CN116185959B - Superimposed tag recognition and processing system - Google Patents
Superimposed tag recognition and processing system Download PDFInfo
- Publication number
- CN116185959B CN116185959B CN202310475841.7A CN202310475841A CN116185959B CN 116185959 B CN116185959 B CN 116185959B CN 202310475841 A CN202310475841 A CN 202310475841A CN 116185959 B CN116185959 B CN 116185959B
- Authority
- CN
- China
- Prior art keywords
- tag
- detection
- label
- type
- list
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/14—Details of searching files based on file metadata
- G06F16/148—File search processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/16—File or folder operations, e.g. details of user interfaces specifically adapted to file systems
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Library & Information Science (AREA)
- Human Computer Interaction (AREA)
- User Interface Of Digital Computer (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a superimposed tag identification and processing system, which comprises a server and a memory storing a computer program, wherein the memory also stores a preset tag type list and a corresponding tag type set threshold value list, and when the server executes the computer program, the following steps are realized: inputting the target file into a preset tag detection model to obtain a first tag list; when the first detection characteristic value of the first tag is greater than the lowest tag threshold value, marking the first tag as a second tag so as to obtain a second tag list; when the second label is not overlapped with other second labels and the first detection characteristic value corresponding to the second label is larger than the label type setting threshold value, the second label is reserved and mapped to the corresponding label type catalogue, and therefore the needed label is obtained more comprehensively.
Description
Technical Field
The invention relates to the field of tag identification, in particular to a superimposed tag identification and processing system.
Background
Tags act as special markers in the document and play an important role in the document identification process. The accurate identification tag is beneficial to correct arrangement of files, particularly, the tags such as signatures, fingerprints and seals in the files are required to be identified and arranged, a lot of resources are required to be consumed, the condition that the tags cannot be identified often occurs, and in practical application, the existing tag identification method mainly comprises the following steps: scoring the labels in the file through the neural network model, and extracting the labels with the scores meeting a preset threshold value; however, the existing tag recognition method cannot accurately recognize the tag of the text, and particularly, in the case that two tags are superimposed together, it often happens that only one tag is recognized.
Disclosure of Invention
Aiming at the technical problems, the invention adopts the following technical scheme: an overlay tag identification and processing system, the system comprising: server and memory storing computer program, said memory also storing preset tag type list t= { T 1 ，T 2 ，…，T j ，…，T n Threshold list c= { C for setting of the tag type corresponding to the same 1 ，C 2 ，…，C j ，…，C n }，T j Is the j-th preset label type, C j Is T j Setting a threshold value for the corresponding tag type, wherein the value range of j is 1 to n, n is the number of preset tag types, and when the server executes a computer program, the following steps are realized:
s100, inputting the target file into a preset tag detection model to obtain a first tag list A= { A 1 ，A 2 ，…，A i ，…，A m -wherein the i-th first tag a i The method at least comprises first position information of the tag in the target file, a first detection characteristic value and a first tag type, wherein the first detection characteristic value is used for representing the probability that the first tag is a real tag, the first tag type is one of a preset tag type list, the value range of i is 1 to m, and m is the number of the first tags in the target file.
S200, when A i Is the first detected characteristic value of (1)>S 0 When marking A i Obtaining a second tag list b= { B for the second tag 1 ，B 2 ，…，B r ，…，B s S, where S 0 Is the lowest label thresholdAnd S is 0 ≤min{C 1 ，C 2 ，…，C j ，…，C n }，B r Is the r second label, the value range of r is 1 to s, and s is the number of the second labels.
S300, based on B r And B k Obtain the first position information of B r And B k Overlap of I r，k Wherein the value range of k is 1 to s, and k is not equal to r.
S400, when B r And any other B than itself k Overlap of 0 and B r Corresponding first detection characteristic value S r >C r When reserve B r And B is combined with r Mapping to a corresponding tag type directory, wherein S r Sum of B r The first detection characteristic values of the corresponding first labels are the same, C r Is with B r And setting a threshold value for the label type corresponding to the corresponding first label type.
The invention has at least the following beneficial effects: the invention provides a superposition label and processing system, which inputs a target file into a preset label detection model, acquires a first label list, marks any first label as a second label when a first detection characteristic value of the first label is larger than a lowest label threshold value, calculates the overlapping degree of the second label and other second labels except the second label, judges whether the first detection characteristic value of the second label is larger than a label type setting threshold value corresponding to a first label type of the second label when the second label and other second labels except the second label are not overlapped, and reserves the second label when the first detection characteristic value of the second label is larger than the label type setting threshold value corresponding to the first label type of the second label and maps the second label to a corresponding label type catalog; in the invention, the lowest label threshold is adopted to screen out all possible labels, and then the required labels are screened out by the preset score threshold corresponding to the label type, so that the screened labels are more comprehensive and meet the requirement.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a superimposed tag identification and processing system according to an embodiment of the present invention when executing a computer program.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to fall within the scope of the invention.
The embodiment of the invention provides a superimposed tag identification and processing system, which comprises: server and memory storing computer program, said memory also storing preset tag type list t= { T 1 ，T 2 ，…，T j ，…，T n Threshold list c= { C for setting of the tag type corresponding to the same 1 ，C 2 ，…，C j ，…，C n }，T j Is the j-th preset label type, C j Is T j The corresponding label types are set with thresholds, the value range of j is 1 to n, and n is the number of preset label types.
Further, when the server executes the computer program, as shown in fig. 1, the following steps are implemented:
s100, inputting the target file into a preset tag detection model to obtain a first tag list A= { A 1 ，A 2 ，…，A i ，…，A m -wherein the i-th first tag a i At least comprises first position information of the tag in the target file, a first detection characteristic value and a first tag type, wherein the first detection characteristic value is used for representing the probability that the first tag is a real tag, and the first tagThe tag type is one of the preset tag type list, the value range of i is 1 to m, and m is the number of first tags in the target file.
In particular, the first label of the present invention includes, but is not limited to, a fingerprint, a signature, and a seal stamp.
Specifically, the preset label detection model is a Yolov5 model after training, the first position information of the first label is a rectangular outer bounding box where the first label is located, the form of the first position information can be the coordinates of the upper left corner point and the coordinates of the lower right corner point of the rectangular outer bounding box where the first label is located, the form of the first position information can also be the coordinates of the central point and the preset length and width of the rectangular outer bounding box where the first label is located, and of course, the form of the first position information is not limited to this and can be any form which can be realized in the prior art.
Further, those skilled in the art will recognize that any method for obtaining a trained Yolov5 model in the prior art falls within the scope of the present invention, specifically, L 3 Inputting the historical target file into a Yolov5 model, acquiring the historical position, the historical score and the historical type of the historical target file, and passing through and marking L in advance 3 And comparing the marking positions, marking scores and marking types of the historical target files, calculating a loss function, and obtaining the trained Yolov5 model by minimizing the loss function.
S200, when A i Is the first detected characteristic value of (1)>S 0 When marking A i Obtaining a second tag list b= { B for the second tag 1 ，B 2 ，…，B r ，…，B s S, where S 0 Is the lowest tag threshold and S 0 ≤min{C 1 ，C 2 ，…，C j ，…，C n }，B r Is the r second label, the value range of r is 1 to s, and s is the number of the second labels.
Specifically, the invention also includes a lowest tag threshold S 0 The method comprises the following steps of:
s001, initialize S 0 =S d Wherein S is d Is S 0 Is a numerical value for initialization.
S002, testing the target file set U= { U 1 ，U 2 ，…，U y ，…，U L1 Inputting a preset label detection model to obtain a first test label set V= { V 1 ，V 2 ，…，V y ，…，V L1 }, wherein V is y ={V y1 ，V y2 ，…，V yw ，…，V yW }，U y Is the y-th test object file in U, V y Is U y Corresponding first test tag list, V y W first test label V yw At least comprises the test label in a test target file U y In (1), a first detection characteristic value, a first label type, y ranging from 1 to L1, W ranging from 1 to W, L1 being the number of test target files in U, W being V y Is the number of first test tags in the database.
S003, when the first detection characteristic value corresponding to any one of the first test labels is greater than S 0 The first test tag is then marked as a second test tag.
S004, obtaining the number E of the second test labels 1 。
Specifically, the number E of the second test tags 1 Is the number of all second test tags in the test object file set.
S005, when (E 1 /E 2 ）>E 0 At the time, acquire S 0 Wherein E is 2 Is that the true characteristic value of U is larger than S 0 E, all numbers of tags of (2) 0 Is a preset recall threshold.
Specifically, the real characteristic value can be determined by a manual method; the preset recall threshold E 0 Can be determined according to actual requirements.
S006, when E 1 /E 2 ≤E 0 At the time S 0 =S 0 +S t Return to execution S002, wherein S t Is a preset growth factor.
In summary, the invention is implemented by initializing S 0 Test object textInputting the part set into a preset label detection model to obtain a first test label set, and when a first detection characteristic value corresponding to any one of the first test labels is greater than S 0 Marking the first test label as a second test label, obtaining the number of the second test labels, and when the number of the second test labels and the true characteristic value are greater than S 0 When the ratio of the number of all tags is larger than the preset recall threshold, the recall at the moment is considered to meet the use requirement, and S at the moment is considered to be 0 As the lowest label threshold, the recall rate of the second labels meets the requirement by the method, and the second labels are recalled as much as possible.
Further, in an embodiment of the present invention, S 0 =0.2。
S300, based on B r And B k Obtain the first position information of B r And B k Overlap of I r，k Wherein the value range of k is 1 to s, and k is not equal to r.
Specifically, the B r And B k Is the value B r And B k IoU value of (2).
Further, B r And B k Overlap of I r，k The method comprises the following steps of:
s301 based on B r And B k Obtain the first position of B r Area X of (2) r And B k Area X of (2) k 。
S303, obtain I r，k =（X r ∩X k ）/（X r ∪X k ）。
S400, when B r And any other B than itself k Overlap of 0 and B r Corresponding first detection characteristic value S r >C r When reserve B r And B is combined with r Mapping to a corresponding tag type directory, wherein S r Sum of B r The first detection characteristic values of the corresponding first labels are the same, C r Is with B r And setting a threshold value for the label type corresponding to the corresponding first label type.
Specifically, B r And removingAny one B other than itself k The overlapping degree of the first label is 0, and it can be understood that the r second label and any one of the other second labels except the r second label are not overlapped, the first detection characteristic value corresponding to the second label is larger than the label type setting threshold corresponding to the first label type, and the label is reserved, namely the label is considered to be the label required in the invention.
In summary, the present invention provides a system for stacking labels and processing, which inputs a target file into a preset label detection model, obtains a first label list, marks a first label as a second label when a first detection feature value of any first label is greater than a lowest label threshold, calculates overlapping degrees of the second label and other second labels except the second label, judges whether the first detection feature value of the second label is greater than a label type setting threshold corresponding to a first label type of the second label when the second label and other second labels except the second label are not overlapped, and reserves the second label when the first detection feature value of any first label is greater than the first label type setting threshold, and maps the second label to a corresponding label type directory; in the invention, the lowest label threshold is adopted to screen out all possible labels, and then the required labels are screened out by the preset score threshold corresponding to the label type, so that the screened labels are more comprehensive and meet the requirement.
Further, after S400, the method further includes:
s401, when I r，k When not equal to 0, obtain HT r，k =C r -q*I r，k *（1-CD r，k /DD r，k ) Q is a preset weight parameter, CD r，k Is B r 、B k Center point distance, DD, of two corresponding rectangular bounding boxes r，k Is B r 、B k The length of a diagonal line of the minimum circumscribed rectangular outer bounding box enclosed by the corresponding two rectangular bounding boxes.
Specifically, the B r And B k The corresponding rectangular bounding box is determined according to the first position information output by the preset label detection model, and the B is that r And B k The corresponding rectangular bounding box may be the coordinates of the upper left corner and the coordinates of the lower right corner, or may be the coordinates of the center point of the rectangular bounding box and the preset length and width, and of course, the form of the first position information is not limited to this, and may be any form that can be realized in the prior art.
Further, q is more than or equal to 0.1 and less than or equal to 0.2.
S403, when B r Corresponding first detection characteristic value>HT r，k When reserve B r And B is combined with r Mapping to the corresponding label type directory.
To sum up, when B r And B k Acquisition of HT when superimposed r，k When B r The corresponding first detection characteristic value is larger than HT r，k When reserve B r And B is combined with r Mapping to a corresponding label type directory; it can be understood that when B r And B k When superimposed, B r It is likely that a smaller first detection feature value will be obtained due to superposition, and it is likely that B will be r Missing, therefore, the method considers the center point distance of the rectangular bounding boxes of the two labels which are overlapped and the length of one diagonal line of the minimum circumscribed rectangular outer bounding box which is enclosed by the rectangular bounding boxes of the two labels, and calculates HT r，k Thereby obtaining the required label more accurately.
Specifically, the invention also comprises a label type setting threshold C j The method comprises the following steps of:
s010, obtain T j Corresponding initial score list H j ={H j1 ，H j2 ，…，H jg ，…，H jz }，H jg Is T j The corresponding g initial score, g has a value ranging from 1 to z, z being T j Number of corresponding initial scores.
Specifically, the initial score is T j The corresponding score, which may be thresholded for the tag type, may be manually determined based on actual demand.
S020, the detection target file set cu= { CU 1 ，CU 2 ，…，CU β ，…，CU L2 Inputting a preset label detection model to obtain a first detection label set CV= { CV 1 ，CV 2 ，…，CV β ，…，CV L2 And (3) with the beta detection target file CU β Corresponding first detection tag list CV β ={CV β1 ，CV β2 ，…，CV βη ，…，CV βη1 }，CU β In eta first detection label CV βη At least comprises the detection label in CU β The value range of beta is 1 to L2, L2 is the number of detection target files in the CU, the value range of eta is 1 to eta 1, eta 1 is the CU β The number of first detection tags in the database.
S030, as CV βη Corresponding first detection characteristic value>S 0 At the time, CV is marked βη A second list of detection tags is obtained for the second detection tag.
Specifically, inputting the detection target file set into a preset tag detection model to obtain a first detection tag set, and when the first detection characteristic value of any first detection tag is greater than S 0 The first detection tag is marked as a second detection tag, so that a second detection tag list is obtained.
S040, when the first tag type=t corresponding to the second detection tag j And the first detection characteristic value corresponding to the second detection label is more than H jg The second detection tag is then marked as a third detection tag.
S050, pair H jg Obtaining H jg Corresponding harmonic mean F jg =2*P jg *R jg /（P jg +R jg ) Thereby obtaining a harmonic mean list F j ={F j1 ，F j2 ，…，F jg ，…，F jz (wherein R is jg =TP jg /（TP jg +FN jg ），P jg =TP jg /（TP jg +FP jg ），TP jg For the number of third detection labels determined to be the same as the true labels in the L2 detection target files, FP jg For the number of the third detection label which is judged to be different from the actual label in the L2 detection target files, TN jg For determining the number of the detection target file which is not the third detection label and is different from the true label, FN jg The number of the detection target file which is not determined to be the third detection label and is different from the actual label is L2.
Specifically, the real tag may be determined manually.
S060, obtain C j =H j0 ，H j0 Is F j0 Corresponding initial score, wherein F j0 =max{F j1 ，F j2 ，…，F jg ，…，F jz }。
Based on S010-S060, the method and the device acquire the reconciliation average value corresponding to each initial score under each preset label type by presetting an initial score list and detecting a detection target file set, and take the initial score corresponding to the largest reconciliation average value as a label type setting threshold corresponding to the preset label type.
While certain specific embodiments of the invention have been described in detail by way of example, it will be appreciated by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the invention. Those skilled in the art will also appreciate that many modifications may be made to the embodiments without departing from the scope and spirit of the invention. The scope of the invention is defined by the appended claims.
Claims (8)
1. An overlay tag identification and processing system, the system comprising: server and memory storing computer program, said memory also storing preset tag type list t= { T 1 ，T 2 ，…，T j ，…，T n Threshold list c= { C for setting of the tag type corresponding to the same 1 ，C 2 ，…，C j ，…，C n }，T j Is the j-th preset label type, C j Is T j Setting a threshold value for the corresponding tag type, wherein the value range of j is 1 to n, n is the number of preset tag types, and when the server executes a computer program, the following steps are realized:
s100, inputting the target file into a preset tag detection model to obtain a first tag list A= { A 1 ，A 2 ，…，A i ，…，A m -wherein the i-th first tag a i The method comprises the steps of at least including first position information of a first tag in a target file, a first detection characteristic value and a first tag type, wherein the first detection characteristic value is used for representing the probability that the first tag is a real tag, the first tag type is one of a preset tag type list, the value range of i is 1 to m, and m is the number of the first tags in the target file;
s200, when A i Is the first detected characteristic value of (1)>S 0 When marking A i Obtaining a second tag list b= { B for the second tag 1 ，B 2 ，…，B r ，…，B s S, where S 0 Is the lowest tag threshold and S 0 ≤min{C 1 ，C 2 ，…，C j ，…，C n }，B r Is the r second label, the value range of r is 1 to s, and s is the number of the second labels;
s300, based on B r And B k Obtain the first position information of B r And B k Overlap of I r，k Wherein the value range of k is 1 to s, and k is not equal to r;
s400, when B r And any other B than itself k Overlap of 0 and B r Corresponding first detection characteristic value S r >C r When reserve B r And B is combined with r Mapping to a corresponding tag type directory, wherein S r Sum of B r The first detection characteristic values of the corresponding first labels are the same, C r Is with B r And setting a threshold value for the label type corresponding to the corresponding first label type.
2. The superimposed tag identification and processing system according to claim 1, wherein a lowest tag threshold S 0 The method comprises the following steps of:
s001, initialize S 0 =S d Wherein S is d Is S 0 Is a initialization value of (1);
s002, testing the target file set U= { U 1 ，U 2 ，…，U y ，…，U L1 Inputting a preset label detection model to obtain a first test label set V= { V 1 ，V 2 ，…，V y ，…，V L1 }, wherein V is y ={V y1 ，V y2 ，…，V yw ，…，V yW }，U y Is the y-th test object file in U, V y Is U y Corresponding first test tag list, V y W first test label V yw At least comprises the test label in a test target file U y In (1), a first detection characteristic value, a first label type, y ranging from 1 to L1, W ranging from 1 to W, L1 being the number of test target files in U, W being V y The number of first test tags in (a);
s003, when the first detection characteristic value corresponding to any one of the first test labels is greater than S 0 When the first test label is marked as a second test label;
s004, obtaining the number E of the second test labels 1 ；
S005, when (E 1 /E 2 ）>E 0 At the time, acquire S 0 Wherein E is 2 Is that the true characteristic value of U is larger than S 0 E, all numbers of tags of (2) 0 A recall threshold is preset;
s006, when E 1 /E 2 ≤E 0 At the time S 0 =S 0 +S t Return to execution S002, wherein S t Is a preset growth factor.
3. The superimposed tag identification and processing system according to claim 2, wherein S 0 =0.2。
4. The superimposed tag identification and processing system according to claim 1, wherein the tag type sets a threshold C j The method comprises the following steps of:
s010, obtain T j Corresponding initial score list H j ={H j1 ，H j2 ，…，H jg ，…，H jz }，H jg Is T j The corresponding g initial score, g has a value ranging from 1 to z, z being T j The number of corresponding initial scores;
s020, the detection target file set cu= { CU 1 ，CU 2 ，…，CU β ，…，CU L2 Inputting a preset label detection model to obtain a first detection label set CV= { CV 1 ，CV 2 ，…，CV β ，…，CV L2 And (3) with the beta detection target file CU β Corresponding first detection tag list CV β ={CV β1 ，CV β2 ，…，CV βη ，…，CV βη1 }，CU β In eta first detection label CV βη At least comprises the detection label in CU β The value range of beta is 1 to L2, L2 is the number of detection target files in the CU, the value range of eta is 1 to eta 1, eta 1 is the CU β The number of first detection tags in (a);
s030, as CV βη Corresponding first detection characteristic value>S 0 At the time, CV is marked βη Obtaining a second detection tag list for the second detection tag;
s040, when the first tag type=t corresponding to the second detection tag j And the first detection characteristic value corresponding to the second detection label is more than H jg Marking the second detection tag as a third detection tag;
s050, pair H jg Obtaining H jg Corresponding harmonic mean F jg =2*P jg *R jg /（P jg +R jg ) Thereby obtaining a harmonic mean list F j ={F j1 ，F j2 ，…，F jg ，…，F jz (wherein R is jg =TP jg /（TP jg +FN jg ），P jg =TP jg /（TP jg +FP jg ），TP jg For the number of third detection labels determined to be the same as the true labels in the L2 detection target files, FP jg FN for the number of the third detection label determined as the third detection label but different from the true label in the L2 detection target file jg The number of the third detection labels which are not judged to be the third detection labels and are different from the number of the real labels in the L2 detection target files;
s060, obtain C j =H j0 ，H j0 Is F j0 Corresponding initial score, wherein F j0 =max{F j1 ，F j2 ，…，F jg ，…，F jz }。
5. The superimposed tag identification and processing system according to claim 1, further comprising, after S400:
s401, when I r，k When not equal to 0, obtain HT r，k =C r -q*I r，k *（1-CD r，k /DD r，k ) Q is a preset weight parameter, CD r，k Is B r 、B k Center point distance, DD, of two corresponding rectangular bounding boxes r，k Is B r 、B k The length of a diagonal line of the minimum circumscribed rectangular outer bounding box enclosed by the two corresponding rectangular bounding boxes;
s403, when B r Corresponding first detection characteristic value>HT r，k When reserve B r And B is combined with r Mapping to the corresponding label type directory.
6. The superimposed tag identification and processing system according to claim 1, wherein in S300, B r And B k Overlap of I r，k The method comprises the following steps of:
s301 based on B r And B k Obtain the first position of B r Area X of (2) r And B k Area X of (2) k ；
S303, obtain I r，k =（X r ∩X k ）/（X r ∪X k ）。
7. The superimposed tag identification and processing system of claim 5, wherein 0.1.ltoreq.q.ltoreq.0.2.
8. The superimposed tag identification and processing system of claim 1, wherein the predetermined tag detection model is a trained Yolov5 model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310475841.7A CN116185959B (en) | 2023-04-28 | 2023-04-28 | Superimposed tag recognition and processing system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310475841.7A CN116185959B (en) | 2023-04-28 | 2023-04-28 | Superimposed tag recognition and processing system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116185959A CN116185959A (en) | 2023-05-30 |
| CN116185959B true CN116185959B (en) | 2023-06-30 |
Family
ID=86449363
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310475841.7A Active CN116185959B (en) | 2023-04-28 | 2023-04-28 | Superimposed tag recognition and processing system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116185959B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2000028676A1 (en) * | 1998-11-05 | 2000-05-18 | Mitsubishi Materials Corporation | Method of identifying overlapping tags |
| CN107122810A (en) * | 2017-03-10 | 2017-09-01 | 杨玉芹 | A kind of label authentic authentication method based on RFID physical layer feature |
| CN115063564A (en) * | 2022-07-13 | 2022-09-16 | 如你所视(北京)科技有限公司 | Article label display method, device and medium for two-dimensional display image |
-
2023
- 2023-04-28 CN CN202310475841.7A patent/CN116185959B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2000028676A1 (en) * | 1998-11-05 | 2000-05-18 | Mitsubishi Materials Corporation | Method of identifying overlapping tags |
| CN107122810A (en) * | 2017-03-10 | 2017-09-01 | 杨玉芹 | A kind of label authentic authentication method based on RFID physical layer feature |
| CN115063564A (en) * | 2022-07-13 | 2022-09-16 | 如你所视(北京)科技有限公司 | Article label display method, device and medium for two-dimensional display image |
Non-Patent Citations (1)
| Title |
|---|
| UHF RFID标签防碰撞算法分析与研究;吴凡;《硕士电子期刊》;1-82 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116185959A (en) | 2023-05-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111860674B (en) | Sample category identification method, sample category identification device, computer equipment and storage medium | |
| US20210182494A1 (en) | Post-filtering of named entities with machine learning | |
| WO2019238063A1 (en) | Text detection and analysis method and apparatus, and device | |
| CN109658584B (en) | Bill information identification method and device | |
| TWI484423B (en) | Face recognition apparatus and methods | |
| US20120143856A1 (en) | Method for detecting object | |
| CN114419736B (en) | Experiment scoring method, system, equipment and readable storage medium | |
| TWI752455B (en) | Image classification model training method, image processing method, data classification model training method, data processing method, computer device, and storage medium | |
| CN110705952A (en) | Contract auditing method and device | |
| CN105930873B (en) | A kind of walking across mode matching method certainly based on subspace | |
| CN108614895B (en) | Abnormal data access behavior identification method and data processing device | |
| CN108509988B (en) | Test paper score automatic statistical method and device, electronic equipment and storage medium | |
| CN107392221B (en) | Training method of classification model, and method and device for classifying OCR (optical character recognition) results | |
| CN114937179B (en) | Junk image classification method and device, electronic equipment and storage medium | |
| CN111325069A (en) | Production line data processing method and device, computer equipment and storage medium | |
| CN112580108A (en) | Signature and seal integrity verification method and computer equipment | |
| CN111126268A (en) | Key point detection model training method and device, electronic equipment and storage medium | |
| CN116185959B (en) | Superimposed tag recognition and processing system | |
| CN111461143A (en) | Picture copying identification method and device and electronic equipment | |
| CN111178203A (en) | Signature verification method and device, computer equipment and storage medium | |
| CN112036362A (en) | Image processing method, image processing device, computer equipment and readable storage medium | |
| CN108764344B (en) | Method and device for identifying card based on edge and storage equipment | |
| CN111104846B (en) | Data detection method and device, computer equipment and storage medium | |
| US20220222568A1 (en) | System and Method for Ascertaining Data Labeling Accuracy in Supervised Learning Systems | |
| CN114220103A (en) | Image recognition method, device, equipment and computer readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |