CN115018492A - Smart campus automatic checkout method and system based on artificial intelligence - Google Patents
Smart campus automatic checkout method and system based on artificial intelligence Download PDFInfo
- Publication number
- CN115018492A CN115018492A CN202210838618.XA CN202210838618A CN115018492A CN 115018492 A CN115018492 A CN 115018492A CN 202210838618 A CN202210838618 A CN 202210838618A CN 115018492 A CN115018492 A CN 115018492A
- Authority
- CN
- China
- Prior art keywords
- commodity
- image
- detected
- template
- images
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/08—Payment architectures
- G06Q20/18—Payment architectures involving self-service terminals [SST], vending machines, kiosks or multimedia terminals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/38—Payment protocols; Details thereof
- G06Q20/40—Authorisation, e.g. identification of payer or payee, verification of customer or shop credentials; Review and approval of payers, e.g. check credit lines or negative lists
- G06Q20/401—Transaction verification
- G06Q20/4014—Identity check for transactions
- G06Q20/40145—Biometric identity checks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/15—Cutting or merging image elements, e.g. region growing, watershed or clustering-based techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/153—Segmentation of character regions using recognition of characters or words
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/18—Extraction of features or characteristics of the image
- G06V30/18143—Extracting features based on salient regional features, e.g. scale invariant feature transform [SIFT] keypoints
- G06V30/18152—Extracting features based on a plurality of salient regional features, e.g. "bag of words"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/19—Recognition using electronic means
- G06V30/19007—Matching; Proximity measures
- G06V30/19093—Proximity measures, i.e. similarity or distance measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
Abstract
The invention discloses an automatic intelligent campus account-settling method and system based on artificial intelligence, and relates to the technical field of data processing. The method comprises the following steps: establishing a commodity template image library, and recording price information of a corresponding commodity; acquiring and optimizing an image of a commodity to be settled out to obtain an image of the commodity to be detected; carrying out example segmentation on the commodity image to be detected to obtain a plurality of commodity segmentation area images to be detected; screening to obtain a primary commodity template image corresponding to each commodity segmentation area image to be detected; performing character recognition to determine a target commodity to be settled; counting the total consumption amount; acquiring and importing a target face image into a preset face recognition model to generate an identity recognition result; and extracting the corresponding checkout account from the corresponding smart campus checkout system, and deducting money from the corresponding checkout account according to the total consumption amount. The invention can greatly improve the commodity and face recognition precision, thereby ensuring the precision and effectiveness of automatic checkout.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to an automatic intelligent campus account-settling method and system based on artificial intelligence.
Background
With the development of the times and the progress of science and technology, the construction of the smart campus is more and more concerned. As an important component in a smart campus, the campus automatic checkout system can provide great convenience for teachers and students, but the existing campus automatic checkout system has the problems of commodity error identification, identity error identification and the like, and the practical application value of the campus automatic checkout system is obviously reduced.
In recent years, with the continuous update of a plurality of technologies in the field of artificial intelligence, the technology can provide direct support for the construction of an automatic checkout system of a smart campus. Therefore, how to combine the artificial intelligence technology to improve the accuracy of the smart campus automatic checkout and promote the smart campus construction becomes a new problem.
Disclosure of Invention
In order to overcome the above problems or at least partially solve the above problems, embodiments of the present invention provide an automated checkout method and system for a smart campus based on artificial intelligence, which can greatly improve the precision of recognition of goods and faces, and further ensure the precision and effectiveness of automated checkout.
The embodiment of the invention is realized by the following steps:
in a first aspect, an embodiment of the present invention provides an artificial intelligence based automated checkout method for a smart campus, including the following steps:
acquiring and establishing a commodity template image library according to the images of the commodities sold in the campus, and recording price information of each corresponding commodity in the commodity template image library;
acquiring an image of a commodity to be settled, and optimizing the image of the commodity to be settled by using an image enhancement method to obtain an image of the commodity to be detected;
carrying out example segmentation on the commodity image to be detected by adopting an example segmentation method to obtain a plurality of commodity segmentation area images to be detected;
respectively calculating the similarity of each to-be-detected commodity segmentation area image and all commodity images in a commodity template image library by using a sparse matching similarity detection method based on an image pyramid so as to obtain a primary commodity template image corresponding to each to-be-detected commodity segmentation area image through screening;
respectively carrying out character recognition on the preliminary commodity template images corresponding to the to-be-detected commodity segmentation area images by adopting a multi-scale OCR recognition method so as to obtain and determine a target to-be-checked commodity according to the character recognition results of the to-be-detected commodity segmentation area images and the character recognition results of the preliminary commodity templates under each scale;
acquiring and counting price information of each target commodity to be settled to obtain total consumption amount;
acquiring and importing a target face image into a preset face recognition model to generate an identity recognition result;
and extracting a corresponding checkout account from the corresponding smart campus checkout system according to the identity recognition result, and deducting money from the corresponding checkout account according to the total consumption amount to finish automatic checkout.
In order to solve the problem that errors exist in automatic checkout due to the fact that the problems of commodity misidentification, identity misidentification and the like exist in the prior art, the method and the device perform deep optimization and commodity example segmentation processing on a plurality of commodity images to be detected by means of an image enhancement technology and an example segmentation technology, and provide effective support for accurate identification of each commodity; and the similarity between the image of the commodity segmentation area to be detected and the template commodity image is calculated by using a sparse matching similarity detection method based on the image pyramid, so that the template commodity image with the highest similarity with the image of the commodity segmentation area to be detected, namely a primary commodity template image, can be more accurately retrieved. Meanwhile, character recognition is carried out on the segmented region image of the commodity to be detected and the primary commodity template image by utilizing a multi-scale OCR technology, so that the character recognition and comparison precision is remarkably improved; the automatic extraction type face recognition model is utilized to complete the recognition of the identity of the account holder, so that the account holder does not need to spend extra time for collecting face images, and the accuracy of face recognition is ensured. The invention can greatly improve the commodity and face recognition precision, thereby ensuring the precision and effectiveness of automatic checkout.
Based on the first aspect, in some embodiments of the present invention, the method for calculating the similarity between each segmented region image of the to-be-detected commodity and all commodity images in the commodity template image library by using the image pyramid-based sparse matching similarity detection method includes the following steps:
respectively carrying out sparse coding on each commodity image in each commodity segmentation area image to be detected and each commodity image in the commodity template image library to obtain a corresponding commodity image code to be detected and a corresponding template image code;
calculating the similarity between each to-be-detected commodity segmentation area image and each commodity image in the commodity template image library according to each to-be-detected commodity image code and each template image code so as to determine a first commodity template image corresponding to each to-be-detected commodity segmentation area image;
respectively carrying out multi-equal division processing on each to-be-detected commodity segmentation area image and the corresponding first commodity template image to obtain the to-be-detected commodity image and the first commodity template area image of the corresponding multiple equal division areas;
respectively carrying out sparse coding on the commodity image to be detected and the first commodity template area image of each equally divided area to obtain corresponding commodity image codes to be detected and first template image codes of a plurality of equally divided areas;
and calculating the similarity between each to-be-detected commodity segmentation area image and the corresponding first commodity template image according to the to-be-detected commodity image code and each template image code of each equally divided area so as to determine a primary commodity template image corresponding to each to-be-detected commodity segmentation area image.
Based on the first aspect, in some embodiments of the present invention, the method for obtaining the preliminary product template image corresponding to each segmented region image of the product to be detected by screening includes the following steps:
and counting and sequencing the corresponding preliminary commodity template images according to the number of similarity equal-dividing areas in the preliminary commodity template images corresponding to the commodity segmentation area images to be detected so as to obtain the optimal preliminary commodity template image by screening.
Based on the first aspect, in some embodiments of the present invention, the method for respectively performing character recognition on the preliminary product template images corresponding to the segmented region images of each to-be-detected product by using the multi-scale OCR recognition method includes the following steps:
carrying out multi-scale division on the preliminary commodity template image corresponding to each commodity segmentation area image to be detected so as to obtain the corresponding preliminary commodity template image under multiple scales;
and performing character recognition on the primary commodity template image under each scale by adopting an OCR recognition method.
Based on the first aspect, in some embodiments of the present invention, the method for determining a target commodity to be settled according to the character recognition result of each segmented area image of the commodity to be detected and the character recognition result of the preliminary commodity template under each scale includes the following steps:
counting character recognition results of each to-be-detected commodity segmentation area image under each scale to determine a first recognition result;
counting character recognition results of the primary commodity template under each scale to determine a second recognition result;
and determining the target commodity to be checked out according to the first recognition result and the second recognition result.
Based on the first aspect, in some embodiments of the present invention, the method for automated checkout in a smart campus based on artificial intelligence further includes the following steps:
acquiring a plurality of face images of a target person, and processing the plurality of face images by using a peak signal-to-noise ratio detection method based on super-resolution reconstruction so as to obtain an initial target face image by screening;
and comparing and matching the initial target face image with a preset standard face template image by using a multi-region accurate matching verification method so as to obtain a target face image by screening.
Based on the first aspect, in some embodiments of the invention, the example segmentation method includes one or more of Mask-RCNN algorithm, PANet example segmentation model, yolcat model, polarmack algorithm, and CenterMask example segmentation algorithm.
In a second aspect, an embodiment of the present invention provides an artificial intelligence-based smart campus automatic checkout system, including a basic recording module, an image optimization module, an instance segmentation module, a similarity detection module, a character recognition module, an amount statistics module, an identity recognition module, and a checkout module, where:
the basic recording module is used for acquiring and establishing a commodity template image library according to the images of the commodities sold in the campus, and recording the price information of each corresponding commodity in the commodity template image library;
the image optimization module is used for acquiring an image of the commodity to be settled, and optimizing the image of the commodity to be settled by using an image enhancement method to obtain an image of the commodity to be detected;
the example segmentation module is used for carrying out example segmentation on the commodity image to be detected by adopting an example segmentation method so as to obtain a plurality of commodity segmentation area images to be detected;
the similarity detection module is used for respectively calculating the similarity of each to-be-detected commodity segmentation area image and all commodity images in the commodity template image library by using a sparse matching similarity detection method based on an image pyramid so as to obtain a primary commodity template image corresponding to each to-be-detected commodity segmentation area image through screening;
the character recognition module is used for respectively carrying out character recognition on the preliminary commodity template images corresponding to the segmented area images of the commodities to be detected by adopting a multi-scale OCR recognition method so as to obtain and determine target commodities to be checked out according to the character recognition results of the segmented area images of the commodities to be detected and the character recognition results of the preliminary commodity templates under all scales;
the amount counting module is used for acquiring and counting price information of each target commodity to be settled, so as to obtain total consumption amount;
the identity recognition module is used for acquiring and importing a target face image into a preset face recognition model to generate an identity recognition result;
and the payment module is used for extracting a corresponding payment account from the corresponding smart campus payment system according to the identity recognition result, deducting money from the corresponding payment account according to the total consumption amount, and completing automatic payment.
In order to solve the problem that errors exist in automatic checkout due to the fact that the problems of commodity misidentification, identity misidentification and the like exist in the prior art, the system conducts deep optimization and commodity example segmentation processing on a plurality of commodity images to be detected through the cooperation of various modules such as a basic recording module, an image optimization module, an example segmentation module, a similarity detection module, a character recognition module, an amount counting module, an identity recognition module and a checkout module by means of an image enhancement technology and an example segmentation technology, and effective support is provided for accurate identification of each commodity; and the similarity between the image of the commodity segmentation area to be detected and the template commodity image is calculated by using a sparse matching similarity detection method based on the image pyramid, so that the template commodity image with the highest similarity with the image of the commodity segmentation area to be detected, namely a primary commodity template image, can be more accurately retrieved. Meanwhile, character recognition is carried out on the segmented region image of the commodity to be detected and the primary commodity template image by utilizing a multi-scale OCR technology, so that the character recognition and comparison precision is remarkably improved; the automatic extraction type face recognition model is utilized to complete the recognition of the identity of the account holder, so that the account holder does not need to spend extra time for collecting face images, and the accuracy of face recognition is ensured. The invention can greatly improve the commodity and face recognition precision, thereby ensuring the precision and effectiveness of automatic checkout.
In a third aspect, an embodiment of the present application provides an electronic device, which includes a memory for storing one or more programs; a processor. The program or programs, when executed by a processor, implement the method of any of the first aspects as described above.
In a fourth aspect, embodiments of the present application provide a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the method according to any one of the first aspect described above.
The embodiment of the invention at least has the following advantages or beneficial effects:
the embodiment of the invention provides an artificial intelligence-based smart campus automatic checkout method and system, which solve the problem that in the prior art, errors exist in automatic checkout due to the fact that the problems of commodity misidentification, identity misidentification and the like exist; and the similarity between the image of the commodity segmentation area to be detected and the template commodity image is calculated by using a sparse matching similarity detection method based on the image pyramid, so that the template commodity image with the highest similarity with the image of the commodity segmentation area to be detected, namely a primary commodity template image, can be more accurately retrieved. Meanwhile, character recognition is carried out on the commodity segmentation area image to be detected and the preliminary commodity template image by utilizing a multi-scale OCR technology, so that the character recognition and comparison precision is remarkably improved; the automatic extraction type face recognition model is utilized to complete recognition of the identity of the account holder, so that the account holder does not need to spend extra time for collecting face images, and the accuracy of face recognition is ensured. The invention can greatly improve the commodity and face recognition precision, thereby ensuring the precision and effectiveness of automatic checkout.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used 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 for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
FIG. 1 is a flowchart of an automated intelligent campus checkout method according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating similarity detection in an automated intelligent campus settlement method according to an embodiment of the present invention;
FIG. 3 is a flow chart of text recognition in an automated intelligent campus checkout method according to an embodiment of the present invention;
FIG. 4 is a schematic block diagram of an automated checkout system for a smart campus based on artificial intelligence in accordance with an embodiment of the present invention;
fig. 5 is a block diagram of an electronic device according to an embodiment of the present invention.
Description of reference numerals: 100. a basic recording module; 200. an image optimization module; 300. an instance partitioning module; 400. a similarity detection module; 500. a character recognition module; 600. a money amount counting module; 700. an identity recognition module; 800. a checkout module; 101. a memory; 102. a processor; 103. a communication interface.
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.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The embodiment is as follows:
as shown in fig. 1 to fig. 3, in a first aspect, an embodiment of the present invention provides an artificial intelligence based smart campus automated checkout method, including the following steps:
s1, collecting and establishing a commodity template image library according to the images of the commodities sold on the campus, and recording the price information of each corresponding commodity in the commodity template image library; and for each commodity sold in the campus, shooting a template image of each commodity sold in the campus, storing the template image into the system to obtain a comprehensive commodity template image library, and recording the price of each commodity in the system.
S2, acquiring an image of the commodity to be checked out, and optimizing the image of the commodity to be checked out by using an image enhancement method to obtain an image of the commodity to be detected; a campus teacher can select a plurality of commodities in the shopping process, and in the account settling process of the teacher and the student, an account settling system directly shoots pictures of the commodities which are put together to serve as images of the commodities to be settled; and carrying out depth optimization on each commodity image to be settled by utilizing an image enhancement technology to obtain a clearer image.
S3, performing example segmentation on the commodity image to be detected by adopting an example segmentation method to obtain a plurality of commodity segmentation area images to be detected; the example segmentation method comprises one or more of a Mask-RCNN algorithm, a PANET example segmentation model, a YOLACT model, a Polarmask algorithm and a CenterMask example segmentation algorithm. The example segmentation method adopted by the invention is the conventional image example segmentation method, and is not described herein again. And (3) carrying out example segmentation on each commodity image to be detected to obtain a plurality of commodity segmentation area images to be detected (each commodity corresponds to one commodity segmentation area image to be detected).
S4, respectively calculating the similarity of each to-be-detected commodity segmentation area image and all commodity images in a commodity template image library by using a sparse matching similarity detection method based on an image pyramid so as to obtain a primary commodity template image corresponding to each to-be-detected commodity segmentation area image through screening;
further, as shown in fig. 2, the method includes:
s41, respectively carrying out sparse coding on each commodity image in the commodity segmentation area image to be detected and each commodity image in the commodity template image library to obtain the corresponding commodity image code to be detected and the corresponding template image code;
s42, calculating the similarity between each commodity image to be detected and each commodity image in the commodity template image library according to each commodity image code to be detected and each template image code, so as to determine a first commodity template image corresponding to each commodity image to be detected;
s43, respectively carrying out multi-equal division processing on each to-be-detected commodity segmentation area image and the corresponding first commodity template image to obtain to-be-detected commodity images and first commodity template area images of a plurality of corresponding equal division areas;
s44, respectively carrying out sparse coding on the commodity image to be detected and the first commodity template area image of each equally divided area to obtain the commodity image code to be detected and the first template image code of a plurality of corresponding equally divided areas;
and S45, calculating the similarity between each to-be-detected commodity segmentation area image and the corresponding first commodity template image according to the to-be-detected commodity image code and each template image code of each equally divided area so as to determine the primary commodity template image corresponding to each to-be-detected commodity segmentation area image.
Further, counting and sorting the corresponding preliminary commodity template images according to the number of similarity equal-dividing areas in the preliminary commodity template images corresponding to the commodity segmentation area images to be detected so as to obtain the optimal preliminary commodity template image through screening.
In some embodiments of the invention, for a specific commodity segmentation area image to be detected, the similarity between the specific commodity segmentation area image and all template commodity images is calculated by using a sparse matching similarity detection method based on an image pyramid, and the template commodity image with the highest similarity, namely a primary commodity template image, is found.
Specifically, the sparse matching similarity detection method based on the image pyramid is as follows:
(a) and carrying out sparse coding on a specific commodity segmentation area image to be detected and a specific template commodity image, and calculating the similarity between the specific commodity segmentation area image and the specific template commodity image by using the Euclidean distance. If the similarity is low, directly determining that the two images are not similar; if the similarity is higher, the next step is continued.
(b) And (3) dividing the commodity segmentation region image to be detected and the template commodity image into 4 equal parts, respectively carrying out sparse coding on each equal divided region, and respectively calculating the similarity of 4 corresponding regions (the upper left corresponds to the upper left, and the lower right corresponds to the lower right, and the like). If the similarity degree of the 4 areas is higher than 3, directly determining that the two images are not similar; if the similarity in the 4 areas is higher than or equal to 3, the next step is continued.
(c) And (3) dividing the to-be-detected commodity segmentation region image and the template commodity image into 16 equal parts, respectively carrying out sparse coding on each equal-divided region, and respectively calculating the similarity of 16 corresponding regions (the upper left corresponds to the upper left, and the lower right corresponds to the lower right, and the like). If the similarity degree of the 16 areas is higher and is lower than 12, directly determining that the two images are not similar; and if the similarity of the segmented area image of the commodity to be detected is higher than 12 in the 16 areas, determining that the similarity of the segmented area image of the commodity to be detected and the template commodity image is higher.
And calculating the similarity between the image of the specific commodity segmentation area to be detected and all the template commodity images by utilizing the steps. And (c) sequencing all template commodity images which are determined to have higher similarity with the segmented area images of the commodity to be detected, determining the template commodity image which has the most similar area (for example, 15 areas are similar in 16 areas) with the segmented area image of the commodity to be detected in the step (c) as the template commodity image which has the highest similarity with the segmented area images of the commodity to be detected, and calling the template commodity image as a primary commodity template image.
S5, respectively carrying out character recognition on the preliminary commodity template images corresponding to the to-be-detected commodity segmentation area images by adopting a multi-scale OCR recognition method to obtain and determine target to-be-settled commodities according to character recognition results of the to-be-detected commodity segmentation area images and character recognition results of the preliminary commodity templates under various scales;
further, as shown in fig. 3, the method includes:
s51, carrying out multi-scale division on the primary commodity template images corresponding to the commodity segmentation region images to be detected to obtain corresponding primary commodity template images under multiple scales;
and S52, performing character recognition on the preliminary commodity template image under each scale by adopting an OCR recognition method.
Further, counting character recognition results of the segmented region images of the to-be-detected commodities under each scale to determine a first recognition result; counting character recognition results of the primary commodity template under each scale to determine a second recognition result; and determining the target commodity to be checked out according to the first recognition result and the second recognition result.
In some embodiments of the invention, the segmented region image of the to-be-detected commodity and the commodity image of the preliminary matching template are subjected to character recognition by utilizing a multi-scale OCR technology, and recognized character results are compared. And if the character similarity is very high and is larger than a preset similarity matching threshold, finally identifying the same commodity, determining the target commodity to be settled, and obtaining the price of the target commodity to be settled in the segmented area image of the commodity to be detected. Character recognition is accomplished using OCR technology at a variety of image scales. If the recognition results of the characters under multiple scales are inconsistent, a minority of the characters are used for complying with the principle of majority, for example, a certain character is recognized as 'king' under the most scales, a certain character is recognized as 'jade' under the least scales, and finally the character is recognized as 'king'.
S6, obtaining and counting the price information of each target commodity to be settled to obtain the total consumption amount; the method is used for identifying the images of the segmented areas of all the commodities to be detected, obtaining the prices of all the commodities and calculating the total consumption price.
S7, acquiring and importing the target face image into a preset face recognition model to generate an identity recognition result;
further, still include: acquiring a plurality of face images of a target person, and processing the plurality of face images by using a peak signal-to-noise ratio detection method based on super-resolution reconstruction so as to obtain an initial target face image by screening; and comparing and matching the initial target face image with a preset standard face template image by using a multi-region accurate matching verification method so as to obtain a target face image by screening.
In some embodiments of the present invention, in order to further improve the accuracy of face identification, the face image is preprocessed and screened to obtain a more accurate image. In the checkout process, a checkout system directly extracts a plurality of face images (5-10), and the high-noise face images are directly removed by using a peak signal-to-noise ratio detection method based on super-resolution reconstruction. Specifically, super-resolution reconstruction is respectively carried out on a plurality of face images, the peak signal-to-noise ratio of each face image is detected on the basis, and the face images with lower peak signal-to-noise ratios are directly deleted. And comparing the plurality of face images with the standard face template image respectively by using a multi-region accurate matching verification method (one standard face template image is needed, and a front face and no shielding are needed). Specifically, all face images and standard face template images are respectively divided into 4 equal parts. Any one face image is matched with the standard face template image in 4 areas, and finally, one face image with the highest matching degree with the standard face template image is reserved and used as a face image to be recognized, namely a target face image, so that the final face recognition is carried out.
And S8, extracting a corresponding payment account from the corresponding smart campus payment system according to the identity recognition result, and deducting money from the corresponding payment account according to the total consumption amount to finish automatic payment.
In order to solve the problem that errors exist in automatic checkout due to the fact that the problems of commodity misidentification, identity misidentification and the like exist in the prior art, the method and the device perform deep optimization and commodity example segmentation processing on a plurality of commodity images to be detected by means of an image enhancement technology and an example segmentation technology, and provide effective support for accurate identification of each commodity; and the similarity between the image of the commodity segmentation area to be detected and the template commodity image is calculated by using a sparse matching similarity detection method based on the image pyramid, so that the template commodity image with the highest similarity with the image of the commodity segmentation area to be detected, namely a primary commodity template image, can be more accurately retrieved. Meanwhile, character recognition is carried out on the segmented region image of the commodity to be detected and the primary commodity template image by utilizing a multi-scale OCR technology, so that the character recognition and comparison precision is remarkably improved; the automatic extraction type face recognition model is utilized to complete the recognition of the identity of the account holder, so that the account holder does not need to spend extra time for collecting face images, and the accuracy of face recognition is ensured. The invention can greatly improve the commodity and face recognition precision, thereby ensuring the precision and effectiveness of automatic checkout.
As shown in fig. 4, in a second aspect, an embodiment of the present invention provides an artificial intelligence based smart campus automated checkout system, which includes a basic record module 100, an image optimization module 200, an example segmentation module 300, a similarity detection module 400, a text recognition module 500, a money amount statistics module 600, an identity recognition module 700, and a checkout module 800, wherein:
the basic recording module 100 is used for acquiring and establishing a commodity template image library according to the images of the commodities sold on the campus, and recording the price information of each corresponding commodity in the commodity template image library;
the image optimization module 200 is configured to obtain an image of a commodity to be settled, and optimize the image of the commodity to be settled by using an image enhancement method to obtain an image of the commodity to be detected;
the example segmentation module 300 is configured to perform example segmentation on the to-be-detected commodity image by using an example segmentation method to obtain a plurality of to-be-detected commodity segmentation region images;
the similarity detection module 400 is configured to calculate similarities of all the commodity images in each to-be-detected commodity segmentation area image and the commodity template image library respectively by using a sparse matching similarity detection method based on an image pyramid, so as to obtain a preliminary commodity template image corresponding to each to-be-detected commodity segmentation area image through screening;
the character recognition module 500 is configured to perform character recognition on the preliminary commodity template images corresponding to the segmented area images of the commodities to be detected by using a multi-scale OCR recognition method, so as to obtain and determine a target commodity to be settled according to character recognition results of the segmented area images of the commodities to be detected and character recognition results of the preliminary commodity templates under various scales;
the amount counting module 600 is configured to obtain and count price information of each target commodity to be settled, so as to obtain a total consumption amount;
the identity recognition module 700 is configured to obtain and import a target face image into a preset face recognition model, and generate an identity recognition result;
and the payment module 800 is configured to extract a corresponding payment account from the corresponding smart campus payment system according to the identity recognition result, deduct money from the corresponding payment account according to the total consumption amount, and complete automatic payment.
In order to solve the problem that errors exist in automatic checkout due to the fact that the problems of commodity misidentification, identity misidentification and the like exist in the prior art, the system carries out deep optimization and commodity example segmentation processing on a plurality of commodity images to be detected through the cooperation of various modules such as a basic recording module 100, an image optimization module 200, an example segmentation module 300, a similarity detection module 400, a character recognition module 500, an amount of money statistics module 600, an identity recognition module 700 and a checkout module 800 by utilizing an image enhancement technology and an example segmentation technology, and provides effective support for accurate identification of each commodity; and the similarity between the image of the commodity segmentation area to be detected and the template commodity image is calculated by using a sparse matching similarity detection method based on the image pyramid, so that the template commodity image with the highest similarity with the image of the commodity segmentation area to be detected, namely a primary commodity template image, can be more accurately retrieved. Meanwhile, character recognition is carried out on the segmented region image of the commodity to be detected and the primary commodity template image by utilizing a multi-scale OCR technology, so that the character recognition and comparison precision is remarkably improved; the automatic extraction type face recognition model is utilized to complete the recognition of the identity of the account holder, so that the account holder does not need to spend extra time for collecting face images, and the accuracy of face recognition is ensured. The invention can greatly improve the commodity and face recognition precision, thereby ensuring the precision and effectiveness of automatic checkout.
As shown in fig. 5, in a third aspect, an embodiment of the present application provides an electronic device, which includes a memory 101 for storing one or more programs; a processor 102. The one or more programs, when executed by the processor 102, implement the method of any of the first aspects as described above.
Also included is a communication interface 103, and the memory 101, processor 102 and communication interface 103 are electrically connected to each other, directly or indirectly, to enable transfer or interaction of data. For example, the components may be electrically connected to each other via one or more communication buses or signal lines. The memory 101 may be used to store software programs and modules, and the processor 102 executes the software programs and modules stored in the memory 101 to thereby execute various functional applications and data processing. The communication interface 103 may be used for communicating signaling or data with other node devices.
The Memory 101 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like.
The processor 102 may be an integrated circuit chip having signal processing capabilities. The Processor 102 may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but also Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components.
In the embodiments provided in the present application, it should be understood that the disclosed method and system and method can be implemented in other ways. The method and system embodiments described above are merely illustrative, for example, the flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of methods and systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
In a fourth aspect, embodiments of the present application provide a computer-readable storage medium, on which a computer program is stored, which, when executed by the processor 102, implements the method according to any one of the first aspect described above. The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes will occur to 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.
It will be evident to those skilled in the art that the present application is not limited to the details of the foregoing illustrative embodiments, and that the present application may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Claims (10)
1. An automatic intelligent campus account settlement method based on artificial intelligence is characterized by comprising the following steps:
acquiring and establishing a commodity template image library according to the images of the commodities sold in the campus, and recording price information of each corresponding commodity in the commodity template image library;
acquiring an image of a commodity to be settled, and optimizing the image of the commodity to be settled by using an image enhancement method to obtain an image of the commodity to be detected;
carrying out example segmentation on the commodity image to be detected by adopting an example segmentation method to obtain a plurality of commodity segmentation area images to be detected;
respectively calculating the similarity of each to-be-detected commodity segmentation area image and all commodity images in a commodity template image library by using a sparse matching similarity detection method based on an image pyramid so as to obtain a primary commodity template image corresponding to each to-be-detected commodity segmentation area image through screening;
respectively carrying out character recognition on the preliminary commodity template images corresponding to the to-be-detected commodity segmentation area images by adopting a multi-scale OCR recognition method so as to obtain and determine a target to-be-checked commodity according to the character recognition results of the to-be-detected commodity segmentation area images and the character recognition results of the preliminary commodity templates under each scale;
acquiring and counting price information of each target commodity to be settled to obtain total consumption amount;
acquiring and importing a target face image into a preset face recognition model to generate an identity recognition result;
and extracting a corresponding checkout account from the corresponding smart campus checkout system according to the identity recognition result, and deducting money from the corresponding checkout account according to the total consumption amount to finish automatic checkout.
2. The intelligent campus automatic checkout method based on artificial intelligence as claimed in claim 1, wherein the method for calculating the similarity of each segmented region image of the to-be-detected commodity and all commodity images in the commodity template image library by using the sparse matching similarity detection method based on image pyramid comprises the following steps:
respectively carrying out sparse coding on each commodity image in each commodity segmentation area image to be detected and each commodity image in the commodity template image library to obtain a corresponding commodity image code to be detected and a corresponding template image code;
calculating the similarity between each to-be-detected commodity segmentation area image and each commodity image in the commodity template image library according to each to-be-detected commodity image code and each template image code so as to determine a first commodity template image corresponding to each to-be-detected commodity segmentation area image;
respectively carrying out multi-equal division processing on each to-be-detected commodity segmentation area image and the corresponding first commodity template image to obtain the to-be-detected commodity image and the first commodity template area image of the corresponding multiple equal division areas;
respectively carrying out sparse coding on the commodity image to be detected and the first commodity template area image of each equally divided area to obtain corresponding commodity image codes to be detected and first template image codes of a plurality of equally divided areas;
and calculating the similarity between each to-be-detected commodity segmentation area image and the corresponding first commodity template image according to the to-be-detected commodity image code and each template image code of each equally divided area so as to determine a primary commodity template image corresponding to each to-be-detected commodity segmentation area image.
3. The intelligent campus automatic checkout method based on artificial intelligence as claimed in claim 2, wherein the method for obtaining the preliminary commodity template image corresponding to each segmented region image of the to-be-detected commodity comprises the following steps:
and counting and sequencing the corresponding preliminary commodity template images according to the number of similarity equal division areas in the preliminary commodity template images corresponding to the commodity division area images to be detected so as to screen and obtain the optimal preliminary commodity template image.
4. The intelligent campus automatic checkout method based on artificial intelligence as claimed in claim 1, wherein the method for respectively performing character recognition on the preliminary commodity template images corresponding to the segmented region images of each to-be-detected commodity by using a multi-scale OCR recognition method comprises the following steps:
carrying out multi-scale division on the preliminary commodity template image corresponding to each commodity segmentation area image to be detected so as to obtain the corresponding preliminary commodity template image under multiple scales;
and performing character recognition on the primary commodity template image under each scale by adopting an OCR recognition method.
5. The intelligent campus automatic checkout method based on artificial intelligence as claimed in claim 1, wherein the method for determining the target merchandise to be checked out according to the character recognition result of each segmented region image of the merchandise to be detected and the character recognition result of the preliminary merchandise template under each scale comprises the following steps:
counting character recognition results of each to-be-detected commodity segmentation area image under each scale to determine a first recognition result;
counting character recognition results of the primary commodity template under each scale to determine a second recognition result;
and determining the target commodity to be checked out according to the first recognition result and the second recognition result.
6. The smart campus automated checkout method based on artificial intelligence of claim 1 further comprising the steps of:
acquiring a plurality of face images of a target person, and processing the plurality of face images by using a peak signal-to-noise ratio detection method based on super-resolution reconstruction so as to obtain an initial target face image by screening;
and comparing and matching the initial target face image with a preset standard face template image by using a multi-region accurate matching verification method so as to obtain a target face image by screening.
7. The intelligent campus automation checkout method based on artificial intelligence of claim 1, wherein the example segmentation method comprises one or more of Mask-RCNN algorithm, PANet example segmentation model, yoact model, polarmack algorithm, and CenterMask example segmentation algorithm.
8. The utility model provides a wisdom campus automatic checkout system based on artificial intelligence, its characterized in that, includes basic record module, image optimization module, example segmentation module, similarity detection module, word recognition module, amount of money statistics module, identity recognition module and the module of settling accounts, wherein:
the basic recording module is used for acquiring and establishing a commodity template image library according to the images of the commodities sold in the campus, and recording the price information of each corresponding commodity in the commodity template image library;
the image optimization module is used for acquiring an image of the commodity to be settled, and optimizing the image of the commodity to be settled by using an image enhancement method to obtain an image of the commodity to be detected;
the example segmentation module is used for carrying out example segmentation on the commodity image to be detected by adopting an example segmentation method so as to obtain a plurality of commodity segmentation area images to be detected;
the similarity detection module is used for respectively calculating the similarity of each to-be-detected commodity segmentation area image and all commodity images in the commodity template image library by using a sparse matching similarity detection method based on an image pyramid so as to obtain a primary commodity template image corresponding to each to-be-detected commodity segmentation area image through screening;
the character recognition module is used for respectively carrying out character recognition on the preliminary commodity template images corresponding to the to-be-detected commodity segmentation area images by adopting a multi-scale OCR recognition method so as to obtain and determine a target to-be-settled commodity according to the character recognition results of the to-be-detected commodity segmentation area images and the character recognition results of the preliminary commodity templates under each scale;
the amount counting module is used for acquiring and counting price information of each target commodity to be settled, so as to obtain total consumption amount;
the identity recognition module is used for acquiring and importing a target face image into a preset face recognition model to generate an identity recognition result;
and the payment module is used for extracting a corresponding payment account from the corresponding smart campus payment system according to the identity recognition result, deducting money from the corresponding payment account according to the total consumption amount, and completing automatic payment.
9. An electronic device, comprising:
a memory for storing one or more programs;
a processor;
the one or more programs, when executed by the processor, implement the method of any of claims 1-7.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210838618.XA CN115018492A (en) | 2022-07-18 | 2022-07-18 | Smart campus automatic checkout method and system based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210838618.XA CN115018492A (en) | 2022-07-18 | 2022-07-18 | Smart campus automatic checkout method and system based on artificial intelligence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115018492A true CN115018492A (en) | 2022-09-06 |
Family
ID=83080489
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210838618.XA Pending CN115018492A (en) | 2022-07-18 | 2022-07-18 | Smart campus automatic checkout method and system based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115018492A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115205950A (en) * | 2022-09-16 | 2022-10-18 | 北京吉道尔科技有限公司 | Intelligent traffic subway passenger detection and settlement method and system based on block chain |
| CN115661784A (en) * | 2022-10-12 | 2023-01-31 | 北京惠朗时代科技有限公司 | Intelligent traffic-oriented traffic sign image big data identification method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106781121A (en) * | 2016-12-14 | 2017-05-31 | 朱明� | The supermarket self-checkout intelligence system of view-based access control model analysis |
| CN108345912A (en) * | 2018-04-25 | 2018-07-31 | 电子科技大学中山学院 | Commodity rapid settlement system based on RGBD information and deep learning |
| CN111027554A (en) * | 2019-12-27 | 2020-04-17 | 创新奇智(重庆)科技有限公司 | System and method for accurately detecting and positioning commodity price tag characters |
| CN111524150A (en) * | 2020-07-03 | 2020-08-11 | 支付宝(杭州)信息技术有限公司 | Image processing method and device |
-
2022
- 2022-07-18 CN CN202210838618.XA patent/CN115018492A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106781121A (en) * | 2016-12-14 | 2017-05-31 | 朱明� | The supermarket self-checkout intelligence system of view-based access control model analysis |
| CN108345912A (en) * | 2018-04-25 | 2018-07-31 | 电子科技大学中山学院 | Commodity rapid settlement system based on RGBD information and deep learning |
| CN111027554A (en) * | 2019-12-27 | 2020-04-17 | 创新奇智(重庆)科技有限公司 | System and method for accurately detecting and positioning commodity price tag characters |
| CN111524150A (en) * | 2020-07-03 | 2020-08-11 | 支付宝(杭州)信息技术有限公司 | Image processing method and device |
Non-Patent Citations (3)
| Title |
|---|
| 倪尧天: "基于内容的商品图像分类算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 李秀利: "基于深度学习的无人超市商品图像检测识别方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 肖亮等: "《基于图像先验建模的超分辨增强理论与算法 变分PDE、稀疏正则化与贝叶斯方法》", 北京国防工业出版社 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115205950A (en) * | 2022-09-16 | 2022-10-18 | 北京吉道尔科技有限公司 | Intelligent traffic subway passenger detection and settlement method and system based on block chain |
| CN115661784A (en) * | 2022-10-12 | 2023-01-31 | 北京惠朗时代科技有限公司 | Intelligent traffic-oriented traffic sign image big data identification method and system |
| CN115661784B (en) * | 2022-10-12 | 2023-08-22 | 北京惠朗时代科技有限公司 | Intelligent traffic-oriented traffic sign image big data identification method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115018492A (en) | Smart campus automatic checkout method and system based on artificial intelligence | |
| RU2695489C1 (en) | Identification of fields on an image using artificial intelligence | |
| KR20150059793A (en) | Detecting embossed characters on cards | |
| CN114581207B (en) | Commodity image big data accurate pushing method and system for E-commerce platform | |
| US11132576B2 (en) | Text recognition method and apparatus, electronic device, and storage medium | |
| CN112395996A (en) | Financial bill OCR recognition and image processing method, system and readable storage medium | |
| CN108717744B (en) | Method and device for identifying seal serial number on financial document and terminal equipment | |
| CN113963147B (en) | Key information extraction method and system based on semantic segmentation | |
| CN115019374B (en) | Intelligent classroom student concentration degree low-consumption detection method and system based on artificial intelligence | |
| CN115147253A (en) | Block chain-based smart campus book big data borrowing management method and system | |
| CN113313217B (en) | Method and system for accurately identifying dip angle characters based on robust template | |
| CN115100640A (en) | Artificial intelligence-based intelligent supermarket commodity sales big data detection method and system | |
| CN115600249A (en) | Meta-universe e-commerce shopping big data security protection method and system based on block chain | |
| CN116739704A (en) | E-commerce platform interest analysis type commodity recommendation method and system based on artificial intelligence | |
| CN111199240A (en) | Training method of bank card identification model, and bank card identification method and device | |
| CN115661833A (en) | Target commodity big data accurate identification method and system based on simple photographing | |
| CN114723536B (en) | E-commerce platform cheap commodity selection method and system based on image big data comparison | |
| CN111428725A (en) | Data structuring processing method and device and electronic equipment | |
| CN116562991A (en) | Commodity big data information identification method and system for meta-space electronic commerce platform | |
| CN115860860A (en) | Smart community choice commodity big data group purchase method and system based on artificial intelligence | |
| CN115965468A (en) | Transaction data-based abnormal behavior detection method, device, equipment and medium | |
| CN115713630A (en) | Low-quality seal image big data identification method and system based on artificial intelligence | |
| CN114925239A (en) | Intelligent education target video big data retrieval method and system based on artificial intelligence | |
| CN114299509A (en) | Method, device, equipment and medium for acquiring information | |
| CN113239031A (en) | Big data denoising processing method |
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 |