CN115409979A - Epidemic prevention kit marked line region detection method, device, medium and equipment - Google Patents
Epidemic prevention kit marked line region detection method, device, medium and equipment Download PDFInfo
- Publication number
- CN115409979A CN115409979A CN202211015182.0A CN202211015182A CN115409979A CN 115409979 A CN115409979 A CN 115409979A CN 202211015182 A CN202211015182 A CN 202211015182A CN 115409979 A CN115409979 A CN 115409979A
- Authority
- CN
- China
- Prior art keywords
- point
- picture
- marking
- kit
- network
- 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 42
- 230000002265 prevention Effects 0.000 title claims abstract description 25
- 238000012549 training Methods 0.000 claims abstract description 70
- 238000000034 method Methods 0.000 claims abstract description 36
- 238000012545 processing Methods 0.000 claims abstract description 8
- 238000002372 labelling Methods 0.000 claims description 30
- 230000004913 activation Effects 0.000 claims description 24
- 230000006870 function Effects 0.000 claims description 23
- 230000015654 memory Effects 0.000 claims description 12
- 238000010606 normalization Methods 0.000 claims description 6
- 238000004873 anchoring Methods 0.000 claims description 3
- 238000013461 design Methods 0.000 abstract description 6
- 239000003153 chemical reaction reagent Substances 0.000 description 11
- 238000010586 diagram Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 239000000427 antigen Substances 0.000 description 2
- 102000036639 antigens Human genes 0.000 description 2
- 108091007433 antigens Proteins 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 241000711573 Coronaviridae Species 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
- G06V10/225—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition based on a marking or identifier characterising the area
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/24—Aligning, centring, orientation detection or correction of the image
- G06V10/242—Aligning, centring, orientation detection or correction of the image by image rotation, e.g. by 90 degrees
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/24—Aligning, centring, orientation detection or correction of the image
- G06V10/245—Aligning, centring, orientation detection or correction of the image by locating a pattern; Special marks for positioning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/32—Normalisation of the pattern dimensions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of image processing, in particular to a method, a device, a medium and equipment for detecting a marked line region of an epidemic prevention kit, wherein compared with the prior art, the method comprises the steps of setting a rectangular marking frame to enable the marked line region to be positioned in the rectangular marking frame, and acquiring absolute coordinate data of a first point and a second point on the rectangular marking frame, wherein the first point and the second point are in a mutual diagonal position relation on the rectangular marking frame, and the first point and the second point have a specific relation relative to a first reference system of a kit picture; and then, the absolute coordinate data of the first point and the second point are sequentially stored as the marking information, so that the rotation angle information of the marking line region is hidden in the marking information, and the design of a subsequent detection training model can be further simplified while the marking information is simplified.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a method, a device, a medium and equipment for detecting a marked line region of an epidemic prevention kit.
Background
The new coronavirus antigen detection kit is a simple and rapid self-service detection method, and non-medical professionals can interpret the detection result through marked lines on the kit. Generally, all levels of health departments require uploading of photo pictures of detection results, verification and statistics of the detection results, but manual interpretation of massive detection results is repetitive work with huge workload. Therefore, interpretation of antigen detection kit results using a deep learning model-based approach is an alternative approach to alleviating the pressure of manual interpretation. Before the position detection of the kit is carried out by using the deep learning model, the model needs to be trained. Therefore, how to construct the training data set has a non-negligible effect on the detection capability and effect of the model. In general, in order to improve the generalization capability of a data set, pictures in various possible scenes in a real environment need to be collected as a training data set.
In addition, the data set picture needs to be labeled, so that the trained model can detect the position of the kit line area. In the prior art, for example, a chinese patent application with publication number CN109902680a (publication date is 2019, 6 months and 18 days) discloses a method for detecting and correcting an image rotation angle based on a convolutional neural network, which includes the following steps: A. constructing a trained dataset of the network; b, constructing a network structure and setting training parameters; C. model training and model parameter storage after training; D. detecting and correcting an angle; E. rotation angle calculation and picture rotation correction. The method combines image processing and a convolutional neural network, improves the efficiency and the accuracy of angle detection by a deep learning technical means, performs angle detection and rotation correction on the picture by utilizing the method, can perfectly solve the problem of poor text detection effect of the large-angle picture, and effectively improves the recognition rate of the text OCR. In the prior art, although the method is widely applied to information such as a rotation angle of a certain object, an efficient method for representing the information of the rotation angle is not provided.
Data marking requires marking the position of the reagent kit marking area. Since the reagent cartridge is rotated when the data set image is generated, the rotation angle information also needs to be labeled when labeling. The traditional method is to label the angle information separately from the position coordinate information.
In summary, in the prior art, the direct labeling of the angle information to the region would lead to the complication of the subsequent training design of the detection model, and the present invention aims to provide a new method for labeling the detection region of the training data set and the design of the training model to solve the above problems.
Disclosure of Invention
In order to solve the defects in the prior art, the invention provides a method for detecting the marked line region of an epidemic prevention kit, which comprises the following steps:
s10: acquiring a training data picture, wherein the training data picture comprises a kit picture for displaying a marking area;
s20: setting a rectangular marking frame to enable the marking line area to be located in the rectangular marking frame;
s30: acquiring absolute coordinate data of a first point and a second point on the rectangular marking frame, wherein the first point and the second point are in a mutual diagonal position relation on the rectangular marking frame, and the first point and the second point have a specific relation relative to a first reference system of the kit picture;
s40: respectively carrying out normalization processing on the absolute coordinate data of the first point and the absolute coordinate data of the second point, and then sequentially storing the normalized absolute coordinate data as the labeling information of the training data pictures;
s50: inputting the training data picture and the corresponding labeling information into a detection training model, wherein the detection training model comprises an activation function, and the activation function comprises a sigmoid activation function so as to output the normalized labeling information.
In some embodiments, the first reference frame is fixed relative to the orientation of the reagent box picture, and the longitudinal axis of the first reference frame points in the direction of the letters "T" to "C" in the reagent box picture.
In some embodiments, the specific relationship comprises: for the label box in different training data pictures, a vector formed by the first point A and the second point B in the first reference system
Is the vector for a particular quadrant.
In some embodiments, the particular quadrant includes a fourth quadrant.
In some embodiments, step S40 includes:
s41: acquiring the width and height of the training data picture;
s41: and normalizing the absolute coordinate data (Xa, ya) of the first point A and the absolute coordinate data (Xb, yb) of the second point B according to the width and the height of the training data picture.
In some embodiments, the normalization process includes calculating floating point numbers Xa2, ya2, xb2 and Yb2 by the following formulas, and storing the floating point number columns (Xa 2, ya2, xb2, yb 2) in the information of the corresponding training data pictures; wherein,
Xa2＝Xa/width；
Ya2＝Ya/height；
Xb2＝Xb/width；
Yb2＝Yb/height。
in some embodiments, the detection training model further comprises a backbone network, a neck network, and a head network;
the backbone network comprises one or more of VggNet, resNet, denseNet and MobileNet, and is used for receiving the training data picture and the labeling information;
the neck network comprises a FPN network, the neck network receiving the output of the backbone network;
the head network comprises a fully connected network, the head network receiving an output of the backbone network or an output of the neck network, the output of the head network being input to the activation function.
The invention also provides a device for detecting the marked line region of the epidemic prevention kit, which comprises:
the image acquisition module is used for acquiring a training data image, and the training data image comprises a kit image in a display marking area;
the marking frame setting module is used for setting a rectangular marking frame so that the marking line area is positioned in the rectangular marking frame;
the characteristic point anchoring module is used for acquiring absolute coordinate data of a first point and a second point on the rectangular marking frame, the first point and the second point are in a mutual diagonal position relation on the rectangular marking frame, and the first point and the second point have a specific relation relative to a first reference system of the kit picture;
the storage module is used for sequentially storing the absolute coordinate data of the first point and the second point;
and the detection training module is used for inputting the training data picture and the corresponding labeling information into a detection training model, the detection training model comprises an activation function, and the activation function comprises a sigmoid activation function so as to output the normalized labeling information.
The invention also provides a computer readable storage medium, which stores computer instructions, and when the computer is executed by a processor, the method for detecting the marked line region of the epidemic prevention kit is realized.
The invention also provides computer equipment comprising at least one processor and a memory communicatively connected with the processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to cause the processor to execute a method for epidemic prevention kit reticle field detection as described in any one of the above.
Based on the above, compared with the prior art, the method for detecting the marked line region of the epidemic prevention kit provided by the invention has the advantages that the marked line region is positioned in the rectangular marking frame by setting the rectangular marking frame, the absolute coordinate data of the first point and the second point on the rectangular marking frame are obtained, the first point and the second point are in a mutual diagonal position relation on the rectangular marking frame, and the first point and the second point have a specific relation relative to the first reference system of the kit picture; and then, the absolute coordinate data of the first point and the second point are sequentially stored as the marking information, so that the rotation angle information of the marking line region is hidden in the marking information, and the design of a subsequent detection training model can be further simplified while the marking information is simplified.
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 practice 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 embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the description in the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts; in the following description, the drawings are illustrated in a schematic view, and the drawings are not intended to limit the present invention.
FIG. 1 is a schematic flow chart of a labeling method for a data set in a marking area of a kit according to the present invention;
FIG. 2 is a schematic diagram of an embodiment of a callout box;
FIG. 3 is a schematic diagram of a first point and a second point selection of an embodiment;
FIG. 4 is a diagram of an exemplary test training model;
FIG. 5 is a schematic view of a labeling apparatus for a data set in a reticle area of a kit according to an embodiment of the present invention;
FIG. 6 is a diagram of a computer device according to an embodiment of the present invention.
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 embodiments of the present invention, but not all embodiments; the technical features designed in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other; 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.
In the description of the present invention, it is to be noted that all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs, and are not to be construed as limiting the present invention; it will be further understood that terms used herein should be interpreted as having a meaning that is consistent with their meaning in the context of this specification and the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
The following description is given by way of specific examples.
To achieve at least one of the above advantages or other advantages, the present invention provides a method for detecting a marked line region of an epidemic prevention kit, as shown in fig. 1, comprising the following steps:
s10: acquiring a training data picture, wherein the training data picture comprises a kit picture for displaying a marking area;
s20: setting a rectangular marking frame to enable the marking line area to be located in the rectangular marking frame;
s30: acquiring absolute coordinate data of a first point and a second point on the rectangular marking frame, wherein the first point and the second point are in a mutual diagonal position relation on the rectangular marking frame, and the first point and the second point have a specific relation relative to a first reference system of the kit picture; fig. 2 is a diagram of a training data picture and its labeled box in an embodiment.
S40: respectively carrying out normalization processing on the absolute coordinate data of the first point and the absolute coordinate data of the second point, and then sequentially storing the normalized absolute coordinate data as the labeling information of the training data pictures;
s50: and inputting the training data picture and the corresponding labeling information into a detection training model, wherein the detection training model comprises an activation function, and the activation function comprises a sigmoid activation function so as to output the normalized labeling information.
In some embodiments, the first reference frame is fixed relative to the orientation of the reagent box picture, and the longitudinal axis of the first reference frame points in the direction of the letters "T" to "C" in the reagent box picture.
In some embodiments, the specific relationship comprises: for the label box in different training data pictures, a vector formed by the first point A and the second point B in the first reference system
Is the vector for a particular quadrant.
As shown in fig. 3, the first reference frame is positioned to illustrate the relationship between the first point and the second point relative to the reagent box picture, and the first reference frame is fixed relative to the reagent box picture, which means that even if the reagent box picture rotates, the first reference frame rotates, so that the position selection of the first point a and the second point B changes along with the placing direction of the reagent box picture. For example, for a certain first kit picture, the placing direction of the first kit picture is that the letter "T" is above, and the letter "C" is below, at this time, assuming that the specific quadrant is the fourth quadrant, the selected first point a is the upper left vertex of the labeling frame, and the second point B is the lower right vertex of the labeling frame, and the first point a and the second point B meet the specific relationship preset in the first reference system; for a certain second kit picture, the placing direction of the second kit picture is rotated 180 degrees relative to the placing direction of the first kit picture, and the second kit picture is an inverted angle, according to the preset specific relationship, the first point a should be the lower right vertex of the labeling frame, and the second point B should be the upper left vertex of the labeling frame.
In some embodiments, the particular quadrant includes a fourth quadrant.
In some embodiments, step S40 includes:
s41: obtaining the width and height of the training data picture;
s41: and normalizing the absolute coordinate data (Xa, ya) of the first point A and the absolute coordinate data (Xb, yb) of the second point B according to the width and the height of the training data picture.
In some embodiments, the normalization process includes calculating floating point numbers Xa2, ya2, xb2 and Yb2 by the following formulas, and storing the floating point number columns (Xa 2, ya2, xb2, yb 2) in the information of the corresponding training data pictures; wherein,
Xa2＝Xa/width；
Ya2＝Ya/height；
Xb2＝Xb/width；
Yb2＝Yb/height。
in some embodiments, the rotation angle of the kit picture can be estimated according to the labeling information of the above embodiments, and the estimation process includes the following steps:
1. the obtained picture marking information is (Xa 2, ya2, xb2, yb 2)
2. If Xa2< Xb2, jump to step 3; otherwise, jumping to the step 4;
3. if Ya2< Yb2, then the rotation angle rotate _ angle =0; otherwise rotate _ angle =90, and jump to step 5;
4. if Ya2< Yb2, then the rotation angle rotate _ angle =270; otherwise rotate _ angle =180;
5. at this time, rotate _ angle is the counterclockwise rotation angle of the reagent kit.
In some embodiments, the training data picture comprises a picture taken from a kit from a different manufacturer or synthesized.
In some embodiments, as shown in fig. 4, the detection training model further comprises a backbone network, a neck network, and a head network;
the backbone network comprises one or more of VggNet, resNet, denseNet and MobileNet, and is used for receiving the training data picture and the labeling information;
the neck network comprises a FPN network, the neck network receiving the output of the backbone network;
the head network comprises a fully connected network, the head network receiving an output of the backbone network or receiving an output of the neck network, the output of the head network being input to the activation function.
In a traditional target detection network, a plurality of data such as coordinates, angles, categories and the like need to be returned, the activation modes of the data are different, and numerical regression generally uses a relu activation function as output. In the data labeling of the training set, because data normalization processing is performed, the last layer of the head network needs to be output by using a sigmoid activation function, the output data is ensured to be in the range of [0,1], only 4 data need to be output, and the design of the head network is simplified. The network model actually used by the users uses ResNet50, the neck network uses FPN, the head network uses a plurality of full connection layers, and the last layer outputs 4 data activated by sigmoid. The normalized coordinates of the two points A, B are obtained.
The invention also provides a device for detecting the marked line region of the epidemic prevention kit, which is shown in figure 5 and comprises:
the image acquisition module is used for acquiring a training data image, and the training data image comprises a kit image in a display marking area;
the marking frame setting module is used for setting a rectangular marking frame so that the marking line area is positioned in the rectangular marking frame;
the characteristic point anchoring module is used for acquiring absolute coordinate data of a first point and a second point on the rectangular marking frame, the first point and the second point are in a mutual diagonal position relation on the rectangular marking frame, and the first point and the second point have a specific relation relative to a first reference system of the kit picture;
the storage module is used for sequentially storing the absolute coordinate data of the first point and the second point;
and the detection training module is used for inputting the training data picture and the corresponding labeling information into a detection training model, the detection training model comprises an activation function, and the activation function comprises a sigmoid activation function so as to output the normalized labeling information.
The invention also provides a computer readable storage medium, which stores computer instructions, and when the computer is executed by a processor, the method for detecting the marked line region of the epidemic prevention kit is realized.
In specific implementation, the computer-readable storage medium is a magnetic Disk, an optical Disk, a Read-only Memory (ROM), a Random Access Memory (RAM), a Flash Memory (Flash Memory), a Hard Disk (Hard Disk Drive, abbreviated as HDD), or a Solid-State Drive (SSD); the computer readable storage medium may also include a combination of memories of the above kinds.
The present invention also provides a computer device, as shown in fig. 6, comprising at least one processor, and a memory communicatively connected to the processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor, so that the processor performs a method for detecting a reticle field of an epidemic prevention kit according to any one of the above.
In particular, the number of processors may be one or more, and the processor may be a Central Processing Unit (CPU). The Processor may also be other general purpose Processor, digital Signal Processor (DSP), application Specific Integrated Circuit (ASIC), field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component, or a combination thereof. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory and the processor can be connected through a bus or other communication mode, the memory stores instructions which can be executed by at least one processor, and the instructions are executed by at least one processor, so that the processor executes the epidemic prevention kit marked line region detection method.
Based on the above, compared with the prior art, the method for detecting the marked line region of the epidemic prevention kit provided by the invention has the advantages that the marked line region is positioned in the rectangular marking frame by setting the rectangular marking frame, the absolute coordinate data of the first point and the second point on the rectangular marking frame are obtained, the first point and the second point are in a mutual diagonal position relation on the rectangular marking frame, and the first point and the second point have a specific relation relative to the first reference system of the kit picture; and then, the absolute coordinate data of the first point and the second point are sequentially stored as the marking information, so that the rotation angle information of the marking line region is hidden in the marking information, and the design of a subsequent detection training model can be further simplified while the marking information is simplified.
In addition, it will be appreciated by those skilled in the art that, although there may be many problems with the prior art, each embodiment or aspect of the present invention may be improved only in one or several respects, without necessarily simultaneously solving all the technical problems listed in the prior art or in the background. It will be understood by those skilled in the art that nothing in a claim should be taken as a limitation on that claim.
Although terms such as training data pictures, reticle fields, kit pictures, rectangular labeling boxes, etc. are used more often herein, the possibility of using other terms is not excluded. These terms are used merely to more conveniently describe and explain the nature of the present invention; they are to be construed as being without limitation to any one or more of the appended limitations; the terms "first," "second," and the like in the description and in the claims, and in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and these modifications or substitutions do not depart from the spirit of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A method for detecting the marked line area of an epidemic prevention kit is characterized by comprising the following steps:
s10: acquiring a training data picture, wherein the training data picture comprises a kit picture for displaying a marking area;
s20: setting a rectangular marking frame to enable the marking line area to be located in the rectangular marking frame;
s30: acquiring absolute coordinate data of a first point and a second point on the rectangular marking frame, wherein the first point and the second point are in a mutual diagonal position relation on the rectangular marking frame, and the first point and the second point have a specific relation relative to a first reference system of the kit picture;
s40: respectively normalizing the absolute coordinate data of the first point and the absolute coordinate data of the second point, and then sequentially storing the normalized absolute coordinate data as the labeling information of the training data picture;
s50: and inputting the training data picture and the corresponding labeling information into a detection training model, wherein the detection training model comprises an activation function, and the activation function comprises a sigmoid activation function so as to output the normalized labeling information.
2. The method for detecting the marking area of the epidemic prevention kit according to claim 1, wherein the method comprises the following steps: the first reference system is fixed relative to the arrangement direction of the kit picture, and the longitudinal axis of the first reference system points to the directions of letters from 'T' to 'C' in the kit picture.
3. The method for detecting the marking area of the epidemic prevention kit according to claim 2, wherein the method comprises the following steps: the specific relationship includes: for the labeled boxes in different training data pictures, a vector formed by the first point A and the second point B in the first reference system
Is the vector for a particular quadrant.
4. The method for detecting the marking area of the epidemic prevention kit according to claim 3, wherein the method comprises the following steps: the particular quadrant includes a fourth quadrant.
5. The method for detecting the marking area of the epidemic prevention kit according to claim 3, wherein the method comprises the following steps: the step S40 includes:
s41: acquiring the width and height of the training data picture;
s41: and normalizing the absolute coordinate data (Xa, ya) of the first point A and the absolute coordinate data (Xb, yb) of the second point B according to the width and the height of the training data picture.
6. The method for detecting the marked line region of the epidemic prevention kit according to claim 5, wherein the method comprises the following steps: the normalization processing comprises the steps of calculating floating point numbers Xa2, ya2, xb2 and Yb2 through the following formulas, and storing the floating point number sequences (Xa 2, ya2, xb2 and Yb 2) into the information of the corresponding training data pictures; wherein,
Xa2＝Xa/width；
Ya2＝Ya/height；
Xb2＝Xb/width；
Yb2＝Yb/height。
7. the method for detecting the marking area of the epidemic prevention kit according to claim 6, wherein the method comprises the following steps: the detection training model further comprises a backbone network, a neck network and a head network;
the backbone network comprises one or more of VggNet, resNet, denseNet and Mobi LeNet, and is used for receiving the training data picture and the labeling information;
the neck network comprises a FPN network, the neck network receiving the output of the backbone network;
the head network comprises a fully connected network, the head network receiving an output of the backbone network or receiving an output of the neck network, the output of the head network being input to the activation function.
8. The utility model provides an epidemic prevention kit marking area detection device which characterized in that includes:
the image acquisition module is used for acquiring a training data image, and the training data image comprises a kit image in a display marking area;
the marking frame setting module is used for setting a rectangular marking frame so that the marking line area is positioned in the rectangular marking frame;
the characteristic point anchoring module is used for acquiring absolute coordinate data of a first point and a second point on the rectangular marking frame, the first point and the second point are in a mutual diagonal position relation on the rectangular marking frame, and the first point and the second point have a specific relation relative to a first reference system of the kit picture;
the storage module is used for sequentially storing the absolute coordinate data of the first point and the second point;
and the detection training module is used for inputting the training data picture and the corresponding labeling information into a detection training model, the detection training model comprises an activation function, and the activation function comprises a sigmoid activation function so as to output the normalized labeling information.
9. A computer-readable storage medium characterized by: the computer readable storage medium stores computer instructions, and when executed by a processor, the computer implements a method for detecting the marking area of an epidemic prevention kit according to any one of claims 1-7.
10. A computer device, characterized by: comprising at least one processor, and a memory communicatively coupled to the processor, wherein the memory stores instructions executable by the at least one processor, the instructions being executable by the at least one processor to cause the processor to perform a method of epidemic prevention kit reticle field detection as recited in any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211015182.0A CN115409979A (en) | 2022-08-23 | 2022-08-23 | Epidemic prevention kit marked line region detection method, device, medium and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211015182.0A CN115409979A (en) | 2022-08-23 | 2022-08-23 | Epidemic prevention kit marked line region detection method, device, medium and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115409979A true CN115409979A (en) | 2022-11-29 |
Family
ID=84162582
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211015182.0A Pending CN115409979A (en) | 2022-08-23 | 2022-08-23 | Epidemic prevention kit marked line region detection method, device, medium and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115409979A (en) |
-
2022
- 2022-08-23 CN CN202211015182.0A patent/CN115409979A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US12073542B2 (en) | Image processing method, image processing apparatus, and non-transitory storage medium | |
| US10769487B2 (en) | Method and device for extracting information from pie chart | |
| US9349076B1 (en) | Template-based target object detection in an image | |
| WO2019104879A1 (en) | Information recognition method for form-type image, electronic device and readable storage medium | |
| US11314989B2 (en) | Training a generative model and a discriminative model | |
| CN110674804A (en) | Text image detection method and device, computer equipment and storage medium | |
| CN110163087B (en) | Face gesture recognition method and system | |
| CN110490198A (en) | Text orientation bearing calibration, device, computer equipment and storage medium | |
| WO2015074521A1 (en) | Devices and methods for positioning based on image detection | |
| CN108446227A (en) | A kind of virtual machine browser compatibility test method based on selenium | |
| US20200184697A1 (en) | Image Modification Using Detected Symmetry | |
| CN110264523A (en) | A kind of method and apparatus of the location information of target image in determining test image | |
| CN111985469A (en) | Method and device for recognizing characters in image and electronic equipment | |
| US20120170834A1 (en) | Determining the Uniqueness of a Model for Machine Vision | |
| WO2020097461A1 (en) | Convolutional neural networks with reduced attention overlap | |
| US20200394808A1 (en) | Aligning digital images by selectively applying pixel-adjusted-gyroscope alignment and feature-based alignment models | |
| CN117115823A (en) | Tamper identification method and device, computer equipment and storage medium | |
| CN115409979A (en) | Epidemic prevention kit marked line region detection method, device, medium and equipment | |
| CN117253022A (en) | Object identification method, device and inspection equipment | |
| CN116628250A (en) | Image generation method, device, electronic equipment and computer readable storage medium | |
| CN114862866B (en) | Calibration plate detection method and device, computer equipment and storage medium | |
| CN110503110A (en) | Feature matching method and device | |
| CN115439711A (en) | Kit marking area data set labeling method, device, medium and equipment | |
| CN115409978A (en) | Kit marking area detection loss evaluation method, device, medium and equipment | |
| Wang et al. | Surface defect detection method for electronic panels based on attention mechanism and dual detection heads |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20240116 Address after: 361008 Room 502, No.18, guanri Road, phase II, software park, Xiamen City, Fujian Province Applicant after: YLZ INFORMATION TECHNOLOGY Co.,Ltd. Address before: 361008 room 504, No.18, guanri Road, phase II, software park, Xiamen City, Fujian Province Applicant before: XIAMEN YILIANZHONG YIHUI TECHNOLOGY CO.,LTD. Applicant before: YLZ INFORMATION TECHNOLOGY Co.,Ltd. |