CN114549993A - Method, system and device for scoring line segment image in experiment and readable storage medium - Google Patents
Method, system and device for scoring line segment image in experiment and readable storage medium Download PDFInfo
- Publication number
- CN114549993A CN114549993A CN202210186378.XA CN202210186378A CN114549993A CN 114549993 A CN114549993 A CN 114549993A CN 202210186378 A CN202210186378 A CN 202210186378A CN 114549993 A CN114549993 A CN 114549993A
- Authority
- CN
- China
- Prior art keywords
- information
- image
- line segment
- processing
- model
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
The invention relates to the technical field of information, in particular to a method, a system, equipment and a readable storage medium for grading a line segment image in an experiment, wherein the method is used for acquiring image information of a line segment and historical line segment image information; sending and inputting the image information of the line segment and the historical image information of the line segment into a training model for training to obtain all clear images in the image information of the line segment; extracting pixel points in all the clear images and filtering to obtain filtered segment pixel point information; preprocessing the filtered segment pixel point information to obtain end point information of the segment and cross point information of the segment; and then scoring the experimental images according to a preset scoring rule to obtain scoring information of each experimental image. The invention reduces the area of the line segment to be detected, avoids the interference of the environment to the line segment state detection, improves the detection precision, reduces the noise reduction difficulty and increases the line segment detection precision.
Description
Technical Field
The invention relates to the technical field of information, in particular to a method, a system and equipment for scoring a line segment image in an experiment and a readable storage medium.
Background
In the current middle school experiment examination, an artificial field scoring mode is mainly adopted, and professional teachers judge the correctness of the student experiment operation steps on the field, so that the method has the defects of difficult professional teacher and resource allocation, large invigilation workload, large subjective scoring difference and the like, and the realization of artificial intelligent examination reading is particularly urgent; in addition, many experiments comprise a step of drawing and connecting lines by paper pens, for example, the step of drawing is needed in the plane mirror imaging experiment, and the system capable of automatically judging the line segment state in the drawing of the examinee is needed for automatically scoring the drawing image of the plane mirror imaging.
Disclosure of Invention
The invention aims to provide a method, a system and equipment for grading a line segment image in an experiment and a readable storage medium, so as to improve the problems. In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
in one aspect, the present application provides a method for scoring a line segment image in an experiment, where the method includes: acquiring first information, wherein the first information is all image information acquired by a camera in an experimental process; sending the first information to a preprocessing model for processing to obtain second information, wherein the preprocessing model is a model for preprocessing the first information and deleting a preprocessed blurred image, and the second information is all clear images in the first information; sending the second information to a clustering model for processing to obtain third information, wherein the clustering model is a model for performing clustering analysis and binarization processing on pixel points in the second information, and the third information is binarized pixel point information; sending the third information to a refinement model for processing to obtain fourth information, wherein the refinement model is a model for performing at least one expansion processing and at least one corrosion processing on the third information, and the fourth information is line segment pixel point information with the width of only one pixel point; and grading the fourth information according to a preset grading rule to obtain fifth information, wherein the fifth information is the score information of each experimental image.
In a second aspect, the present application provides a system for detecting an image of a line segment in an experiment, where the system includes:
the first acquisition unit is used for acquiring first information, wherein the first information is all image information acquired by a camera in an experimental process;
the first processing unit is used for sending the first information to a preprocessing model for processing to obtain second information, the preprocessing model is a model for preprocessing the first information and deleting a preprocessed blurred image, and the second information is all clear images in the first information;
the first clustering unit is used for sending the second information to a clustering model for processing to obtain third information, the clustering model is a model for carrying out clustering analysis and binarization processing on pixel points in the second information, and the third information is binarized pixel point information;
the second processing unit is used for sending the third information to a refinement model for processing to obtain fourth information, the refinement model is a model for performing at least one expansion processing and at least one corrosion processing on the third information, and the fourth information is line segment pixel point information with the width of only one pixel point;
and the third processing unit is used for grading the fourth information according to a preset grading rule to obtain fifth information, wherein the fifth information is score information of each experimental image.
In a third aspect, an embodiment of the present application provides an apparatus for detecting a line segment image in an experiment, where the apparatus includes a memory and a processor. The memory is used for storing a computer program; the processor is used for implementing the steps of the grading method of the line segment images in the experiment when the computer program is executed.
In a fourth aspect, the present application provides a readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the method for scoring a line segment image in the above experiment.
The beneficial effects of the invention are as follows:
1. according to the invention, the image in the whole picture is cut by using the target detection model, so that the area of the line segment to be detected is reduced, the interference of the environment on the line segment state detection is avoided, and the detection precision is improved.
2. The invention inputs the clearly detected picture into the semantic segmentation model for semantic segmentation and then carries out binarization processing, thereby reducing the requirement on the shooting environment and improving the accuracy of line segment state detection in a complex environment.
3. According to the method, the expansion pretreatment, the corrosion pretreatment and the refinement pretreatment are carried out on the picture, then noise points and interference points are filtered through a clustering algorithm, and error identification pixel points caused by a shooting machine and a shooting environment are also filtered together, so that the robustness of line segment state detection is further improved.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the embodiments of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and those skilled in the art can also obtain other related drawings based on the drawings without inventive efforts.
FIG. 1 is a schematic flow chart illustrating a method for scoring an image of a line segment in an experiment according to an embodiment of the present invention;
FIG. 2 is a schematic structural diagram of a system for detecting line segment images in an experiment according to an embodiment of the present disclosure;
fig. 3 is a schematic structural diagram of a device for detecting a line segment image in an experiment according to an embodiment of the present invention.
The labels in the figure are: 701. a first acquisition unit; 702. a first processing unit; 703. a first clustering unit; 704. a second processing unit; 705. a third processing unit; 7021. a first processing subunit; 7022. a second processing subunit; 7023. a third processing subunit; 7031. a first filtering subunit; 7032. a fourth processing subunit; 7033. a first clustering subunit; 7034. a fifth processing subunit; 7041. a sixth processing subunit; 7042. a seventh processing subunit; 7051. an eighth processing subunit; 7052. a first marker subunit; 7053. a first judgment subunit; 7054. a first comparison subunit; 7055. a second marker subunit; 70511. a third judgment subunit; 70512. a third judgment subunit; 70513. a fourth judgment subunit 70513; 800. an experimental scoring device; 801. a processor; 802. a memory; 803. a multimedia component; 804. an input/output (I/O) interface; 805. a communication component.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present invention, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.
Example 1
As shown in fig. 1, the present embodiment provides a method for scoring a line segment image in an experiment, where the method includes step S1, step S2, step S3, step S4, and step S5.
Step S1, acquiring first information, wherein the first information is all image information acquired by a camera in the experimental process;
it can be understood that the experimental data is acquired by the camera during the experiment, the experimental data is image data made by an experimenter according to the experiment, and the experimental data is sent to the preprocessing module of the invention for preprocessing.
Step S2, sending the first information to a preprocessing model for processing to obtain second information, wherein the preprocessing model is a model for preprocessing the first information and deleting a preprocessed blurred image, and the second information is all clear images in the first information;
it can be understood that the invention acquires the image information only containing line segments by performing gray processing and edge detection on the image in the first information, and then extracts the sharp image only containing line segments.
Step S3, sending the second information to a clustering model for processing to obtain third information, wherein the clustering model is a model for performing clustering analysis and binarization processing on pixel points in the second information, and the third information is binarized pixel point information;
it can be understood that the invention filters the pixels in the second information, then divides the pixels into background pixels and segment pixels based on semantic segmentation, then clusters the segment pixels through a clustering model, and then extracts the most cluster clusters, and since the segments required to be made in the experiment are connected together, other individual segments wrongly drawn are filtered through a clustering algorithm.
Step S4, sending the third information to a thinning model for processing to obtain fourth information, wherein the thinning model is a model for performing at least one expansion processing and at least one corrosion processing on the third information, and the fourth information is segment pixel point information with the width of only one pixel point;
it can be understood that the invention carries out multiple times of expansion processing and corrosion processing to thin the line segment pixel points, and obtains the line segment pixel point information with only one pixel point in width, thereby obtaining a clear pixel point image.
And step S5, scoring the fourth information according to a preset scoring rule to obtain fifth information, wherein the fifth information is score information of each experimental image.
It can be understood that the invention reduces the area of the line segment to be detected by processing the image in the whole picture, avoids the interference of the environment on the line segment state detection and improves the detection precision.
The method can be understood that the method converts the detected clear picture into the gray-scale image, then inputs the semantic segmentation model for semantic segmentation, and compares the threshold value binarization, so that the method reduces the requirements on the shooting environment and improves the accuracy of line segment state detection in the complex environment.
The method can be understood that the robustness of the line segment state detection is further improved by performing expansion preprocessing, corrosion preprocessing and refinement preprocessing on the picture, then filtering out noise points and interference points through a clustering algorithm, and also filtering out mistakenly identified pixel points caused by a shooting machine and a shooting environment.
In a specific embodiment of the present disclosure, the step S2 includes a step S21, a step S22, and a step S23.
Step S21, performing gray scale processing on each frame image in the first information, to obtain a gray scale processed image, where the gray scale processed image includes an image in which an average value of three RGB component luminances of each frame image is used as a gray scale value;
it can be understood that the present invention obtains a gray scale map by performing gray scale processing on each frame image in the first information, and then using the average value of the three RGB component luminances of each frame image as an image of gray scale values.
Step S22, performing edge detection on the grayed images by adopting an edge detection algorithm to obtain edge information of each grayed image, and performing differential processing on the grayed images based on each edge information to obtain image information after edge detection;
it can be understood that in the above steps, the range of line segments in all the gray-scale images is detected through an edge detection algorithm, and then the pixel points of the line segments are extracted to obtain the image information of each line segment.
And step S23, sending the first information and the image information after the edge detection to a trained convolutional neural network classification model for classification, and deleting all fuzzy images in the first information to obtain all clear images in the first information.
The first information is subjected to graying processing and then processed by an edge detection algorithm, so that the dimensionality of classification training is increased, whether the image is clear or not and whether the brightness and the like meet requirements or not are determined more accurately, the judgment error of a training model is reduced, and the accuracy of the training model is improved.
The method can be understood that the historical experiment fuzzy image information, the historical experiment clear image information, the image information after the historical experiment fuzzy image edge detection and the image information after the historical experiment clear image edge detection are overlapped, wherein the overlapping mode is that the image information after the edge detection and the image information before the edge detection are overlapped in a one-to-one correspondence mode, and then the overlapped image information is input into a network model to be trained to obtain the convolutional neural network classification model.
In a specific embodiment of the present disclosure, the step S3 includes steps S31, S32, S33 and S34.
Step S31, filtering the second information by adopting a target detection model to obtain first sub information, wherein the first sub information is image information after filtering is obtained after pixel points which are not connected together in the second information are deleted;
step S32, inputting the first sub-information into a semantic segmentation model for semantic segmentation to obtain second sub-information, wherein the second sub-information comprises background pixel points and line segment pixel points of the first sub-information;
the semantic segmentation model is obtained by opening a picture to be segmented by a data labeling tool, manually labeling an area to be semantically segmented, generating a corresponding label file by data labeling tool labeling software, wherein the label file contains a polygonal coordinate frame of an object to be manually labeled, obtaining polygonal drawing of the object according to coordinates in the label file, and inputting the picture to be segmented and the polygonal drawing of the object into a training model for training.
Step S33, performing clustering analysis on the line segment pixel points in the second sub-information by using a DBSAN (database server application server) clustering model to obtain at least one pixel point cluster, marking and calling the pixel point cluster with the largest number of the line segment pixel points to obtain line segment clustering pixel points, and deleting other pixel point clusters to obtain filtered line segment pixel points;
it can be understood that the invention determines the area where the connection line is located by filtering the second information, obtains the connection line coordinate and cuts the connection line, then divides the cut image semantically, divides the pixel points in the image into two types, one is the background pixel point and the other is the segment pixel point, and binarizes all the pixel points in the image, and preferably uses the DBSAN to perform noise reduction and deletion on the binarized pixel points, so as to obtain the filtered segment pixel point information.
Step S34, performing binarization processing on the background pixel points and the filtered segment pixel points to obtain third sub-information, where the third sub-information is binarized pixel point information, and in the third sub-information, the background pixel points are set to 0, and the filtered segment pixel points are set to 255.
It can be understood that the invention can also open the picture needing target detection by using the data labeling tool, manually label the object needing model detection, and generate the corresponding label file by using the data labeling tool labeling software, wherein the label file contains the rectangular coordinate frame of the manually labeled object. And sending the picture file and the label file which correspond to each other into the target detection model, and training to obtain the trained target detection model.
It can be understood that the noise points which need to be processed after the pixel points are processed through the steps are fewer, and the difficulty is lower than that of unprocessed noise points.
It can be understood that the detection area is reduced by firstly using the target detection model to obtain the connecting line graph, so that the class with the largest clustering quantity is the line segment, while the straight ruler pencil appearing at the corner of the picture occupies fewer pixels, and the number of the pixel points is less than that of the pixel points of the line segment.
In a specific embodiment of the present disclosure, the step S4 includes steps S41 and S42.
Step S41, performing pixel interpolation processing on the edge of each binarized pixel point in the third information by adopting a preset first convolution kernel to obtain fourth sub information, and performing corrosion processing on the fourth sub information to obtain preprocessed third information;
and step S42, carrying out iterative expansion processing and secondary corrosion processing on the preprocessed third information by adopting an iterative refinement algorithm, and refining the processed third information to obtain segment pixel point information with the width of only one pixel point.
It can be understood that the invention obtains the segment pixel point information of only one pixel point by performing expansion processing, corrosion processing and thinning processing on the image information, and the segment pixel point information is a segment, and can also calculate the endpoint information of the segment and the intersection point information of the segment.
In a specific embodiment of the present disclosure, the step S5 includes steps S51, S52, S53, S54, and S55.
Step S51, obtaining the end point information of the line segment and the intersection point information of the line segment according to the line segment pixel point information of which the width is only one pixel point;
step S52, marking the line segment passing through the three intersection points as a first line segment, and marking the line segment passing through the two end points and one intersection point as a second line segment; the second line segments comprise at least two, and at least two of the second line segments are parallel to each other;
step S53, judging whether the included angle between the second line segment and the coordinate axis Y is smaller than a preset first threshold value, and if so, calculating the distance between two adjacent second line segments;
step S54, comparing the distance between any two second line segments, determining whether the difference between the distance between the two second line segments is smaller than a preset second threshold, and if so, determining that the image is a correct image;
the method can be understood that the intersection point and the endpoint information of each line segment pixel point are extracted, and whether the line segments formed by the line segment pixel points are parallel or vertical is judged through the intersection point and the endpoint information, so that the straight line distance between every two parallel line segments is determined, and whether the graph drawn by experimenters is correct is judged.
And step S55, marking the correct image into one score to obtain the scoring information of the first information.
It can be understood that the invention judges the line segment state through the end points and the intersection points, and scores the line segment state according to a specific rule, so as to obtain the scoring condition of the physical experiment.
In a specific embodiment of the present disclosure, the step S51 includes a step S511, a step S512, and a step S513.
Step S511, traversing the binarized pixel point information around the pixel points on all the line segments, and judging whether the binarized pixel point information around the pixel points on the line segments is 255 or not;
step S512, counting the number of the binarized pixel point information around the pixel points on the line segment, which is 255, to obtain the total number of the binarized pixel point information around the pixel points on the line segment, which is 255;
it can be understood that the invention judges whether the binarized pixel point information around the pixel points on the line segment has line segment pixel points or not, if so, judges that several pixel points exist around the line segment, and judges whether the pixel points are end points or cross points or not based on the surrounding several pixel points.
Step S513, determining whether the sum of the numbers is 4 or 1, if the total value of the pixels around the line segment pixel is 4, determining that the pixel is a cross point, and if the total value of the pixels around the line segment pixel is 1, determining that the pixel is an end point.
It can be understood that the present invention judges whether there is the same pixel point by performing 8 angles around each pixel point, and determines whether the pixel point is the end point and the cross point of the line segment according to a certain rule.
Example 2
As shown in fig. 2, the present embodiment provides a system for detecting a line segment image in an experiment, where the system includes a first obtaining unit 701, a first processing unit 702, a first clustering unit 703, a second processing unit 704, and a third processing unit 705.
A first obtaining unit 701, configured to obtain first information, where the first information is all image information acquired by a camera in an experimental process;
a first processing unit 702, configured to send the first information to a preprocessing model for processing to obtain second information, where the preprocessing model is a model for preprocessing the first information and deleting a preprocessed blurred image, and the second information is all sharp images in the first information;
the first clustering unit 703 is configured to send the second information to a clustering model for processing to obtain third information, where the clustering model is a model for performing clustering analysis and binarization on pixels in the second information, and the third information is binarized pixel information;
the second processing unit 704 is configured to send the third information to a refinement model for processing to obtain fourth information, where the refinement model is a model for performing at least one expansion processing and at least one corrosion processing on the third information, and the fourth information is segment pixel point information with a width of only one pixel point;
the third processing unit 705 is configured to score the fourth information according to a preset scoring rule, so as to obtain fifth information, where the fifth information is score information of each experimental image.
In a specific embodiment of the present disclosure, the first processing unit 702 includes a first processing subunit 7021, a second processing subunit 7022, and a third processing subunit 7023.
A first processing subunit 7021, configured to perform gray processing on each frame of image in the first information, respectively, to obtain a grayed image, where the grayed image includes an image in which an average value of three RGB component luminances of each frame of image is used as a gray value;
a second processing subunit 7022, configured to perform edge detection on the grayed image by using an edge detection algorithm to obtain edge information of each grayed image, and perform difference processing on the grayed image based on each edge information to obtain image information after edge detection;
a third processing subunit 7023, configured to send the first information and the image information after the edge detection to a trained convolutional neural network classification model for classification, and delete all the blurred images in the first information to obtain all the sharp images in the first information.
In a specific embodiment of the present disclosure, the first clustering unit 703 includes a first filtering subunit 7031, a fourth processing subunit 7032, a first clustering subunit 7033, and a fifth processing subunit 7034.
A first filtering subunit 7031, configured to filter the second information by using a target detection model to obtain first sub information, where the first sub information is image information obtained after deleting pixel points that are not connected together in the second information;
a fourth processing subunit 7032, configured to input the first sub information to a semantic segmentation model for semantic segmentation to obtain second sub information, where the second sub information includes background pixel points and line segment pixel points of the first sub information;
a first clustering subunit 7033, configured to perform clustering analysis on the segment pixels in the second sub-information by using a DBSAN clustering model to obtain at least one pixel cluster, mark and call the pixel cluster with the largest number of segment pixels to obtain a segment clustering pixel, and delete other pixel clusters to obtain filtered segment pixels;
a fifth processing subunit 7034, configured to perform binarization processing on the background pixel points and the filtered line segment pixel points to obtain third sub information, where the third sub information is binarized pixel point information, in the third sub information, the background pixel points are set to 0, and the filtered line segment pixel points are set to 255.
In a specific embodiment of the present disclosure, the second processing unit 704 includes a sixth processing subunit 7041 and a seventh processing subunit 7042.
A sixth processing subunit 7041, configured to perform pixel interpolation processing on the edge of each binarized pixel point in the third information by using a preset first convolution kernel to obtain fourth sub information, and perform corrosion processing on the fourth sub information to obtain preprocessed third information;
a seventh processing subunit 7042, configured to perform iterative expansion processing and secondary corrosion processing on the preprocessed third information by using an iterative refinement algorithm, and refine the processed third information to obtain segment pixel point information with a width of only one pixel point.
In a specific embodiment of the present disclosure, the third processing unit 705 includes an eighth processing subunit 7051, a first marking subunit 7052, a first determining subunit 7053, a first comparing subunit 7054, and a second marking subunit 7055.
An eighth processing subunit 7051, configured to obtain, according to the segment pixel information of which the width is only one pixel, end point information of the segment and intersection point information of the segment;
a first marking subunit 7052, configured to mark a line segment passing through three of the intersection points as a first line segment, and mark a line segment passing through two of the end points and one of the intersection points as a second line segment; the second line segments comprise at least two, and at least two of the second line segments are parallel to each other;
a first determining subunit 7053, configured to determine whether an included angle between the second line segment and the coordinate axis Y is smaller than a preset first threshold, and if the included angle is smaller than the first threshold, calculate a distance between two adjacent second line segments;
a first comparing subunit 7054, configured to compare the distance between any two of the second line segments, determine whether a difference between the distance between the two second line segments is smaller than a preset second threshold, and if the difference is smaller than the second threshold, determine that the image is a correct image;
and a second marking subunit 7055, configured to mark the correct image into one score to obtain the scoring information of the first information.
In one embodiment of the present disclosure, the eighth processing sub-unit 7051 includes a third judging sub-unit 70511, a ninth processing sub-unit 70512, and a fourth judging sub-unit 70513.
A third determining subunit 70511, configured to traverse the binarized pixel information around the pixel points on all the line segments, and determine whether the binarized pixel information around the pixel points on the line segments is 255;
a ninth processing subunit 70512, configured to count the number of the binarized pixel information around the pixel point on the line segment, which is 255, to obtain a total number of the binarized pixel information around the pixel point on the line segment, which is 255;
a fourth determining subunit 70513, configured to determine whether the sum of the numbers is 4 or 1, determine that the pixel is a cross point if the total value of the pixels around the line segment pixel is 4, and determine that the pixel is an end point if the total value of the pixels around the line segment pixel is 1.
It should be noted that, regarding the system in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated herein.
Example 3
Corresponding to the above method embodiment, the embodiment of the present disclosure further provides a device for detecting an experimental line segment image, and the device for detecting an experimental line segment image described below and the method for scoring an experimental line segment image described above may be referred to correspondingly.
Fig. 3 is a block diagram illustrating an apparatus 800 for detecting an image of a line segment in an experiment according to an exemplary embodiment. As shown in fig. 3, the apparatus 800 for detecting line segment images in the experiment may include: a processor 801, a memory 802. The apparatus 800 for detecting line segment images in an experiment may further include one or more of a multimedia component 803, an input/output (I/O) interface 804, and a communication component 805.
The processor 801 is configured to control the overall operation of the device 800 for detecting line segment images in an experiment, so as to complete all or part of the steps in the above method for scoring line segment images in an experiment. The memory 802 is used to store various types of data to support the operation of the inspection device 800 of line segment images in the experiment, which data may include, for example, instructions for any application or method operating on the inspection device 800 of line segment images in the experiment, as well as application-related data, such as contact data, messages sent or received, pictures, audio, video, and the like. The Memory 802 may be implemented by any type of volatile or non-volatile Memory device or combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic Memory, flash Memory, magnetic disk or optical disk. The multimedia components 803 may include screen and audio components. Wherein the screen may be, for example, a touch screen and the audio component is used for outputting and/or inputting audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may further be stored in the memory 802 or transmitted through the communication component 805. The audio assembly further comprises at least one speaker for outputting audio signals. The I/O interface 804 provides an interface between the processor 801 and other interface modules, such as a keyboard, mouse, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 805 is used for wired or wireless communication between the detection device 800 of the line segment image and other devices in the experiment. Wireless communication, such as Wi-Fi, bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or a combination of one or more of them, so that the corresponding communication component 805 may include: Wi-Fi module, bluetooth module, NFC module.
In an exemplary embodiment, the Device 800 for detecting line segment images in an experiment can be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components, and is used to perform one of the above-mentioned methods for scoring line segment images in an experiment.
In another exemplary embodiment, a computer readable storage medium including program instructions, which when executed by a processor, implement the steps of the above-described method for scoring a line segment image in an experiment is also provided. For example, the computer readable storage medium may be the above-mentioned memory 802 including program instructions executable by the processor 801 of the device 800 for detecting an image of a line segment in an experiment to perform the above-mentioned method for scoring an image of a line segment in an experiment.
Example 4
Corresponding to the above method embodiment, the embodiment of the present disclosure further provides a readable storage medium, and a readable storage medium described below and the above scoring method for the line segment image in the experiment may be referred to correspondingly.
A readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the steps of the method for scoring a line segment image in an experiment according to the above-described method embodiment.
The readable storage medium may be a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and various other readable storage media capable of storing program codes.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. A grading method for line segment images in experiments is characterized by comprising the following steps:
acquiring first information, wherein the first information is image information acquired by a camera in an experimental process;
sending the first information to a preprocessing model for processing to obtain second information, wherein the preprocessing model is a model for preprocessing the first information and deleting a preprocessed blurred image, and the second information is all clear images in the first information;
sending the second information to a clustering model for processing to obtain third information, wherein the clustering model is a model for performing clustering analysis and binarization processing on pixel points in the second information, and the third information is binarized pixel point information;
sending the third information to a refinement model for processing to obtain fourth information, wherein the refinement model is a model for performing at least one expansion processing and at least one corrosion processing on the third information, and the fourth information is line segment pixel point information with the width of only one pixel point;
and grading the fourth information according to a preset grading rule to obtain fifth information, wherein the fifth information is the score information of each experimental image.
2. The method for scoring the line segment image in the experiment as claimed in claim 1, wherein the step of sending the first information to a preprocessing model for processing to obtain second information comprises:
performing gray processing on each frame of image in the first information respectively to obtain a gray processed image, wherein the gray processed image comprises an image taking the average value of the brightness of three RGB components of each frame of image as a gray value;
performing edge detection on the grayed images by adopting an edge detection algorithm to obtain edge information of each grayed image, and performing differential processing on the grayed images based on each edge information to obtain image information after edge detection;
and sending the first information and the image information after the edge detection to a trained convolutional neural network classification model for classification, and deleting all fuzzy images in the first information to obtain all clear images in the first information.
3. The method for scoring the line segment image in the experiment according to claim 1, wherein the step of sending the second information to a clustering model for clustering processing to obtain third information comprises:
filtering the second information by adopting a target detection model to obtain first sub information, wherein the first sub information is image information after filtering is obtained after pixel points which are not connected together in the second information are deleted;
inputting the first sub information into a semantic segmentation model for semantic segmentation to obtain second sub information, wherein the second sub information comprises background pixel points and line segment pixel points of the first sub information;
performing clustering analysis on the line segment pixel points in the second sub-information by using a DBSAN (database server application server) clustering model to obtain at least one pixel point cluster, marking and calling the pixel point cluster with the largest number of the line segment pixel points to obtain line segment clustering pixel points, and deleting other pixel point clusters to obtain filtered line segment pixel points;
and carrying out binarization processing on the background pixel points and the filtered line segment pixel points to obtain third sub information, wherein the third sub information is binarized pixel point information, in addition, in the third sub information, the background pixel points are set to be 0, and the filtered line segment pixel points are set to be 255.
4. The method for scoring the line segment image in the experiment as claimed in claim 3, wherein the step of sending the third information to a refinement model for refinement processing to obtain fourth information comprises the steps of:
performing pixel interpolation processing on the edge of each binarized pixel point in the third information by adopting a preset first convolution kernel to obtain fourth sub information, and performing corrosion processing on the fourth sub information to obtain preprocessed third information;
and carrying out iterative expansion processing and secondary corrosion processing on the preprocessed third information by adopting an iterative refinement algorithm, and refining the processed third information to obtain segment pixel point information with the width of only one pixel point.
5. A detection system for line segment images in experiments is characterized by comprising:
the first acquisition unit is used for acquiring first information, wherein the first information is all image information acquired by a camera in an experimental process;
the first processing unit is used for sending the first information to a preprocessing model for processing to obtain second information, the preprocessing model is a model for preprocessing the first information and deleting a preprocessed blurred image, and the second information is all clear images in the first information;
the first clustering unit is used for sending the second information to a clustering model for processing to obtain third information, the clustering model is a model for carrying out clustering analysis and binarization processing on pixel points in the second information, and the third information is binarized pixel point information;
the second processing unit is used for sending the third information to a refinement model for processing to obtain fourth information, the refinement model is a model for performing at least one expansion processing and at least one corrosion processing on the third information, and the fourth information is line segment pixel point information with the width of only one pixel point;
and the third processing unit is used for grading the fourth information according to a preset grading rule to obtain fifth information, wherein the fifth information is score information of each experimental image.
6. The system for detecting an image of a line segment in an experiment as claimed in claim 5, wherein the system comprises:
the first processing subunit is configured to perform gray processing on each frame of image in the first information, so as to obtain a grayed image, where the grayed image includes an image in which an average value of three RGB component luminances of each frame of image is used as a gray value;
the second processing subunit is configured to perform edge detection on the grayed image by using an edge detection algorithm to obtain edge information of each grayed image, and perform differential processing on the grayed image based on each edge information to obtain image information after edge detection;
and the third processing subunit is configured to send the first information and the image information after the edge detection to a trained convolutional neural network classification model for classification, and delete all blurred images in the first information to obtain all sharp images in the first information.
7. The system for detecting an image of a line segment in an experiment as claimed in claim 5, wherein the system comprises:
the first filtering subunit is configured to filter the second information by using a target detection model to obtain first sub information, where the first sub information is image information obtained after deleting pixel points that are not connected together in the second information;
the fourth processing subunit is configured to input the first sub-information to a semantic segmentation model for semantic segmentation to obtain second sub-information, where the second sub-information includes background pixel points and line segment pixel points of the first sub-information;
the first clustering subunit is used for performing clustering analysis on the line segment pixel points in the second sub-information by using a DBSAN (database server application server) clustering model to obtain at least one pixel point cluster, marking and calling the pixel point cluster with the largest number of the line segment pixel points to obtain line segment clustering pixel points, and deleting other pixel point clusters to obtain filtered line segment pixel points;
and the fifth processing subunit is configured to perform binarization processing on the background pixel points and the filtered segment pixel points to obtain third sub information, where the third sub information is binarized pixel point information, in the third sub information, the background pixel points are set to 0, and the filtered segment pixel points are set to 255.
8. The system for detecting the image of the line segment in the experiment as claimed in claim 5, wherein the system comprises:
a sixth processing subunit, configured to perform pixel interpolation processing on an edge of each binarized pixel point in the third information by using a preset first convolution kernel to obtain fourth sub information, and perform corrosion processing on the fourth sub information to obtain preprocessed third information;
and the seventh processing subunit is used for performing iterative expansion processing and secondary corrosion processing on the preprocessed third information by adopting an iterative refinement algorithm, and refining the processed third information to obtain segment pixel point information with the width of only one pixel point.
9. The utility model provides a check out test set of line segment image in experiment which characterized in that includes:
a memory for storing a computer program;
a processor for implementing the steps of the method for scoring line segment images in an experiment as claimed in any one of claims 1 to 4 when executing said computer program.
10. A readable storage medium, characterized by: the readable storage medium has stored thereon a computer program which, when executed by a processor, implements the steps of the method for scoring line segment images in an experiment as claimed in any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210186378.XA CN114549993B (en) | 2022-02-28 | 2022-02-28 | Method, system and device for grading line segment image in experiment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210186378.XA CN114549993B (en) | 2022-02-28 | 2022-02-28 | Method, system and device for grading line segment image in experiment and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114549993A true CN114549993A (en) | 2022-05-27 |
| CN114549993B CN114549993B (en) | 2022-11-11 |
Family
ID=81679243
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210186378.XA Active CN114549993B (en) | 2022-02-28 | 2022-02-28 | Method, system and device for grading line segment image in experiment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114549993B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114882520A (en) * | 2022-07-08 | 2022-08-09 | 成都西交智汇大数据科技有限公司 | Method, system and equipment for detecting circuit diagram and readable storage medium |
| CN114881541A (en) * | 2022-07-06 | 2022-08-09 | 成都西交智汇大数据科技有限公司 | Evaluation method, device and equipment for examinee score and readable storage medium |
| CN115019240A (en) * | 2022-08-04 | 2022-09-06 | 成都西交智汇大数据科技有限公司 | Grading method, device and equipment for chemical experiment operation and readable storage medium |
| CN115359412A (en) * | 2022-10-24 | 2022-11-18 | 成都西交智汇大数据科技有限公司 | Hydrochloric acid neutralization experiment scoring method, device, equipment and readable storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102629325A (en) * | 2012-03-13 | 2012-08-08 | 深圳大学 | Image characteristic extraction method, device thereof, image copy detection method and system thereof |
| CN110189344A (en) * | 2019-04-24 | 2019-08-30 | 河海大学常州校区 | A kind of line segment extraction method in power knife switch image |
| CN110781885A (en) * | 2019-10-24 | 2020-02-11 | 泰康保险集团股份有限公司 | Text detection method, device, medium and electronic equipment based on image processing |
| CN111353957A (en) * | 2020-02-28 | 2020-06-30 | 北京三快在线科技有限公司 | Image processing method, image processing device, storage medium and electronic equipment |
| CN111881310A (en) * | 2019-12-07 | 2020-11-03 | 杭州华冬人工智能有限公司 | Chinese character hard-stroke writing intelligent guidance and scoring method and guidance scoring system |
| CN112614074A (en) * | 2020-12-28 | 2021-04-06 | 自行科技(武汉)有限公司 | Robust vanishing point detection method and device based on response graph and clustering |
| CN113988563A (en) * | 2021-10-21 | 2022-01-28 | 上海大风实验室设备有限公司 | A intelligent endow branch system for level crossing imaging experiment |
-
2022
- 2022-02-28 CN CN202210186378.XA patent/CN114549993B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102629325A (en) * | 2012-03-13 | 2012-08-08 | 深圳大学 | Image characteristic extraction method, device thereof, image copy detection method and system thereof |
| CN110189344A (en) * | 2019-04-24 | 2019-08-30 | 河海大学常州校区 | A kind of line segment extraction method in power knife switch image |
| CN110781885A (en) * | 2019-10-24 | 2020-02-11 | 泰康保险集团股份有限公司 | Text detection method, device, medium and electronic equipment based on image processing |
| CN111881310A (en) * | 2019-12-07 | 2020-11-03 | 杭州华冬人工智能有限公司 | Chinese character hard-stroke writing intelligent guidance and scoring method and guidance scoring system |
| CN111353957A (en) * | 2020-02-28 | 2020-06-30 | 北京三快在线科技有限公司 | Image processing method, image processing device, storage medium and electronic equipment |
| CN112614074A (en) * | 2020-12-28 | 2021-04-06 | 自行科技(武汉)有限公司 | Robust vanishing point detection method and device based on response graph and clustering |
| CN113988563A (en) * | 2021-10-21 | 2022-01-28 | 上海大风实验室设备有限公司 | A intelligent endow branch system for level crossing imaging experiment |
Non-Patent Citations (1)
| Title |
|---|
| 吴一全等: ""基于视觉的车道线检测方法研究进展"", 《仪器仪表学报》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114881541A (en) * | 2022-07-06 | 2022-08-09 | 成都西交智汇大数据科技有限公司 | Evaluation method, device and equipment for examinee score and readable storage medium |
| CN114882520A (en) * | 2022-07-08 | 2022-08-09 | 成都西交智汇大数据科技有限公司 | Method, system and equipment for detecting circuit diagram and readable storage medium |
| CN115019240A (en) * | 2022-08-04 | 2022-09-06 | 成都西交智汇大数据科技有限公司 | Grading method, device and equipment for chemical experiment operation and readable storage medium |
| CN115359412A (en) * | 2022-10-24 | 2022-11-18 | 成都西交智汇大数据科技有限公司 | Hydrochloric acid neutralization experiment scoring method, device, equipment and readable storage medium |
| CN115359412B (en) * | 2022-10-24 | 2023-03-03 | 成都西交智汇大数据科技有限公司 | Hydrochloric acid neutralization experiment scoring method, device, equipment and readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114549993B (en) | 2022-11-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114549993B (en) | Method, system and device for grading line segment image in experiment and readable storage medium | |
| CN109726643B (en) | Method and device for identifying table information in image, electronic equipment and storage medium | |
| CN110942074B (en) | Character segmentation recognition method and device, electronic equipment and storage medium | |
| TW201405440A (en) | Form recognition method and device | |
| CN111738041A (en) | Video segmentation method, device, equipment and medium | |
| CN105427696A (en) | Method for distinguishing answer to target question | |
| CN111242899B (en) | Image-based flaw detection method and computer-readable storage medium | |
| CN111626249B (en) | Method and device for identifying geometric figure in topic image and computer storage medium | |
| CN110135225B (en) | Sample labeling method and computer storage medium | |
| CN112149667A (en) | Method for automatically reading pointer type instrument based on deep learning | |
| CA3035387C (en) | Digitization of industrial inspection sheets by inferring visual relations | |
| CN111652117A (en) | Method and medium for segmenting multi-document image | |
| WO2022237117A1 (en) | Touch control method and system for interactive electronic whiteboard, and readable medium | |
| CN112613367A (en) | Bill information text box acquisition method, system, equipment and storage medium | |
| CN112052730A (en) | 3D dynamic portrait recognition monitoring device and method | |
| CN111832551A (en) | Text image processing method and device, electronic scanning equipment and storage medium | |
| CN111008635A (en) | OCR-based multi-bill automatic identification method and system | |
| CN115393616A (en) | Target tracking method, device, equipment and storage medium | |
| CN114240928A (en) | Board quality partition detection method, device and equipment and readable storage medium | |
| CN114049686A (en) | Signature recognition model training method and device and electronic equipment | |
| CN112949621A (en) | Method and device for marking test paper answering area, storage medium and electronic equipment | |
| CN112232282A (en) | Gesture recognition method and device, storage medium and electronic equipment | |
| CN111062377A (en) | Question number detection method, system, storage medium and electronic equipment | |
| CN114882520B (en) | Method, system and equipment for detecting circuit diagram and readable storage medium | |
| CN114998906B (en) | Text detection method, training method and device of model, electronic equipment and 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 |