CN116777903B - Box door detection method and system - Google Patents
Box door detection method and system Download PDFInfo
- Publication number
- CN116777903B CN116777903B CN202311016523.0A CN202311016523A CN116777903B CN 116777903 B CN116777903 B CN 116777903B CN 202311016523 A CN202311016523 A CN 202311016523A CN 116777903 B CN116777903 B CN 116777903B
- Authority
- CN
- China
- Prior art keywords
- corner
- point cloud
- cloud data
- box
- door
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims description 11
- 238000000034 method Methods 0.000 claims abstract description 38
- 230000004044 response Effects 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 9
- 238000005516 engineering process Methods 0.000 abstract description 2
- 230000008569 process Effects 0.000 description 14
- 239000011159 matrix material Substances 0.000 description 9
- 238000012986 modification Methods 0.000 description 8
- 230000004048 modification Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 4
- 230000006872 improvement Effects 0.000 description 3
- 239000000463 material Substances 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000000712 assembly Effects 0.000 description 1
- 238000000429 assembly Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000003607 modifier Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/64—Three-dimensional objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20164—Salient point detection; Corner detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30232—Surveillance
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The present disclosure relates to the field of information technologies, and in particular, to a method and a system for detecting a door of a vehicle, where the method includes: acquiring side face point cloud data corresponding to the side face of the box body; extracting corner features based on the side point cloud data; based on the corner characteristics, judging whether the side surface is the door side of the box body.
Description
Technical Field
The present disclosure relates to the field of information technologies, and in particular, to a method and a system for detecting a door of a vehicle.
Background
In the port automatic operation process, the automatic driving vehicle box unloading operation needs to be accurately stopped below the gantry crane lifting appliance, so that the end face of the container is aligned with the lifting appliance, and the lifting appliance is convenient to grasp. However, the containers are divided into door sides and non-doors, which are slightly different in construction and relative positions of alignment with the spreader.
Accordingly, it is desirable to provide a door detection method that can accurately identify and determine the door side and non-door of a container to facilitate automated operations.
Disclosure of Invention
One of the embodiments of the present disclosure provides a method for detecting a door, including: acquiring side face point cloud data corresponding to the side face of the box body; extracting corner features based on the side point cloud data; based on the corner characteristics, judging whether the side surface is the door side of the box body.
One of the embodiments of the present disclosure provides a door check system, including: the point cloud acquisition module is used for acquiring side point cloud data corresponding to the side of the box body; the corner extraction module is used for extracting corner features based on the side face point cloud data; and the box door judging module is used for judging whether the side surface is the box door side of the box body based on the corner characteristics.
One of the embodiments of the present disclosure provides a door check device, including a processor for executing the door check method described above.
One of the embodiments of the present specification provides a computer-readable storage medium storing computer instructions that when read by a computer in the storage medium, the computer performs the above-described door detection method.
Drawings
The present specification will be further elucidated by way of example embodiments, which will be described in detail by means of the accompanying drawings. The embodiments are not limiting, in which like numerals represent like structures, wherein:
FIG. 1 is an exemplary flow chart of a method of detecting a door according to some embodiments of the present disclosure;
FIG. 2 is a schematic illustration of side point cloud data of a chest door side shown in accordance with some embodiments of the present description;
FIG. 3 is a schematic illustration of non-bin-door side point cloud data shown according to some embodiments of the present description;
FIG. 4 is a schematic illustration of corner features of a bin gate side shown according to some embodiments of the present description;
FIG. 5 is a schematic illustration of corner features of a non-door side shown in accordance with some embodiments of the present disclosure;
FIG. 6 is a statistical plot of corner features of the box gate side in a plurality of data frames according to some embodiments of the present description;
FIG. 7 is a statistical plot of corner features of non-bin sides in a plurality of data frames according to some embodiments of the present description;
FIG. 8 is an exemplary flow chart for determining whether a side is a door side of a box, according to some embodiments of the present disclosure;
FIG. 9 is an exemplary block diagram of a door check system according to some embodiments of the present disclosure.
Detailed Description
In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the drawings that are required to be used in the description of the embodiments will be briefly described below. It is apparent that the drawings in the following description are only some examples or embodiments of the present specification, and it is possible for those of ordinary skill in the art to apply the present specification to other similar situations according to the drawings without inventive effort. Unless otherwise apparent from the context of the language or otherwise specified, like reference numerals in the figures refer to like structures or operations.
It will be appreciated that "system," "apparatus," "unit" and/or "module" as used herein is one method for distinguishing between different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.
As used in this specification and the claims, the terms "a," "an," "the," and/or "the" are not specific to a singular, but may include a plurality, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.
A flowchart is used in this specification to describe the operations performed by the system according to embodiments of the present specification. It should be appreciated that the preceding or following operations are not necessarily performed in order precisely. Rather, the steps may be processed in reverse order or simultaneously. Also, other operations may be added to or removed from these processes.
In some embodiments, in order to identify the door of the container, the image information may be acquired by providing a special camera, and identifying the image information by means of deep learning to determine whether it is the door side. But this approach requires additional sensors and a special camera and running a deep learning model also places demands on the device computational power.
In view of this, some embodiments of the present disclosure provide a door detection method that determines whether a side is a door side by acquiring point cloud data of the side.
Fig. 1 is an exemplary flow chart of a method of detecting a door according to some embodiments of the present disclosure. As shown in fig. 1, the process 100 includes the following steps. In some embodiments, the process 100 may be performed by a processor or a cloud server, etc., and for ease of description, the process will be described below with respect to a processor.
Step 110, acquiring side face point cloud data corresponding to the side face of the box body. In some embodiments, step 110 may be performed by the point cloud acquisition module 910.
The point cloud data refers to a point data set of the product appearance surface obtained through an acquisition device (such as a radar, a scanner and the like), and the point data can comprise information such as the number, the distribution, the density and the like of points. In some embodiments, the point cloud data may also include line segment data, or the like.
Since the container has a rectangular parallelepiped outer contour (6 faces total), the areas of the container door side and the non-door side opposite the door side are obviously identical (or similar), while the areas of the door side and the non-door side are far different from the areas of the remaining 4 sides of the container except the door side and the non-door side (e.g., the areas of the door side and the non-door side are smaller than the areas of the remaining 4 sides). Thus, in some embodiments, the area data of multiple sides of the container may be acquired by the acquisition device, and the data of the sides of the corresponding container may be quickly removed (e.g., removing the sides with larger areas) by, for example, comparing the areas of the sides.
In some embodiments, the preset range may also be determined based on the areas of the door sides and the non-door sides of the multiple types of boxes, and the sides with the acquired side areas outside the preset range may be eliminated, that is, the acquired point cloud data only includes the door sides or the non-door sides of the container, so that in the following, the sides of the boxes default to represent the door sides or the non-door sides of the opposite sides of the container.
In some embodiments, the processor may collect an observation map of the container body via the collection device and obtain side point cloud data based on the observation map. Taking radar as an example, the side point cloud data is an observation diagram of a certain side of the box body acquired by the radar, and the corresponding side point cloud data is obtained based on the observation diagram.
Referring to fig. 2 and 3, there are shown side point cloud data of the door side (fig. 2) and the non-door side (fig. 3) of a container body acquired by a radar, respectively. In some embodiments, the radar may be installed on a vehicle (such as an IMV) in a port, or may be a radar originally provided on the vehicle, such as the RS-Bpearl blind-complement laser radar on the aforementioned IMV. It should be noted that the foregoing vehicle and radar model are only examples, and in some embodiments, the collecting device may be a laser radar or a camera, and the collecting device may be mounted on other movable devices, such as an unmanned plane, a robot, and the like.
As can be seen from fig. 2 and 3, the two sub-graphs have a high similarity (similar line distribution) from the two figures only, and there is still a certain difficulty in distinguishing the door side and the non-door side of the box, and therefore, further processing is required.
And 120, extracting corner features based on the side point cloud data. In some embodiments, step 120 may be performed by corner extraction module 920.
The corner points represent extreme points of a type in the point cloud data, and the corner points may include one or more of points in the point cloud data where a gradient value and/or a change rate of a gradient direction is high, points where two or more line segments intersect, end points of the line segments, points corresponding to local maxima of the first derivative, and points in the point cloud data where edge changes are discontinuous. The corner features represent information of the corners in the side point cloud data, for example, the corner features may include the number of corners, the positions of the corners, the distribution of the corners, and the like.
In some embodiments, the processor may extract corner features based on the side point cloud data through a specific algorithm (e.g., a corner detection algorithm) or a machine learning model. For example, taking fig. 2 and fig. 3 as examples, the corner features extracted by the processor based on the side point cloud data (e.g. by using a corner detection algorithm) may be shown in fig. 4 and fig. 5, respectively, where the corner features may be the number of all corners in the side point cloud data.
As can be seen from fig. 4 and fig. 5, in the corner features of the door side and the non-door side, the number of corners near the middle of the image is different, so that the processor can extract the number of corners of the target area in the side point cloud data, that is, the corner features include the number of corners of the target area in the side point cloud data.
In some embodiments, the target area may be an area corresponding to the entire side point cloud data, or may be a partial area, for example, an area with a larger difference between the number of corner points on the side of the selected door and the number of corner points on the side of the non-door. The reason for the large difference in the number of corner points includes structural differences between the door side and the non-door side of the cabinet, for example, the door side is provided with a plurality of columns on its outer surface as compared with the non-door side, and the cabinet shown in fig. 4 and 5 is taken as an example, and the target area may be an area of the door side of the cabinet containing the aforementioned plurality of columns, such as an area corresponding to a position of 6m wide and 1.5m high in the side, or a smaller area.
In some embodiments, since the radar may continuously acquire the side point cloud data during operation, if the radar acquires data including multiple sets of side point cloud data, the number of corner points of each set of side point cloud data in the multiple sets of side point cloud data may be determined, an average value of the number of corner points of each set of side point cloud data may be calculated, and further judgment (whether the judgment described later satisfies the first preset condition or the second preset condition) may be performed by using the average value.
In some embodiments, the processor may extract the number of corner points of the target area in the side point cloud data based on a corner point extraction algorithm (such as Harris extraction algorithm or Susan extraction algorithm) or the like.
Fig. 6 and fig. 7 are respectively corresponding to statistical graphs of corner features of a box door side and a non-box door side in a plurality of data frames acquired by a radar, wherein an abscissa in the graph is a data frame number, and an ordinate in the graph is a corner number corresponding to the data frame number.
It will be appreciated that the side point cloud data in fig. 4 and 5 may be side point cloud data corresponding to a certain data frame in fig. 6 and 7. As can be seen from the figure, the number of the corner points of the target area on the door side is basically in the range of 30 to 40, and the number of the corner points of the target area on the non-door side is basically in the range of 5 to 12. At this time, compared with the case door side and the non-case door side of the case body which are simply distinguished by fig. 2 and 3, the judging process by using the number of the corner points is more visual and accurate.
As can be seen from fig. 3 and 4, there is also a difference in the locations of the corner points in the side point cloud data of the bin gate side and the non-bin gate side, and thus, in some embodiments, the corner features may include a corner distribution of the corner points in the side point cloud data.
The corner distribution may represent the position and quantity information of the corner in the side point cloud data, and the corner distribution may be extracted based on the side point cloud data based on a specific algorithm (such as a feature extraction algorithm). In some embodiments, the corner distribution may be in the form of a matrix, for example, the corner distribution may be a 3×3 matrix, which may then divide the side point cloud data into 9 regions, the number of corners in each region corresponding to the values of the elements in the matrix. For the processing of the diagonal distribution, see later on the relevant description of fig. 7.
And 130, judging whether the side surface is the door side of the box body based on the corner characteristics. In some embodiments, step 130 may be performed by the door determination module 930.
In some embodiments, the processor may determine whether one of the two different sides is a door side of the box by comparing the number of corner points based on obtaining corner features of the two different sides (one of the two different sides is a door side of the box). For example, the processor may determine corner features for two different sides, and determine that the more numerous sides of the corner correspond to the sides of the bin door.
In some embodiments, the manner of determining whether the side is the door side of the cabinet may include: responding to the quantity of the corner points meeting a first preset condition, and determining that the side surface is the door side of the box body; and determining that the side surface is the non-box door side of the box body in response to the fact that the number of the corner points does not meet the first preset condition.
In some embodiments, the first preset condition may be whether the value meets a preset threshold. The preset threshold value can be determined based on the number of corner points in the historically acquired box door side point cloud data of the box body. Taking the case corresponding to fig. 6 and 7 as an example, the number of the corner points of the case door side is basically in the range of 30-40, and the number of the corner points of the non-case door side is basically in the range of 5-12, so that the preset threshold value can be set to be 20, and when the number of the corner points extracted from the obtained side point cloud data is greater than 20, the side is the case door side of the case; otherwise, the box body is at the non-box door side.
It should be noted that, for the case, the preset threshold may also be other values, such as 15, 22, 25, etc., as determined by the average value of the numbers of the plurality of groups of corner points in the foregoing description; for other containers, the preset threshold may be set according to practical situations, which is not limited in this specification.
By setting the first preset condition, it is possible to determine whether the side surface is the door side of the case by simple comparison.
Analysis and comparison show that a relatively fixed proportional relationship exists between the numbers of corner points corresponding to the box door side and the non-box door side, and in some embodiments, the processor can determine that the side is the non-box door side of the box body according to the proportional relationship.
In some embodiments, the processor may obtain corner features for two different sides, wherein the two different sides include a first side of corresponding first side point cloud data and a second side of corresponding second side point cloud data. The mode of judging whether the side surface is the door side of the case may include: determining the first side as the box door side in response to the difference characteristic of the number of corner points of the first side point cloud data and the number of corner points of the second side point cloud data meeting a second preset condition;
and determining the second side as the box door side in response to the difference characteristic of the number of the corner points of the second side point cloud data and the number of the corner points of the first side point cloud data meeting a second preset condition.
The difference feature may be a numerical ratio of the number of corner points, and the second preset condition may be that the numerical ratio satisfies a preset ratio condition. For example, a numerical ratio of the number of corner points of the first side and the number of corner points of the second side may be calculated and compared with a preset ratio to determine whether the second preset condition is satisfied.
Taking fig. 4 and fig. 5 as examples, assume that fig. 4 is a first side, and the number of corner points in the target area is 35; fig. 5 shows a second side, in which the number of corner points in the target area is 8, and since the difference feature is a numerical ratio of the number of corner points, the numerical ratio of the number of corner points of the first side point cloud data to the number of corner points of the second side point cloud data is 4.375; accordingly, the difference characteristic of the number of corner points of the second side point cloud data and the number of corner points of the first side point cloud data is 0.23. If the second preset condition is set to be that the ratio of the numerical ratio is not less than 2.5, it can be seen that the difference feature of the number of corner points of the first side point cloud data and the number of corner points of the second side point cloud data satisfies the second preset condition, and therefore, the processor can determine that the first side (i.e., the side shown in fig. 4) is the door side.
The box door side is determined through the difference characteristics of two different sides, so that the result accuracy is higher.
According to the door detection method provided by the process 100, the point cloud data of the side face of the container body can be conveniently acquired through the radar, whether the side face is the door side or not is judged according to the point cloud data, and the judgment result is high in accuracy and low in hardware requirement.
Fig. 8 is an exemplary flow chart for determining whether a side is a door side of a cabinet, according to some embodiments of the present disclosure. As before, in some embodiments, the corner feature may include a corner distribution of corners in the side point cloud data, and fig. 8 illustrates a process 800 of determining whether a side is a door side of a box based on the corner distribution, which specifically includes the following steps. In some embodiments, the process 800 may be performed by the door determination module 930.
And step 810, searching in a corner database based on corner distribution of the side point cloud data, and determining whether reference data of the same type of box exists in the corner database.
The reference data is the preset reference corner distribution corresponding to each type of box in the corner database. The reference data is used for searching to match angular point distribution in the side point cloud data with angular point distribution information of the corresponding box body. The reference data includes a distribution of reference corners on the chest door side and a distribution of reference corners on the non-chest door side.
In some embodiments, the reference corner may be a representative or easily identifiable portion of all the corner points extracted from the side point cloud data. For example, as described above, the corner features in the target area in the side point cloud data are more easily identified due to the difference between the door side and the non-door side structures, and thus, in some embodiments, the door determination module 930 may use the corner points in the target area in the side point cloud data as reference corner points. The reference corner distribution may comprise position information of the reference corners and the manner of determining the reference corner distribution may be referred to as the manner of determining the corner distribution in the foregoing.
The corner databases may be pre-constructed, including databases of corner distribution data for different types of containers. In some embodiments, the corner database may store data for a plurality of containers in the form of [ container model (number) ] [ corresponding to the door side reference corner distribution data, corresponding to the door side reference corner distribution data ].
The box door judging module 930 may quickly determine whether the reference data of the same type of box exists in the corner database by searching based on the corner distribution of the current box side point cloud data acquired by the radar. For example, the box door judging module 930 may search in a corner database based on the corner distribution in the current box side point cloud data acquired by the radar, and when a preset condition is satisfied between the corner distribution in the current side point cloud data and each reference corner distribution in the corner database, it may determine that reference data of the same type box exists in the corner database, and may determine a container type (e.g. 1 CC) of the current box based on the container type corresponding to the search result.
In some embodiments, the manner of searching in the corner database includes, but is not limited to, one or more of inverted index, simHash algorithm, similarity algorithm, and the like, so as to determine whether the same type of box reference data exists in the corner database. Different preset conditions can be determined according to different searching modes, for example, when searching is performed by using a similarity algorithm, a similarity threshold can be set to determine a result meeting the requirement (such as being greater than the similarity threshold), and the method is not limited herein.
Step 820, in response to the existence of the reference data of the same type of box body in the corner database, comparing the corner distribution with the box door side reference corner distribution and the non-box door side reference corner distribution, and judging whether the side corresponding to the side point cloud data is the box door side.
In some embodiments, if the corner database includes reference data of the same type of box, the box door judging module 930 may compare the corner distribution of the current box side point cloud data obtained by the radar with the corresponding box door side reference corner distribution data and/or the corresponding non-box door side reference corner distribution data of the type of box in the corner database, and if the corner distribution of the current box side point cloud data and the corresponding box door side reference corner distribution data meet preset conditions, it may determine that the side corresponding to the side point cloud data is the box door side, and otherwise, is the non-box door side. For example, as before, the corner distribution and the reference corner distribution may be matrices, and thus, in some embodiments, the comparison method may include calculating a matrix distance or a matrix similarity between two matrices, and the preset condition may be that the matrix distance obtained by the comparison is smaller than a preset value or the matrix similarity is larger than a preset value, and so on.
By comparing and searching the corner features with the data in the corner database, the side point cloud data on any side of the box body can be obtained for judgment, and the method is better in adaptability and high in accuracy to areas with various container types.
In some embodiments, the box door judging module 930 may update the reference data in the corner database based on the obtained corner distribution, and the process 800 may further include:
and 830, averaging the angular point distribution of the box door side or non-box door side of the box body and the angular point distribution of the corresponding box door side or non-box door side searched in the angular point database to update the angular point distribution of the box door side or non-box door side of the box body with the same type existing in the angular point database.
For example, in response to the existence of the reference data of the same type of box in the corner database and the fact that the side corresponding to the side point cloud data is the box door side, the corner distribution of the current point cloud data and the box door side reference corner distribution of the type of box in the corner database may be averaged, for example, if the corner distribution is an 8×8 matrix, the two matrices may be added in pairs and each element divided by 2 to obtain a new corner distribution matrix, and the new box door side reference corner distribution of the type of box is updated to the corner database.
The reference corner in the corner database is updated iteratively, so that the reference data in the corner database is more accurate along with the increase of the service time.
In step 840, in response to the corner database not having the reference data of the boxes with different models, second side point cloud data and corner features thereof corresponding to the second side of the box are obtained.
In some embodiments, if the reference data of the same type of box does not exist in the corner database, the second side point cloud data of the other side (i.e., the second side) of the side corresponding to the current side point cloud data needs to be acquired for comparison.
Step 850, determining the box door side and the non-box door side of the box body based on the corner feature of the side point cloud data and the corner feature of the second side point cloud data.
After the corner features of the side point cloud data of the two sides of the box are obtained in step 850, the method of determining the box door side and the non-box door side of the box based on the corner features may adopt the scheme described in step 130, and specifically, refer to fig. 1 and the related description of step 130, which are not repeated herein.
Step 860, the model of the box is obtained, and at least the model of the box and the corner distribution of the box door side and/or the corner distribution of the non-box door side are stored in the corner database.
In some embodiments, the door determination module 930 may obtain the model of the case by requesting the user to fill in or scan other locations (e.g., nameplates) of the case, etc. After determining the model of the box, the box door judging module 930 may determine the box door side reference corner distribution data and the box door side reference corner distribution data based on the corner feature of the side point cloud data and the corner feature of the second side point cloud data obtained in the foregoing steps and the judging result (i.e., the side point cloud data corresponds to the box door side or is not the box door side), and store the box door side reference corner distribution data in the corner database in the form of [ container model (number) ] - [ corresponding to the box door side reference corner distribution data ] and corresponding to the box door side reference corner distribution data ] illustrated in step 810.
Through the steps, the non-input model can be expanded in the corner database, the corner database can be gradually expanded along with the use under the condition that the detection of the box door is not affected, the adaptability of the method to containers of different models is improved, and meanwhile, the accuracy of box door matching can be improved.
It should be noted that the above description of the process 100 and the process 800 is merely for illustration and description, and is not intended to limit the application scope of the present disclosure. Various modifications and changes to the flow may be made by those skilled in the art under the guidance of this specification. However, such modifications and variations are still within the scope of the present description.
FIG. 9 is an exemplary block diagram of a door check system according to some embodiments of the present disclosure.
In some embodiments, the door detection system 900 may include a point cloud acquisition module 910, a corner extraction module 920, and a door determination module 930.
The point cloud acquisition module 910 is configured to acquire side point cloud data corresponding to a side of the box.
For further description of the side point cloud data, reference may be made to step 110, which is not repeated here.
The corner extraction module 920 is configured to extract corner features based on the side point cloud data.
For further description of corner features, reference may be made to step 120, which is not repeated here.
The box door judging module 930 is configured to judge whether the side surface is a box door side of the box body based on the corner feature.
For more description of the determination process, reference may be made to step 130 and related content of flow 800, which are not described herein.
It should be appreciated that the door check system 900 shown in fig. 9 and its modules can be implemented in a variety of ways. It should be noted that the above description of the door check system 900 and its modules is for convenience only, and is not intended to limit the present disclosure to the exemplary embodiments. It will be appreciated by those skilled in the art that, given the principles of the system, various modules may be combined arbitrarily or a subsystem may be constructed in connection with other modules without departing from such principles. In some embodiments, the point cloud acquiring module 910, the corner extracting module 920, and the box door judging module 930 disclosed in fig. 9 may be different modules in one system, or may be one module to implement the functions of two or more modules. For example, each module may share one memory module, or each module may have a respective memory module. Such variations are within the scope of the present description.
While the basic concepts have been described above, it will be apparent to those skilled in the art that the foregoing detailed disclosure is by way of example only and is not intended to be limiting. Although not explicitly described herein, various modifications, improvements, and adaptations to the present disclosure may occur to one skilled in the art. Such modifications, improvements, and modifications are intended to be suggested within this specification, and therefore, such modifications, improvements, and modifications are intended to be included within the spirit and scope of the exemplary embodiments of the present invention.
Meanwhile, the specification uses specific words to describe the embodiments of the specification. Reference to "one embodiment," "an embodiment," and/or "some embodiments" means that a particular feature, structure, or characteristic is associated with at least one embodiment of the present description. Thus, it should be emphasized and should be appreciated that two or more references to "an embodiment" or "one embodiment" or "an alternative embodiment" in various positions in this specification are not necessarily referring to the same embodiment. Furthermore, certain features, structures, or characteristics of one or more embodiments of the present description may be combined as suitable.
Furthermore, the order in which the specification processes elements and sequences, the use of numerical letters, or other designations are used is not intended to limit the order in which the specification flows and methods are performed unless explicitly recited in the claims. While certain presently useful inventive embodiments have been discussed in the foregoing disclosure, by way of various examples, it is to be understood that such details are merely illustrative and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements included within the spirit and scope of the embodiments of the present disclosure. For example, while the system components described above may be implemented by hardware devices, they may also be implemented solely by software solutions, such as installing the described system on an existing server or mobile device.
Likewise, it should be noted that in order to simplify the presentation disclosed in this specification and thereby aid in understanding one or more inventive embodiments, various features are sometimes grouped together in a single embodiment, figure, or description thereof. This method of disclosure, however, is not intended to imply that more features than are presented in the claims are required for the present description. Indeed, less than all of the features of a single embodiment disclosed above.
In some embodiments, numbers describing the components, number of attributes are used, it being understood that such numbers being used in the description of embodiments are modified in some examples by the modifier "about," approximately, "or" substantially. Unless otherwise indicated, "about," "approximately," or "substantially" indicate that the number allows for a 20% variation. Accordingly, in some embodiments, numerical parameters set forth in the specification and claims are approximations that may vary depending upon the desired properties sought to be obtained by the individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and employ a method for preserving the general number of digits. Although the numerical ranges and parameters set forth herein are approximations that may be employed in some embodiments to confirm the breadth of the range, in particular embodiments, the setting of such numerical values is as precise as possible.
Each patent, patent application publication, and other material, such as articles, books, specifications, publications, documents, etc., referred to in this specification is incorporated herein by reference in its entirety. Except for application history documents that are inconsistent or conflicting with the content of this specification, documents that are currently or later attached to this specification in which the broadest scope of the claims to this specification is limited are also. It is noted that, if the description, definition, and/or use of a term in an attached material in this specification does not conform to or conflict with the content of this specification, the description, definition, and/or use of the term in this specification controls.
Finally, it should be understood that the embodiments in this specification are merely illustrative of the principles of the embodiments in this specification. Other variations are possible within the scope of this description. Thus, by way of example, and not limitation, alternative configurations of embodiments of the present specification may be considered as consistent with the teachings of the present specification. Accordingly, the embodiments of the present specification are not limited to only the embodiments explicitly described and depicted in the present specification.
Claims (9)
1. A door check method comprising:
acquiring side face point cloud data corresponding to the side face of the box body;
extracting corner features based on the side point cloud data; the corner features comprise the number of corner points of a target area in the side point cloud data;
and judging whether the side surface is a box door side of the box body or not based on the corner characteristics.
2. The method of claim 1, wherein the determining whether the side is a door side of the cabinet based on the corner feature comprises:
determining the side surface to be the door side of the box body in response to the number of the corner points meeting a first preset condition;
and determining the side surface to be a non-box door side of the box body in response to the number of the corner points not meeting the first preset condition.
3. The method of claim 1, wherein,
the side face point cloud data comprise first side face point cloud data corresponding to a first side of the box body and second side face point cloud data corresponding to a second side of the box body;
based on the corner feature, judging whether the side is a door side of the box body or not, including:
determining that the first side is the box door side in response to the difference characteristic of the number of corner points of the first side point cloud data and the number of corner points of the second side point cloud data meeting a second preset condition;
and determining that the second side is the box door side in response to the difference characteristic of the number of the corner points of the second side point cloud data and the number of the corner points of the first side point cloud data meeting the second preset condition.
4. The method of claim 1, wherein the corner features comprise a corner distribution of corners in the side point cloud data.
5. The method of claim 4, wherein the determining whether the side is a door side of the cabinet based on the corner feature comprises:
searching in a corner database based on the corner distribution of the side point cloud data, and determining whether reference data of the same type of box body exists in the corner database, wherein the reference data comprises box door side reference corner distribution and non-box door side reference corner distribution;
and respectively comparing the angular point distribution with the box door side reference angular point distribution and the non-box door side reference angular point distribution in response to the existence of the reference data of the box bodies with the same type in the angular point database, and judging whether the side corresponding to the side point cloud data is the box door side.
6. The method of claim 5, further comprising:
acquiring second side point cloud data and corner characteristics thereof corresponding to a second side of the box body in response to the fact that reference data of the box body of the same type does not exist in the corner database;
determining a box door side and a non-box door side of the box body based on the corner features of the side point cloud data and the corner features of the second side point cloud data;
and obtaining the model of the box body, and storing at least the model of the box body, the angular point distribution of the box door side and/or the angular point distribution of the non-box door side into the angular point database.
7. A door check system, comprising:
the point cloud acquisition module is used for acquiring side point cloud data corresponding to the side of the box body;
the corner extraction module is used for extracting corner features based on the side face point cloud data; the corner features comprise the number of corner points of a target area in the side point cloud data;
and the box door judging module is used for judging whether the side surface is the box door side of the box body based on the corner characteristics.
8. A door check apparatus comprising a processor for performing the door check method according to any one of claims 1 to 6.
9. A computer-readable storage medium storing computer instructions that, when read by a computer in the storage medium, the computer performs the door detection method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311016523.0A CN116777903B (en) | 2023-08-11 | 2023-08-11 | Box door detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311016523.0A CN116777903B (en) | 2023-08-11 | 2023-08-11 | Box door detection method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116777903A CN116777903A (en) | 2023-09-19 |
| CN116777903B true CN116777903B (en) | 2024-01-26 |
Family
ID=88013641
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311016523.0A Active CN116777903B (en) | 2023-08-11 | 2023-08-11 | Box door detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116777903B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109978954A (en) * | 2019-01-30 | 2019-07-05 | 杭州飞步科技有限公司 | The method and apparatus of radar and camera combined calibrating based on cabinet |
| CN112816976A (en) * | 2021-01-29 | 2021-05-18 | 三一海洋重工有限公司 | Container door orientation detection method and system, storage medium and electronic device |
| CN112991429A (en) * | 2019-12-13 | 2021-06-18 | 顺丰科技有限公司 | Box volume measuring method and device, computer equipment and storage medium |
| CN113415725A (en) * | 2021-06-17 | 2021-09-21 | 三一海洋重工有限公司 | Alignment method, alignment controller and hoisting mechanism |
| CN113705557A (en) * | 2021-08-30 | 2021-11-26 | 上海西井信息科技有限公司 | Method, system, equipment and storage medium for detecting door post of rear door of container |
| CN114180458A (en) * | 2021-11-04 | 2022-03-15 | 三一海洋重工有限公司 | Container hoisting equipment, hoisting control method, computer equipment and storage medium |
| CN115490150A (en) * | 2022-08-31 | 2022-12-20 | 西门子（中国）有限公司 | Lifting device for a container and method for detecting the orientation of a container |
| WO2023024087A1 (en) * | 2021-08-27 | 2023-03-02 | 深圳市速腾聚创科技有限公司 | Method, apparatus and device for processing laser radar point cloud, and storage medium |
| CN116109915A (en) * | 2023-04-17 | 2023-05-12 | 济宁能源发展集团有限公司 | Intelligent recognition method for container door state |
| CN116524166A (en) * | 2023-04-20 | 2023-08-01 | 三一海洋重工有限公司 | Container door orientation identification method, device and port automation system |
-
2023
- 2023-08-11 CN CN202311016523.0A patent/CN116777903B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109978954A (en) * | 2019-01-30 | 2019-07-05 | 杭州飞步科技有限公司 | The method and apparatus of radar and camera combined calibrating based on cabinet |
| CN112991429A (en) * | 2019-12-13 | 2021-06-18 | 顺丰科技有限公司 | Box volume measuring method and device, computer equipment and storage medium |
| CN112816976A (en) * | 2021-01-29 | 2021-05-18 | 三一海洋重工有限公司 | Container door orientation detection method and system, storage medium and electronic device |
| CN113415725A (en) * | 2021-06-17 | 2021-09-21 | 三一海洋重工有限公司 | Alignment method, alignment controller and hoisting mechanism |
| WO2023024087A1 (en) * | 2021-08-27 | 2023-03-02 | 深圳市速腾聚创科技有限公司 | Method, apparatus and device for processing laser radar point cloud, and storage medium |
| CN113705557A (en) * | 2021-08-30 | 2021-11-26 | 上海西井信息科技有限公司 | Method, system, equipment and storage medium for detecting door post of rear door of container |
| CN114180458A (en) * | 2021-11-04 | 2022-03-15 | 三一海洋重工有限公司 | Container hoisting equipment, hoisting control method, computer equipment and storage medium |
| CN115490150A (en) * | 2022-08-31 | 2022-12-20 | 西门子（中国）有限公司 | Lifting device for a container and method for detecting the orientation of a container |
| CN116109915A (en) * | 2023-04-17 | 2023-05-12 | 济宁能源发展集团有限公司 | Intelligent recognition method for container door state |
| CN116524166A (en) * | 2023-04-20 | 2023-08-01 | 三一海洋重工有限公司 | Container door orientation identification method, device and port automation system |
Non-Patent Citations (2)
| Title |
|---|
| Smart port and machine vision;Weijian Mi，Yuan Liu;《Smart Ports》;99–125页 * |
| 点云的边界提取及角点检测算法研究;杨雪娇;《中国优秀硕士学位论文全文数据库 信息科技辑》;第2011年卷(第05期);第5章 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116777903A (en) | 2023-09-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11403839B2 (en) | Commodity detection terminal, commodity detection method, system, computer device, and computer readable medium | |
| CN114981840A (en) | Ground segmentation method and device based on point cloud data and computer equipment | |
| Guo et al. | A method of automatic cell counting based on microscopic image | |
| CN111126393A (en) | Vehicle appearance refitting judgment method and device, computer equipment and storage medium | |
| CN111310753B (en) | Meter alignment method and device | |
| CN107004266A (en) | Method for detecting defects on the surface of a tyre | |
| KR101997048B1 (en) | Method for recognizing distant multiple codes for logistics management and code recognizing apparatus using the same | |
| US20170178341A1 (en) | Single Parameter Segmentation of Images | |
| Galsgaard et al. | Circular hough transform and local circularity measure for weight estimation of a graph-cut based wood stack measurement | |
| CN115546202B (en) | Tray detection and positioning method for unmanned forklift | |
| CN108664970A (en) | A kind of fast target detection method, electronic equipment, storage medium and system | |
| CN108805872A (en) | Product detection method and device | |
| CN113128346B (en) | Target identification method, system and device for crane construction site and storage medium | |
| CN112686162B (en) | Method, device, equipment and storage medium for detecting clean state of warehouse environment | |
| CN116777903B (en) | Box door detection method and system | |
| CN103544495A (en) | Method and system for recognizing of image categories | |
| CN117745780A (en) | Outdoor large scene 3D point cloud registration method based on isolated cluster removal | |
| CN117237681A (en) | Image processing method, device and related equipment | |
| CN114202548A (en) | Forklift pallet positioning method and device, storage medium and electronic equipment | |
| CN112668365A (en) | Material warehousing identification method, device, equipment and storage medium | |
| CN115984759A (en) | Substation switch state identification method and device, computer equipment and storage medium | |
| CN115100271A (en) | Method and device for detecting goods taking height, computer equipment and storage medium | |
| Gollub et al. | A partitioned approach for efficient graph–based place recognition | |
| US10339364B2 (en) | Apparatus and method for rejecting erroneous objects of extrusion in point cloud data | |
| CN114581360A (en) | Photovoltaic module label detection method, device, equipment and computer storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |