CN117495843A - Aggregate grading intelligent detection method system and equipment with different depth of field in storage bin - Google Patents
Aggregate grading intelligent detection method system and equipment with different depth of field in storage bin Download PDFInfo
- Publication number
- CN117495843A CN117495843A CN202311687791.5A CN202311687791A CN117495843A CN 117495843 A CN117495843 A CN 117495843A CN 202311687791 A CN202311687791 A CN 202311687791A CN 117495843 A CN117495843 A CN 117495843A
- Authority
- CN
- China
- Prior art keywords
- aggregate
- perimeter
- field
- real
- determining
- 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.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 57
- 238000003860 storage Methods 0.000 title description 9
- 230000011218 segmentation Effects 0.000 claims abstract description 46
- 239000002245 particle Substances 0.000 claims abstract description 41
- 238000000034 method Methods 0.000 claims description 34
- 230000009466 transformation Effects 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 12
- 239000011159 matrix material Substances 0.000 claims description 8
- 238000007476 Maximum Likelihood Methods 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 5
- 238000001914 filtration Methods 0.000 claims description 4
- 238000002372 labelling Methods 0.000 claims 1
- 238000004519 manufacturing process Methods 0.000 abstract description 5
- 230000007547 defect Effects 0.000 abstract description 2
- 238000006243 chemical reaction Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000012360 testing method Methods 0.000 description 5
- 238000012937 correction Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 239000010426 asphalt Substances 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 239000008187 granular material Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000004570 mortar (masonry) Substances 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000013519 translation Methods 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
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- 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
- G06T2207/30108—Industrial image inspection
- G06T2207/30132—Masonry; Concrete
-
- 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
- G06T2207/30244—Camera pose
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Geometry (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the field of intelligent manufacturing, and provides an intelligent aggregate grading detection method, system and equipment with different depth of field in a bin, wherein the intelligent detection method comprises the following steps: carrying out semantic segmentation on the acquired image based on a pre-constructed semantic segmentation model, and determining the pixel area and the pixel perimeter of each aggregate in the image; determining the real area of the aggregate based on the corresponding relation between the pixel area and the real area of the aggregate and the pixel area of the aggregate under different depths of field; determining the real perimeter of the aggregate based on the corresponding relation between the pixel perimeter and the real perimeter of the aggregate under different depths of field; and determining the equivalent particle size of each aggregate based on the real area and the real perimeter of the aggregate, and finishing the intelligent detection of the aggregate grading. The automatic detection device is used for solving the defects that in the prior art, aggregate grading cannot be detected timely, production efficiency is low, and aggregate grading accuracy is low, and accurate intelligent detection can be carried out on aggregates in different depths in a bin.
Description
Technical Field
The invention relates to the technical field of intelligent manufacturing, in particular to an aggregate grading intelligent detection method, system and equipment with different depth of field in a bin.
Background
In the fields of construction, road construction and the like, the grading distribution of aggregate is an important quality parameter, which influences the performance and stability of materials such as concrete, asphalt and the like. Currently, most aggregate grading detection methods rely on laboratory testing, which greatly limits the ability to monitor and control in real time. In addition, variations in aggregate grading at different depths of field within the silo can also be a challenge for quality control. Conventionally, the detection of aggregate gradation mainly depends on manual sampling and analysis in a laboratory, and laboratory testing requires a certain time, so that the change condition of aggregate gradation cannot be obtained in real time, which results in difficulty in timely adjustment in the production process. The increase or decrease of the aggregate in the bin inevitably leads to the change of the height of the aggregate in the bin, but no method or device for the intelligent non-contact type aggregate grading detection aiming at different depth of field exists at present.
Disclosure of Invention
The invention provides an intelligent detection method, system and equipment for aggregate gradation of different depth of field in a bin, which are used for solving the defects that in the prior art, aggregate gradation cannot be detected in time, production efficiency is low, and aggregate gradation accuracy is low, and accurate intelligent detection can be carried out on aggregates of different depth of field in the bin.
The invention provides an intelligent aggregate grading detection method with different depth of field in a bin, which comprises the following steps:
collecting images of aggregates in a bin;
performing semantic segmentation on the image based on a pre-constructed semantic segmentation model, and determining the pixel area and the pixel perimeter of each aggregate in the image;
determining the real area of the aggregate based on the corresponding relation between the pixel area and the real area of the aggregate and the pixel area of the aggregate under different depths of field;
determining the real perimeter of the aggregate based on the corresponding relation between the pixel perimeter and the real perimeter of the aggregate under different depths of field;
and determining the equivalent particle diameter of each aggregate based on the real area and the real perimeter of the aggregate, and determining the aggregate grading in the bin based on the equivalent particle diameter of each aggregate in the bin.
According to the aggregate gradation intelligent detection method for different depth of field in the bin, provided by the invention, the aggregate gradation in the bin is determined based on the equivalent particle size of each aggregate, and the method comprises the following steps:
and generating a grading curve of the aggregates in the bin based on the equivalent particle size of each aggregate.
According to the aggregate grading intelligent detection method with different depth of field in the bin, provided by the invention, the semantic segmentation model is obtained by training in the following manner:
constructing a semantic segmentation data set based on aggregate image data with different particle sizes;
training the semantic segmentation data set pair to obtain a semantic segmentation model.
According to the aggregate grading intelligent detection method with different depth of field in the bin, provided by the invention, the semantic segmentation data set is constructed based on aggregate image data with different particle sizes, and the method comprises the following steps:
aggregate images with different particle sizes are collected, and median filtering and Laplacian transformation are carried out on the aggregate images;
and marking and expanding the aggregate image to obtain a semantic segmentation data set.
According to the aggregate grading intelligent detection method for different depth of field in the bin, provided by the invention, the corresponding relation between the pixel area and the real area of the aggregate under different depth of field is determined by the following modes:
shooting a standard cube with a determined size under different depth of field;
determining the real area and the pixel area of cubes in images shot under different depths of field;
based on the ratio of the real area to the pixel area of the cube under different depth of field, the corresponding relation between the pixel area and the real area of the aggregate is established.
According to the aggregate grading intelligent detection method with different depth of field in the bin, provided by the invention, the corresponding relation between the pixel perimeter and the real perimeter of the aggregate under different depth of field is determined by the following modes:
shooting a standard cube with a determined size under different depth of field;
determining the real perimeter and the pixel perimeter of a cube in images shot under different depths of field;
based on the ratio of the real perimeter to the pixel perimeter of the cubic block under different depth of field, the corresponding relation between the pixel perimeter and the real perimeter of the aggregate is established.
The aggregate grading intelligent detection method for different depth of field in the bin provided by the invention further comprises the following steps:
determining a transformation matrix and a distortion coefficient of a camera used for acquiring images of aggregate in the bin, and converting a coordinate system of the camera;
and correcting the image shot by the camera based on a least square method and a maximum likelihood optimization method.
According to the aggregate grading intelligent detection method with different depth of field in the bin, provided by the invention, the camera is subjected to coordinate system conversion, and the method comprises the following steps:
coordinate system conversion is performed using the following formula (1):
（1）
wherein the method comprises the steps ofRepresenting coordinates of a point in the image in the three-dimensional camera coordinate system,/for>Transformation matrix representing camera,/, for>Representing coordinates of a point in the image in a three-dimensional world coordinate system,/->The coordinates of the camera in a three-dimensional world coordinate system are shown.
According to the aggregate grading intelligent detection method with different depth of field in the bin, provided by the invention, the equivalent particle diameter of each aggregate is determined based on the real area and the real perimeter of the aggregate, and the method comprises the following steps:
the equivalent particle diameter of the aggregate is determined by the following formulas (2), (3) and (4):
（2）
（3）
（4）
wherein S is the real area of the aggregate, and L is the real perimeter of the aggregate.
The invention also provides an aggregate grading intelligent detection system with different depth of field in the bin, which comprises: the collecting module is used for collecting images of aggregate in the bin;
the first determining module is used for carrying out semantic segmentation on the image based on a pre-constructed semantic segmentation model and determining the pixel area and the pixel perimeter of each aggregate in the image;
the second determining module is used for determining the real area of the aggregate based on the corresponding relation between the pixel area and the real area of the aggregate and the pixel area of the aggregate under different depths of field;
the third determining module is used for determining the real perimeter of the aggregate based on the corresponding relation between the pixel perimeter and the real perimeter of the aggregate and the pixel perimeter of the aggregate under different depths of field;
and a fourth determining module for determining the equivalent particle diameter of each aggregate based on the real area and the real perimeter of the aggregate, thereby completing the intelligent detection of the aggregate grading.
The invention also provides an electronic device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the intelligent aggregate grading detection method with different depth of field in any bin when executing the program.
The invention also provides a non-transitory computer readable storage medium, on which a computer program is stored, which when executed by a processor implements the aggregate gradation intelligent detection method of different depths of field in any bin as described above.
The invention also provides a computer program product, which comprises a computer program, and the computer program realizes the intelligent aggregate grading detection method with different depth of field in any bin when being executed by a processor.
According to the intelligent aggregate grading detection method for the different depth of field in the bin, provided by the invention, the image of the aggregate collected by the camera can be subjected to semantic segmentation, single aggregate can be segmented from the image, the pixel area and the pixel perimeter of the segmented aggregate can be calculated, and the calculated pixel area and pixel perimeter are not necessarily consistent with the real area and the real perimeter of the aggregate, so that the real area and the real perimeter of the aggregate are required to be determined based on the pixel area and the pixel perimeter, and then the equivalent particle diameter of the aggregate can be accurately determined based on the real area and the real perimeter, and further the intelligent aggregate grading detection is accurately completed.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of an intelligent detection method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a grading curve provided by an embodiment of the present invention;
FIG. 3 is a schematic diagram of an intelligent detection system according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Fig. 1 is a schematic flow chart of an intelligent detection method according to an embodiment of the present invention.
As shown in fig. 1, the embodiment provides an intelligent aggregate grading detection method with different depth of field in a bin, which comprises the following steps:
step 101, collecting images of aggregates in a bin;
102, carrying out semantic segmentation on an image based on a pre-constructed semantic segmentation model, and determining the pixel area and the pixel perimeter of each aggregate in the image;
step 103, determining the real area of the aggregate based on the corresponding relation between the pixel area and the real area of the aggregate and the pixel area of the aggregate under different depths of field;
104, determining the real perimeter of the aggregate based on the corresponding relation between the pixel perimeter and the real perimeter of the aggregate under different depths of field and the pixel perimeter of the aggregate;
and 105, determining the equivalent particle size of each aggregate based on the real area and the real perimeter of the aggregate, and determining the aggregate grading in the bin based on the equivalent particle size of each aggregate in the bin.
Wherein aggregate refers to granular materials in concrete and mortar which play a role in framework and filling.
The aggregate grading refers to the proportion relationship of different aggregates matched with each other according to the size relationship.
The depth of field refers to the distance between the foreground and the background in the image taken by the camera, which is related to where the plane in which the camera is focused, and in this embodiment, the distance of the camera from the aggregate may be used to represent the depth of field.
In implementation, the camera for shooting the image can be arranged right above the bin, and the image of the aggregate in the bin is shot from top to bottom, so that the situation that the distance between the camera and the aggregate changes can be generated along with the increase and the decrease of the aggregate in the bin.
Fig. 2 is a schematic diagram of a grading curve provided by an embodiment of the present invention.
In practical application, as shown in fig. 2, after determining the equivalent particle diameter of each aggregate, a grading curve can be generated based on the equivalent particle diameter of each aggregate, and the grading curve can be used for representing grading conditions of aggregates in the bin.
Wherein figure 2 provides a grading curve for aggregate in a standard case 10mm-20mm silo.
In the aggregate grading intelligent detection method with different depth of field in the bin, semantic segmentation can be performed on the image of the aggregate acquired by the camera, single aggregate can be segmented from the image, the pixel area and the pixel perimeter of the segmented aggregate can be calculated, the calculated pixel area and pixel perimeter are not necessarily consistent with the real area and the real perimeter of the aggregate, so that the real area and the real perimeter of the aggregate are required to be determined based on the pixel area and the pixel perimeter, then the equivalent particle diameter of the aggregate can be accurately determined based on the real area and the real perimeter, and further the aggregate grading intelligent detection is accurately completed.
In an exemplary embodiment, a radar camera may be used to measure the distance of the camera from the bin aggregate plane, which may characterize the depth of field of the camera.
In an exemplary embodiment, the semantic segmentation model is trained by:
constructing a semantic segmentation data set based on aggregate image data with different particle sizes;
training the semantic segmentation data set pair to obtain a semantic segmentation model.
In practical application, the training set can be subjected to semantic segmentation training based on the PyTorch deep learning framework and the U-Net network structure, so that the accuracy of the semantic segmentation model obtained through training can be improved.
In an exemplary embodiment, determining the pixel area and the pixel perimeter of each aggregate in the image includes:
extracting the outline in the semantically segmented image by using a findContours module in an OpenCV software library in a Python program, calculating the pixel area of the outline of each aggregate by using a contourArea module in the OpenCV software library according to the outline of the aggregate, and calculating the pixel perimeter of the outline of each aggregate by using an arcLength module in the OpenCV software library.
In an exemplary embodiment, constructing a semantic segmentation dataset based on aggregate image data of different particle sizes includes:
aggregate images with different particle sizes are collected, and median filtering and Laplacian transformation are carried out on the aggregate images;
and marking and expanding the aggregate image to obtain a semantic segmentation data set.
In practical application, aggregates in the image can be marked by a marking tool Labelme, and the data set is expanded by means of overturning, rotating, cutting, scaling and translation conversion of the image. And dividing the data set into a training set, a verification set and a test set, wherein the training set can be used for training in the process of training based on the semantic segmentation data set to obtain a semantic segmentation model, and the verification set and the test set are used for verifying and testing the semantic segmentation model obtained by training.
In an exemplary embodiment, the correspondence between the pixel area and the real area of the aggregate under different depths of field is determined by:
shooting a standard cube with a determined size under different depth of field;
determining the real area and the pixel area of cubes in images shot under different depths of field;
based on the ratio of the real area to the pixel area of the cube under different depth of field, the corresponding relation between the pixel area and the real area of the aggregate is established.
In an exemplary embodiment, the correspondence between the pixel perimeter and the real perimeter of the aggregate at different depths of field is determined by:
shooting a standard cube with a determined size under different depth of field;
determining the real perimeter and the pixel perimeter of a cube in images shot under different depths of field;
based on the ratio of the real perimeter to the pixel perimeter of the cubic block under different depth of field, the corresponding relation between the pixel perimeter and the real perimeter of the aggregate is established.
In practical application, the corresponding relation between the pixel area and the real area of the aggregate and the corresponding relation between the pixel perimeter and the real perimeter of the aggregate can be established simultaneously, specifically, the same 1cm×1cm black standard block can be shot by a camera at different distances H, an image is shot every 10cm, and black is calculatedThe ratio of the actual area of the standard block to the pixel area and the ratio of the actual circumference of the black standard block to the circumference of the pixel are respectively fitted to obtain the ratio of the actual area of the aggregate to the pixel area through two sets of dataFunctional relation to camera and aggregate plane distance H>And the ratio of the actual perimeter of the aggregate to the perimeter of the pixel +.>Functional relation to camera and aggregate plane distance H>。
Thus, the actual area of each aggregate in the bin isThe actual perimeter of the aggregate is。
In an exemplary embodiment, further comprising:
determining a transformation matrix and a distortion coefficient of a camera used for acquiring images of aggregate in the bin, and converting a coordinate system of the camera;
and correcting the image shot by the camera based on a least square method and a maximum likelihood optimization method.
In an exemplary embodiment, coordinate system conversion of a camera includes:
coordinate system conversion is performed using the following formula (1):
（1）
wherein the method comprises the steps ofRepresenting the sitting of points in an image in a three-dimensional camera coordinate systemMark (I) of->Transformation matrix representing camera,/, for>Representing coordinates of a point in the image in a three-dimensional world coordinate system,/->Representing the coordinates of the camera in a three-dimensional world coordinate system.
In an exemplary embodiment, correcting an image captured by a camera based on a least square method and a maximum likelihood optimization method includes:
optimizing and estimating by using a least square method and a maximum likelihood method to obtain actual radial distortion coefficients, internal parameters, external parameters and distortion coefficients;
and carrying out distortion correction on the image shot by the camera based on the radial distortion coefficient, the internal parameter, the external parameter and the distortion coefficient of the camera.
In an exemplary embodiment, determining an equivalent particle size for each aggregate based on the real area and the real perimeter of the aggregate includes:
the equivalent particle diameter of the aggregate is determined by the following formulas (2), (3) and (4):
（2）
（3）
（4）
wherein the method comprises the steps ofFor the real area of aggregate, +.>Is the real perimeter of the aggregate.
The aggregate gradation intelligent detection system with different depth of field in the bin provided by the invention is described below, and the aggregate gradation intelligent detection system with different depth of field in the bin described below and the aggregate gradation intelligent detection method with different depth of field in the bin described above can be referred to correspondingly.
Fig. 3 is a schematic structural diagram of an intelligent detection system according to an embodiment of the present invention.
As shown in fig. 3, the aggregate grading intelligent detection system with different depth of field in the bin provided in this embodiment includes:
the acquisition module 301 is used for acquiring images of aggregate in the bin;
a first determining module 302, configured to perform semantic segmentation on the image based on a pre-constructed semantic segmentation model, and determine a pixel area and a pixel perimeter of each aggregate in the image;
a second determining module 303, configured to determine a real area of the aggregate based on a correspondence between pixel areas of the aggregate and the real area under different depths of field and the pixel areas of the aggregate;
a third determining module 304, configured to determine the real perimeter of the aggregate based on the correspondence between the pixel perimeter and the real perimeter of the aggregate and the pixel perimeter of the aggregate under different depths of field;
and a fourth determining module 305, determining the equivalent particle diameter of each aggregate based on the real area and the real perimeter of the aggregate, and completing the intelligent aggregate grading detection.
In an exemplary embodiment, the second determining module is specifically configured to:
aggregate images with different particle sizes are collected, and median filtering and Laplacian transformation are carried out on the aggregate images;
and marking and expanding the aggregate image to obtain a semantic segmentation data set.
In an exemplary embodiment, the aggregate grading intelligent detection system with different depths of field in the bin further comprises a correction module, wherein the correction module is specifically used for:
determining a transformation matrix and a distortion coefficient of a camera used for acquiring images of aggregate in the bin, and converting a coordinate system of the camera;
and correcting the image shot by the camera based on a least square method and a maximum likelihood optimization method.
In an exemplary embodiment, the correction module is further configured to:
coordinate system conversion is performed using the following formula (1):
（1）
wherein the method comprises the steps ofRepresenting coordinates of a point in the image in the three-dimensional camera coordinate system,/for>Transformation matrix representing camera,/, for>Representing coordinates of a point in the image in a three-dimensional world coordinate system,/->Representing the coordinates of the camera in a three-dimensional world coordinate system.
In an exemplary embodiment, the fourth determining module is specifically configured to:
the equivalent particle diameter of the aggregate is determined by the following formulas (2), (3) and (4):
（2）
（3）
（4）
wherein the method comprises the steps ofFor the real area of aggregate, +.>Is the real perimeter of the aggregate.
The specific implementation method of the aggregate grading intelligent detection system with different depth of field in the bin provided by the embodiment can be implemented with reference to the above embodiment, and will not be repeated here.
Fig. 4 illustrates a physical schematic diagram of an electronic device, as shown in fig. 4, which may include: processor 410, communication interface (Communications Interface) 420, memory 430 and communication bus 440, wherein processor 410, communication interface 420 and memory 430 communicate with each other via communication bus 440. The processor 410 may invoke logic instructions in the memory 430 to perform an intelligent aggregate grading detection method for different depths of field within the silo, the method comprising:
collecting images of aggregates in a bin;
performing semantic segmentation on the image based on a pre-constructed semantic segmentation model, and determining the pixel area and the pixel perimeter of each aggregate in the image;
determining the real area of the aggregate based on the corresponding relation between the pixel area and the real area of the aggregate and the pixel area of the aggregate under different depths of field;
determining the real perimeter of the aggregate based on the corresponding relation between the pixel perimeter and the real perimeter of the aggregate under different depths of field;
and determining the equivalent particle diameter of each aggregate based on the real area and the real perimeter of the aggregate, and determining the aggregate grading in the bin based on the equivalent particle diameter of each aggregate in the bin.
Further, the logic instructions in the memory 430 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method of the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product, where the computer program product includes a computer program, where the computer program can be stored on a non-transitory computer readable storage medium, and when the computer program is executed by a processor, the computer can execute the aggregate grading intelligent detection method for different depths of field in a bin provided by the above methods, where the method includes:
collecting images of aggregates in a bin;
performing semantic segmentation on the image based on a pre-constructed semantic segmentation model, and determining the pixel area and the pixel perimeter of each aggregate in the image;
determining the real area of the aggregate based on the corresponding relation between the pixel area and the real area of the aggregate and the pixel area of the aggregate under different depths of field;
determining the real perimeter of the aggregate based on the corresponding relation between the pixel perimeter and the real perimeter of the aggregate under different depths of field;
and determining the equivalent particle diameter of each aggregate based on the real area and the real perimeter of the aggregate, and determining the aggregate grading in the bin based on the equivalent particle diameter of each aggregate in the bin.
In still another aspect, the present invention further provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the aggregate gradation intelligent detection method for different depths of field in a silo provided by the above methods, the method comprising:
collecting images of aggregates in a bin;
performing semantic segmentation on the image based on a pre-constructed semantic segmentation model, and determining the pixel area and the pixel perimeter of each aggregate in the image;
determining the real area of the aggregate based on the corresponding relation between the pixel area and the real area of the aggregate and the pixel area of the aggregate under different depths of field;
determining the real perimeter of the aggregate based on the corresponding relation between the pixel perimeter and the real perimeter of the aggregate under different depths of field;
and determining the equivalent particle diameter of each aggregate based on the real area and the real perimeter of the aggregate, and determining the aggregate grading in the bin based on the equivalent particle diameter of each aggregate in the bin.
The apparatus embodiments described above are merely illustrative, wherein elements illustrated as separate elements may or may not be physically separate, and elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on such understanding, the foregoing technical solutions may be embodied essentially or in part in the form of a software product, which may be stored in a computer-readable storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to perform the various embodiments or methods of some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. An intelligent aggregate grading detection method with different depth of field in a bin is characterized by comprising the following steps:
collecting images of aggregates in a bin;
performing semantic segmentation on the image based on a pre-constructed semantic segmentation model, and determining the pixel area and the pixel perimeter of each aggregate in the image;
determining the real area of the aggregate based on the corresponding relation between the pixel area and the real area of the aggregate under different depths of field and the pixel area of the aggregate;
determining the real perimeter of the aggregate based on the corresponding relation between the pixel perimeter and the real perimeter of the aggregate under different depth of field and the pixel perimeter of the aggregate;
and determining the equivalent particle diameter of each aggregate based on the real area and the real perimeter of the aggregate, and determining the aggregate grading in the bin based on the equivalent particle diameter of each aggregate in the bin.
2. The intelligent detection method for aggregate gradation of different depth of field in a silo according to claim 1, wherein the determining the aggregate gradation in the silo based on the equivalent particle size of each aggregate comprises:
and generating a grading curve of the aggregate in the bin based on the equivalent particle size of each aggregate.
3. The intelligent aggregate grading detection method for different depths of field in a bin according to claim 1, wherein the semantic segmentation model is trained by the following modes:
constructing a semantic segmentation data set based on aggregate image data with different particle sizes;
training the semantic segmentation data set pair to obtain the semantic segmentation model.
4. The intelligent aggregate gradation detection method of different depths of field in a silo according to claim 3, wherein the constructing a semantic segmentation dataset based on aggregate image data of different particle sizes comprises:
aggregate images with different particle diameters are collected, and median filtering and Laplace transformation are carried out on the aggregate images;
and labeling the aggregate image and performing expansion processing to obtain the semantic segmentation data set.
5. The intelligent aggregate grading detection method for different depths of field in a bin according to claim 1, wherein the correspondence between the pixel area and the real area of the aggregate under the different depths of field is determined by the following method:
shooting a standard cube with a determined size under different depth of field;
determining the real area and the pixel area of the cube in the images shot under different depths of field;
based on the ratio of the real area to the pixel area of the cube under different depth of field, the corresponding relation between the pixel area and the real area of the aggregate is established.
6. The intelligent aggregate grading detection method of different depth of field in a bin according to claim 1, wherein the correspondence between the pixel perimeter and the real perimeter of the aggregate under the different depth of field is determined by the following method:
shooting a standard cube with a determined size under different depth of field;
determining the real perimeter and the pixel perimeter of the cubic block in the images shot under different depths of field;
based on the ratio of the real perimeter to the pixel perimeter of the cubic block under different depth of field, the corresponding relation between the pixel perimeter and the real perimeter of the aggregate is established.
7. The intelligent aggregate grading detection method for different depths of field in a bin according to claim 1, further comprising:
determining a transformation matrix and a distortion coefficient of a camera used for acquiring images of aggregate in a bin, and performing coordinate system transformation on the camera based on the following formula (1)
（1）
Wherein the method comprises the steps ofRepresenting coordinates of a point in the image in the three-dimensional camera coordinate system,/for>A transformation matrix representing the camera, +.>Representing coordinates of a point in the image in a three-dimensional world coordinate system,/->Representing coordinates of the camera in a three-dimensional world coordinate system;
and correcting the image shot by the camera based on a least square method and a maximum likelihood optimization method.
8. The intelligent detection method for aggregate grading with different depth of field in a silo according to claim 1, wherein the determining the equivalent particle diameter of each aggregate based on the real area and the real perimeter of the aggregate comprises:
the equivalent particle diameter of the aggregate was determined by the following formulas (2), (3) and (4):
（2）
（3）
（4）
wherein S is the real area of the aggregate, and L is the real perimeter of the aggregate.
9. Aggregate grading intelligent detection system of different depth of field in feed bin, its characterized in that includes:
the collecting module is used for collecting images of aggregate in the bin;
the first determining module is used for carrying out semantic segmentation on the image based on a pre-constructed semantic segmentation model and determining the pixel area and the pixel perimeter of each aggregate in the image;
the second determining module is used for determining the real area of the aggregate based on the corresponding relation between the pixel area and the real area of the aggregate under different depth of field and the pixel area of the aggregate;
the third determining module is used for determining the real perimeter of the aggregate based on the corresponding relation between the pixel perimeter and the real perimeter of the aggregate under different depth of field and the pixel perimeter of the aggregate;
and a fourth determining module for determining the equivalent particle diameter of each aggregate based on the real area and the real perimeter of the aggregate, thereby completing the intelligent detection of aggregate grading.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the aggregate grading intelligent detection method for different depths of field in a bin according to any one of claims 1 to 8 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311687791.5A CN117495843A (en) | 2023-12-11 | 2023-12-11 | Aggregate grading intelligent detection method system and equipment with different depth of field in storage bin |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311687791.5A CN117495843A (en) | 2023-12-11 | 2023-12-11 | Aggregate grading intelligent detection method system and equipment with different depth of field in storage bin |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117495843A true CN117495843A (en) | 2024-02-02 |
Family
ID=89671029
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311687791.5A Pending CN117495843A (en) | 2023-12-11 | 2023-12-11 | Aggregate grading intelligent detection method system and equipment with different depth of field in storage bin |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117495843A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110174334A (en) * | 2019-06-28 | 2019-08-27 | 华侨大学 | A kind of coarse aggregate quality morphology detection system and method |
| CN111105386A (en) * | 2019-03-28 | 2020-05-05 | 烟台大学 | Coarse aggregate quality image processing and analyzing method based on mobile equipment |
| CN112611690A (en) * | 2020-12-04 | 2021-04-06 | 华侨大学 | Coarse aggregate equivalent particle size grading method based on three-dimensional image |
| CN114092485A (en) * | 2021-09-28 | 2022-02-25 | 华侨大学 | Mask rcnn-based stacked coarse aggregate image segmentation method and system |
-
2023
- 2023-12-11 CN CN202311687791.5A patent/CN117495843A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111105386A (en) * | 2019-03-28 | 2020-05-05 | 烟台大学 | Coarse aggregate quality image processing and analyzing method based on mobile equipment |
| CN110174334A (en) * | 2019-06-28 | 2019-08-27 | 华侨大学 | A kind of coarse aggregate quality morphology detection system and method |
| CN112611690A (en) * | 2020-12-04 | 2021-04-06 | 华侨大学 | Coarse aggregate equivalent particle size grading method based on three-dimensional image |
| CN114092485A (en) * | 2021-09-28 | 2022-02-25 | 华侨大学 | Mask rcnn-based stacked coarse aggregate image segmentation method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111311650B (en) | Point cloud data registration method, device and storage medium | |
| CN108960135B (en) | Dense ship target accurate detection method based on high-resolution remote sensing image | |
| CN113344956B (en) | Ground feature contour extraction and classification method based on unmanned aerial vehicle aerial photography three-dimensional modeling | |
| CN112652015B (en) | BIM-based pavement disease marking method and device | |
| CN110008207B (en) | Airborne L iDAR point cloud data vulnerability rapid detection method based on density histogram | |
| CN111105452B (en) | Binocular vision-based high-low resolution fusion stereo matching method | |
| CN111582093A (en) | Automatic small target detection method in high-resolution image based on computer vision and deep learning | |
| CN108665468B (en) | Device and method for extracting tangent tower insulator string | |
| CN112329726B (en) | Face recognition method and device | |
| CN112163588A (en) | Intelligent evolution-based heterogeneous image target detection method, storage medium and equipment | |
| CN112819066A (en) | Res-UNet single tree species classification technology | |
| CN114494292A (en) | Method and system for extracting building facade glass area | |
| CN104299241A (en) | Remote sensing image significance target detection method and system based on Hadoop | |
| CN111104850A (en) | Remote sensing image building automatic extraction method and system based on residual error network | |
| CN116222381A (en) | Electrode coating size measurement method and device | |
| CN110047146B (en) | Error correction method based on single revolving body image 3D restoration | |
| WO2024125434A1 (en) | Regional-consistency-based building principal angle correction method | |
| CN114595238A (en) | Vector-based map processing method and device | |
| CN117710588A (en) | Three-dimensional target detection method based on visual ranging priori information | |
| CN112070735B (en) | Asphalt core sample image extraction method and system based on special convolution operator | |
| CN107341808B (en) | Visual detection system and measurement method for simulating lunar soil hardness based on rut image | |
| CN117495843A (en) | Aggregate grading intelligent detection method system and equipment with different depth of field in storage bin | |
| CN108335321B (en) | Automatic ground surface gravel size information extraction method based on multi-angle photos | |
| CN111336991A (en) | Tunnel ellipticity analysis method, device and system based on laser point cloud | |
| CN106355576A (en) | SAR image registration method based on MRF image segmentation algorithm |
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 |