of containers. When the program for the intelligent identification and cleaning method of containers is executed by a processor, the steps of the intelligent identification and cleaning method of containers as described in any of the above items are implemented.
[0140] The present invention discloses an intelligent identification and cleaning method, system and medium for a container. The method comprises the following steps: obtaining image data of a space to be cleaned inside a container, pre-processing the image data to generate a color region image; dividing the color region image into regions based on a division strategy to generate point regions; obtaining point cloud data of the space to be cleaned inside the container, analyzing distribution information of the point cloud data, and performing type judgment on the point region based on the distribution information of the point cloud data to obtain the region type; comparing the point region with historical image data to obtain a verification point region, and comparing the point region type with historical point cloud data to obtain verification type information; cleaning point regions of different region types according to the verification type information; and performing data collection and analysis through visual recognition and data scanning to achieve targeted cleaning of different point regions, thereby ensuring comprehensive and thorough cleaning of the entire container and making the cleaning cleaner.
[0141] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.
[0142] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.
[0143] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.
[0144] Those skilled in the art can understand that: all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), disks or optical disks, and other media that can store program codes.
[0145] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.
Claims
1. A method for intelligent identification and cleaning of containers, characterized in that: include: Acquire image data of the space to be cleaned inside the container, and pre-process the image data to generate a color area image; Based on the division strategy, the color region image is divided into regions to generate point regions; Obtain point cloud data about the space to be cleaned inside the container, analyze the distribution information of the point cloud data, and determine the type of the point area based on the distribution information of the point cloud data to obtain the area type; The point area is compared with the historical image data to obtain the verification point area, and the point area type is compared with the historical point cloud data to obtain the verification type information; Clean point areas of different area types according to the verification type information.
2. The intelligent identification and cleaning method for containers according to claim 1 is characterized in that: Obtain image data about the space to be cleaned inside the container, and preprocess the image data to generate a color area image, specifically including: Collect image data from various perspectives inside the container to obtain multi-perspective image data. Perform image stitching on image data from different perspectives to generate a large image stitching of the cleaning space inside the container; Preprocess the image stitching image, extract image features, and calculate the grayscale value based on the image features; Based on the grayscale value, image features of the same or similar colors are fitted into closed color regions to generate several color region images.
3. The intelligent identification and cleaning method for containers according to claim 2 is characterized in that: Based on the division strategy, the color region image is divided into regions to generate point regions, including: Obtain the brightness value of the image in the same color area, set the brightness value interval, and divide the image in the same color area based on the brightness value interval to obtain a number of point areas; Get the area of each point region, compare the area with the set area, and get the area difference; Determine whether the area difference is greater than or equal to a set area difference threshold; If it is greater than, a secondary division rule is generated, and the point area is divided twice based on the secondary division rule; If it is less than, several point areas are generated.
4. The intelligent identification and cleaning method for containers according to claim 3 is characterized in that: Obtain point cloud data about the space to be cleaned inside the container, analyze the distribution information of the point cloud data, and determine the type of the point area based on the distribution information of the point cloud data to obtain the area type, including: Obtain point cloud data of the space to be cleaned inside the container, perform feature extraction on the point cloud data, and extract distribution information about the point cloud data; Generate visualization data according to the point cloud data and its corresponding distribution information, and obtain thickness information based on the visualization data; Determine whether the thickness information is greater than zero, If it is greater than zero, the point area is a convex area, the reflectivity of the convexity is obtained, and the type of the convexity is determined based on the reflectivity of the convexity; If it is less than zero, the point area is a groove area.
5. The intelligent identification and cleaning method for containers according to claim 4 is characterized in that: The type of protrusion is determined based on the reflectivity or scattering characteristics of the protrusion, including: Obtaining the reflectivity of the protrusion, and comparing the reflectivity of the protrusion with a preset standard reflectivity; If the reflectivity of the protrusion is less than the preset standard reflectivity, the protrusion is caused by stains, and the point area is marked as a key cleaning area; If the reflectivity of the protrusion is greater than or equal to the preset standard reflectivity, the protrusion is a deformed protrusion, and the point area is marked as a normal cleaning area.
6. The intelligent identification and cleaning method for containers according to claim 5, characterized in that: By comparing the point area with historical image data, the verification point area is obtained. By comparing the point area type with historical point cloud data, the verification type information is obtained, including: Obtain historical image data, extract the area corresponding to the key cleaning area in the historical image data, and compare the area with the key cleaning area to determine whether the difference between the two images is greater than a set value. If so, the key cleaning area is the check point area; Obtain historical point cloud data, analyze the area corresponding to the historical point cloud data and the key area, and compare the area with the key cleaning area to determine whether the difference in reflectivity between the two is greater than the reflectivity difference threshold. If so, the area is the point area of the verification type information.
7. An intelligent container identification and cleaning system, characterized in that: The system includes: a memory and a processor, wherein the memory includes a program of a method for intelligently identifying and cleaning a container, and when the program of the method for intelligently identifying and cleaning a container is executed by the processor, the following steps are implemented: Acquire image data of the space to be cleaned inside the container, and pre-process the image data to generate a color area image; Based on the division strategy, the color region image is divided into regions to generate point regions; Obtain point cloud data about the space to be cleaned inside the container, analyze the distribution information of the point cloud data, and determine the type of the point area based on the distribution information of the point cloud data to obtain the area type; The point area is compared with the historical image data to obtain the verification point area, and the point area type is compared with the historical point cloud data to obtain the verification type information; Clean point areas of different area types according to the verification type information.
8. The intelligent identification and cleaning system for containers according to claim 7, characterized in that: Obtain image data about the space to be cleaned inside the container, and preprocess the image data to generate a color area image, specifically including: Collect image data from various perspectives inside the container to obtain multi-perspective image data. Perform image stitching on image data from different perspectives to generate a large image stitching of the cleaning space inside the container; Preprocess the image stitching image, extract image features, and calculate the grayscale value based on the image features; Based on the grayscale value, image features of the same or similar colors are fitted into closed color regions to generate several color region images.
9. The intelligent identification and cleaning system for containers according to claim 8, characterized in that: Based on the division strategy, the color region image is divided into regions to generate point regions, including: Obtain the brightness value of the image in the same color area, set the brightness value interval, and divide the image in the same color area based on the brightness value interval to obtain a number of point areas; Get the area of each point region, compare the area with the set area, and get the area difference; Determine whether the area difference is greater than or equal to a set area difference threshold; If it is greater than, a secondary division rule is generated, and the point area is divided twice based on the secondary division rule; If it is less than, several point areas are generated.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a container intelligent identification and cleaning method program, and when the container intelligent identification and cleaning method program is executed by a processor, the steps of the container intelligent identification and cleaning method according to any one of claims 1 to 6 are implemented.
Citation Information
Cited By
AI visual system for shaft part stain identification and cleaning equipment
CN120088209A
