CN116186318A - Image retrieval method, device and storage medium based on association rule learning - Google Patents
Image retrieval method, device and storage medium based on association rule learning Download PDFInfo
- Publication number
- CN116186318A CN116186318A CN202310446006.0A CN202310446006A CN116186318A CN 116186318 A CN116186318 A CN 116186318A CN 202310446006 A CN202310446006 A CN 202310446006A CN 116186318 A CN116186318 A CN 116186318A
- Authority
- CN
- China
- Prior art keywords
- image
- images
- association rule
- rule learning
- retrieved
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 45
- 230000015654 memory Effects 0.000 claims description 26
- 230000004044 response Effects 0.000 claims description 17
- 238000000605 extraction Methods 0.000 claims description 16
- 238000012549 training Methods 0.000 claims description 11
- 230000006870 function Effects 0.000 claims description 8
- 238000013507 mapping Methods 0.000 claims description 6
- 238000011176 pooling Methods 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 5
- 238000005065 mining Methods 0.000 abstract description 5
- 238000004364 calculation method Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 238000007418 data mining Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 239000007787 solid Substances 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 235000008429 bread Nutrition 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 235000013336 milk Nutrition 0.000 description 2
- 239000008267 milk Substances 0.000 description 2
- 210000004080 milk Anatomy 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
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
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- 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)
- Library & Information 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)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an image retrieval method, an image retrieval device and a storage medium based on association rule learning, wherein images to be retrieved and images in a retrieval database are obtained; extracting local correlation characteristics of the images to be searched and the images in the search database based on an association rule learning algorithm; and carrying out feature matching on the local correlation characteristics of the images to be searched and the images in the search database, and determining the search results of the same subcategory as the images to be searched in the search database. Therefore, the image retrieval method is used for mining the associated information in the image through the association rule learning algorithm, and fine-grained feature representation is enhanced, so that the discrimination and the discrimination among different subclasses can be enhanced.
Description
Technical Field
The invention relates to the technical field of image retrieval, in particular to an image retrieval method, an image retrieval device and a storage medium based on association rule learning.
Background
The grid image search aims at a user inputting a query object to a computer, and the computer returns a search result which belongs to the same subcategory as the query object. With the increasing rise of deep learning and artificial intelligence, the power grid image retrieval has wide development and application prospect. For example, the violations of different subclasses can be accurately retrieved, thereby helping the salesman to promote the automation and the accuracy of the electric power operation.
Currently, the challenges faced by grid image retrieval: the difference between different subclasses is smaller, and the variance in the same subclass is larger. This is exactly contrary to what we want to have "big inter-subclass variance, small intra-subclass variance". Thus, how to enhance the differentiation and discrimination between different subcategories is a problem that is currently in need of resolution.
Disclosure of Invention
In view of the above, the embodiments of the present invention provide an image retrieval method, apparatus and storage medium based on association rule learning, so as to solve the technical problem in the prior art that the difference between different subclasses is small when performing grid image retrieval.
The technical scheme provided by the invention is as follows:
the first aspect of the embodiment of the invention provides an image retrieval method based on association rule learning, which comprises the following steps: acquiring an image to be searched and an image in a search database; extracting local correlation characteristics of the images to be searched and the images in the search database based on an association rule learning algorithm; and carrying out feature matching on the local correlation characteristics of the images to be searched and the images in the search database, and determining the search results of the same subcategory as the images to be searched in the search database.
Optionally, extracting the local relevance features of the image to be retrieved and the image in the retrieval database based on the association rule learning algorithm includes: extracting local correlation characteristics of the image to be retrieved based on a correlation rule learning algorithm; and extracting local correlation characteristics of each image in the retrieval database based on an association rule learning algorithm.
Optionally, extracting the local relevance feature of the image to be retrieved based on the association rule learning algorithm includes: extracting a feature map of an image to be retrieved and activated pixels in each feature map; and calculating local correlation characteristics of the image to be searched by adopting an association rule learning algorithm based on the feature map and the activated pixels in each feature map.
Optionally, extracting the feature map of the image to be retrieved and the activated pixels in each feature map includes: mapping the image to be retrieved into a group of feature images by adopting a pre-training model; calculating the average value of the response values of all the pixel points in each feature map; and comparing the response values of all the pixel points in each feature map with the corresponding average value, and taking the pixel points with the response values larger than the corresponding average value as activated pixels in the feature map.
Optionally, calculating local correlation features of the image to be retrieved by adopting an association rule learning algorithm based on the feature map and the activated pixels in each feature map, including: generating a feature map set and an activated pixel set based on the feature map and the activated pixels in each feature map; calculating a frequent mode of the image to be retrieved according to the feature map set and the activated pixel set; and extracting the features in the frequent mode by adopting pooling operation as local correlation features of the images to be retrieved.
Optionally, performing feature matching on local correlation features of the image to be retrieved and the image in the retrieval database, and determining a retrieval result of the same sub-category in the retrieval database and the image to be retrieved, including: calculating the similarity of the local correlation characteristics of the images to be searched and the local correlation characteristics of each image in the search database to obtain a plurality of similarities; and sequencing the multiple similarities, and extracting images corresponding to the preset number of similarities as retrieval results belonging to the same subcategory as the images to be retrieved.
Optionally, the similarity is calculated using a cosine similarity function.
A second aspect of an embodiment of the present invention provides an image retrieval device based on association rule learning, including: the image acquisition module is used for acquiring the image to be searched and the image in the search database; the feature extraction module is used for extracting local correlation features of the images to be searched and the images in the search database based on the association rule learning algorithm; the feature matching module is used for carrying out feature matching on the local correlation features of the images to be searched and the images in the search database, and determining the search results of the same subcategory as the images to be searched in the search database.
Optionally, the feature extraction module includes: the first extraction module is used for extracting local correlation characteristics of the image to be retrieved based on an association rule learning algorithm; and the second extraction module is used for extracting the local correlation characteristic of each image in the retrieval database based on the association rule learning algorithm.
Optionally, the first extraction module includes: the extraction module is used for extracting the feature images of the images to be retrieved and the activated pixels in each feature image; and the feature calculation module is used for calculating local correlation features of the images to be retrieved by adopting an association rule learning algorithm based on the feature graphs and the activated pixels in each feature graph.
Optionally, the extraction module is specifically configured to: mapping the image to be retrieved into a group of feature images by adopting a pre-training model; calculating the average value of the response values of all the pixel points in each feature map; and comparing the response values of all the pixel points in each feature map with the corresponding average value, and taking the pixel points with the response values larger than the corresponding average value as activated pixels in the feature map.
Optionally, the feature calculation module is specifically configured to: generating a feature map set and an activated pixel set based on the feature map and the activated pixels in each feature map; calculating a frequent mode of the image to be retrieved according to the feature map set and the activated pixel set; and extracting the features in the frequent mode by adopting pooling operation as local correlation features of the images to be retrieved.
Optionally, the feature matching module is specifically configured to: calculating the similarity of the local correlation characteristics of the images to be searched and the local correlation characteristics of each image in the search database to obtain a plurality of similarities; and sequencing the multiple similarities, and extracting images corresponding to the preset number of similarities as retrieval results belonging to the same subcategory as the images to be retrieved.
Optionally, the similarity is calculated using a cosine similarity function.
A third aspect of the embodiment of the present invention provides a computer-readable storage medium, where computer instructions are stored, where the computer instructions are configured to cause the computer to perform the image retrieval method based on association rule learning according to any one of the first aspect and the first aspect of the embodiment of the present invention.
A fourth aspect of an embodiment of the present invention provides an electronic device, including: the image retrieval method based on association rule learning according to the first aspect of the embodiment of the invention is characterized by comprising a memory and a processor, wherein the memory and the processor are in communication connection, the memory stores computer instructions, and the processor executes the computer instructions, so that the image retrieval method based on association rule learning according to any one of the first aspect and the first aspect of the embodiment of the invention is executed.
The technical scheme provided by the invention has the following effects:
the image retrieval method, the device and the storage medium based on association rule learning provided by the embodiment of the invention are characterized in that images to be retrieved and images in a retrieval database are obtained; extracting local correlation characteristics of the images to be searched and the images in the search database based on an association rule learning algorithm; and carrying out feature matching on the local correlation characteristics of the images to be searched and the images in the search database, and determining the search results of the same subcategory as the images to be searched in the search database. Therefore, the image retrieval method is used for mining the associated information in the image through the association rule learning algorithm, and fine-grained feature representation is enhanced, so that the discrimination and the discrimination among different subclasses can be enhanced.
The image retrieval method based on association rule learning provided by the embodiment of the invention adopts the pre-training model to map the image to be retrieved into a group of feature images, namely adopts an unsupervised frame and the pre-training model, does not need manual labeling, and has good application value in electric power operation scenes.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of an image retrieval method based on association rule learning according to an embodiment of the present invention;
FIG. 2 is a block diagram of a local correlation feature learning in an image retrieval method based on association rule learning according to an embodiment of the present invention;
fig. 3 is a block diagram of a structure of an image retrieval apparatus based on association rule learning according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a computer-readable storage medium provided according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
The terms first, second, third, fourth and the like in the description and in the claims and in the above drawings are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
According to an embodiment of the present invention, there is provided an image retrieval method based on association rule learning, it being noted that the steps shown in the flowchart of the drawings may be performed in a computer system such as a set of computer-executable instructions, and that although a logical order is shown in the flowchart, in some cases the steps shown or described may be performed in an order different from that herein.
In this embodiment, an image retrieval method based on association rule learning is provided, which may be used in an electronic device, such as a computer, a mobile phone, a tablet computer, etc., fig. 1 is a flowchart of an image retrieval method based on association rule learning according to an embodiment of the present invention, and as shown in fig. 1, the method includes the following steps:
step S101: and acquiring the image to be searched and the image in the search database. Specifically, the image to be retrieved is a query object input by the user. For example, the user inputs an image to be searched to the computer, and the computer returns a search result which belongs to the same sub-category as the image to be searched through the image search method based on the association rule learning. When the image retrieval method is applied to the field of power grids, for example, the illegal behaviors of different subclasses need to be accurately retrieved, and the image to be retrieved is a power grid image. The search database is specifically a database queried by a computer, and a large amount of image data is stored in the database, for example, when the image to be searched is a power grid image, the image stored in the search database is also the power grid image, namely, when the power grid image is searched, the search query is performed from the image database of the power grid.
Step S102: and extracting local correlation characteristics of the images to be searched and the images in the search database based on the association rule learning algorithm. Specifically, the association rule learning algorithm is mainly used for acquiring internal association of the structure, and in this embodiment, the processing of the image by the association rule learning algorithm realizes that the computer vision task is converted into a data mining task, namely, the local correlation feature of the image is extracted.
Among them, the association rule learning algorithm is one of the most important and most deeply studied technologies in data mining, and it refers to mining frequently occurring patterns from a data set, where a pattern refers to a set of items, a sequence, a substructure, or the like. Association rule learning was originally initiated in shopping basket analysis scenarios: the supermarket operators further find the buying habit of customers by analyzing a large number of supermarket shopping tickets, and further formulate an effective trip strategy for supermarket marketing, including discount promotion, shelf placement and the like. For example, supermarket operators find that { milk and bread } always appear frequently through analysis, so that the shelves of milk and bread are placed as close as possible, customers can conveniently and simultaneously purchase two commodities in a short time, and customer satisfaction is improved. For another example, { micro-single camera, lens, memory card } is always frequently appeared, so that the micro-single camera, lens, memory card are often bundled and sold, thereby increasing sales and increasing profits.
The processing of the data by the association rule learning algorithm is realized according to the following flow: first define
Is a set of items, wherein ∈>
Representing the ith item. Definitions->
Is a transaction set, wherein->
Representing the ith transaction, and, at the same time,
also->
Is a subset of the set of (c). Given a set P, let ∈>
The support of P is defined as follows:
Where || represents the number of elements in the collection. When (when)
When P is the frequent pattern. Here, minsupp is a threshold value representing the minimum support.
Extracting local correlation characteristics of the images to be searched and the images in the search database respectively based on the association rule learning algorithm, namely extracting the local correlation characteristics of the images to be searched based on the association rule learning algorithm; and extracting local correlation characteristics of each image in the retrieval database based on an association rule learning algorithm.
Step S103: and carrying out feature matching on the local correlation characteristics of the images to be searched and the images in the search database, and determining the search results of the same subcategory as the images to be searched in the search database. And when the features are matched, respectively matching each image in the images to be searched and the images in the search database, and then determining the search results of the search database and the images to be searched belonging to the same subcategory according to the matching results.
According to the image retrieval method based on association rule learning, images to be retrieved and images in a retrieval database are obtained; extracting local correlation characteristics of the images to be searched and the images in the search database based on an association rule learning algorithm; and carrying out feature matching on the local correlation characteristics of the images to be searched and the images in the search database, and determining the search results of the same subcategory as the images to be searched in the search database. Therefore, the image retrieval method is used for mining the associated information in the image through the association rule learning algorithm, and fine-grained feature representation is enhanced, so that the discrimination and the discrimination among different subclasses can be enhanced.
In one embodiment, extracting local relevance features of an image to be retrieved based on an association rule learning algorithm includes: extracting a feature map of an image to be retrieved and activated pixels in each feature map; and calculating local correlation characteristics of the image to be searched by adopting an association rule learning algorithm based on the feature map and the activated pixels in each feature map. Specifically, when the association rule learning algorithm is adopted, the local correlation characteristics of the image to be searched are determined by extracting the characteristic images and the activated pixels of the image to be searched as a transaction set and a term set respectively and adopting a supporting degree calculation mode.
In an embodiment, extracting a feature map of an image to be retrieved and an activated pixel in each feature map includes: mapping the image to be retrieved into a group of feature images by adopting a pre-training model; calculating the average value of the response values of all the pixel points in each feature map; and comparing the response values of all the pixel points in each feature map with the corresponding average value, and taking the pixel points with the response values larger than the corresponding average value as activated pixels in the feature map. Specifically, as shown in fig. 2, when feature map extraction is performed, an image to be retrieved is input into a pre-training model, and the image to be retrieved is mapped into a set of feature maps through a convolution operation of the pre-training model, where the set of feature maps includes m pieces in total. And then calculating the average value of the response values of all the pixel points for each feature map as a threshold value. The feature maps are then traversed, for each feature map, traversing all pixel points that it contains, if the point response value is above a threshold, the point is considered to be activated, otherwise, not activated. The pixels that are to be activated are then left behind and the pixels that are not to be activated are removed.
In one embodiment, based on the feature map and the activated pixels in each feature map, a correlation rule learning algorithm is used to calculate local correlation features of the image to be retrieved, including the following steps:
step S201: a feature map set and an active pixel set are generated based on the feature map and the active pixels in each feature map. The method comprises the steps of extracting local correlation characteristics by adopting a correlation rule mining technology, and capturing the correlation among components. The most central step is how to transform computer vision tasks into data mining tasksAnd (5) carrying out business. Specifically, the feature map is translated into a transaction and the activated pixels are translated into terms. For example, if 6 activated pixels are included in the jth feature map, then from the perspective of data mining, it may be translated into 6 entries in the jth transaction, noted as
. Thus, as shown in fig. 2, a feature map set and an active pixel set, that is, a transaction set and a term set, are generated based on all feature maps obtained by convolutionally mapping an image to be retrieved and pixel points reserved in each feature map. Then the active pixel set is denoted +.>
Is an active pixel set, wherein +.>
Representing the ith activated pixel. />
Is a feature map set, wherein +.>
Representing the ith feature map, at the same time, < + >>
Also->
Is a subset of the set of (c).
Step S202: and calculating the frequent pattern of the image to be retrieved according to the characteristic image set and the activated pixel set. Specifically, as shown in FIG. 2, frequent pattern computation is performed by first giving a set
So that->
Calculating the set by using the above formula (1)
In this embodiment, m in the formula (1) is a constant representing the number of feature maps, and may be directly added->
. When->
When (I)>
Is a frequent pattern. Here, minsupp is a superparameter, depending on the distribution of the processed dataset.
Step S203: and extracting the features in the frequent mode by adopting pooling operation as local correlation features of the images to be retrieved. Specifically, after the frequent pattern is mined, the pooling operation is adopted to extract the features of the frequent pattern, and the obtained features are the local correlation features of the images to be searched
。/>
It should be noted that, in the above embodiment, a process of extracting the local correlation feature of the image to be searched by using the association rule learning algorithm is provided, and the process of extracting the local correlation feature of each image in the search database based on the association rule learning algorithm is the same as the process of extracting the local correlation feature of the image to be searched, which is not described herein.
In one embodiment, performing feature matching on local correlation features of an image to be retrieved and an image in a retrieval database, and determining a retrieval result in the retrieval database and in which the image to be retrieved belongs to the same subcategory, includes: calculating the similarity of the local correlation characteristics of the images to be searched and the local correlation characteristics of each image in the search database to obtain a plurality of similarities; and sequencing the multiple similarities, and extracting images corresponding to the preset number of similarities as retrieval results belonging to the same subcategory as the images to be retrieved.
Specifically, the similarity is calculated by using a cosine similarity function, and in other embodiments, the similarity can be calculated by using other calculation functions, which is not limited in the present invention. The similarity between the local correlation characteristic of the search image and the local correlation characteristic of each image in the search database is calculated by adopting a cosine similarity function and is realized by the following formula:
Wherein, the liquid crystal display device comprises a liquid crystal display device,
、/>
representing the local correlation characteristics of the image to be searched and each image in the search database respectively.
For example, when the search database includes N images, N similarity values are calculated according to the above formula (2), the N similarity values are sorted from large to small, and then the first K # -after sorting is selected
) The images in the search library corresponding to the similarity are the images most similar to the images to be searched, or the images belonging to the same sub-category with the images to be searched.
The image retrieval method based on association rule learning provided by the embodiment of the invention adopts the pre-training model to map the image to be retrieved into a group of feature images, namely adopts an unsupervised frame and the pre-training model, does not need manual labeling, and has good application value in electric power operation scenes.
The embodiment of the invention also provides an image retrieval device based on association rule learning, as shown in fig. 3, the device comprises:
the image acquisition module is used for acquiring the image to be searched and the image in the search database; the specific content refers to the corresponding parts of the above method embodiments, and will not be described herein.
The feature extraction module is used for extracting local correlation features of the images to be searched and the images in the search database based on the association rule learning algorithm; the specific content refers to the corresponding parts of the above method embodiments, and will not be described herein.
The feature matching module is used for carrying out feature matching on the local correlation features of the images to be searched and the images in the search database, and determining the search results of the same subcategory as the images to be searched in the search database. The specific content refers to the corresponding parts of the above method embodiments, and will not be described herein.
The image retrieval device based on association rule learning provided by the embodiment of the invention is used for acquiring the images to be retrieved and retrieving the images in the database; extracting local correlation characteristics of the images to be searched and the images in the search database based on an association rule learning algorithm; and carrying out feature matching on the local correlation characteristics of the images to be searched and the images in the search database, and determining the search results of the same subcategory as the images to be searched in the search database. Therefore, the image retrieval device mines the associated information in the image through the association rule learning algorithm, and fine-grained characteristic representation is enhanced, so that the discrimination and the discrimination among different subclasses can be enhanced.
The functional description of the image retrieval device based on association rule learning provided by the embodiment of the invention refers to the description of the image retrieval method based on association rule learning in the above embodiment in detail.
Optionally, the feature extraction module includes: the first extraction module is used for extracting local correlation characteristics of the image to be retrieved based on an association rule learning algorithm; and the second extraction module is used for extracting the local correlation characteristic of each image in the retrieval database based on the association rule learning algorithm.
Optionally, the first extraction module includes: the extraction module is used for extracting the feature images of the images to be retrieved and the activated pixels in each feature image; and the feature calculation module is used for calculating local correlation features of the images to be retrieved by adopting an association rule learning algorithm based on the feature graphs and the activated pixels in each feature graph.
Optionally, the extraction module is specifically configured to: mapping the image to be retrieved into a group of feature images by adopting a pre-training model; calculating the average value of the response values of all the pixel points in each feature map; and comparing the response values of all the pixel points in each feature map with the corresponding average value, and taking the pixel points with the response values larger than the corresponding average value as activated pixels in the feature map.
Optionally, the feature calculation module is specifically configured to: generating a feature map set and an activated pixel set based on the feature map and the activated pixels in each feature map; calculating a frequent mode of the image to be retrieved according to the feature map set and the activated pixel set; and extracting the features in the frequent mode by adopting pooling operation as local correlation features of the images to be retrieved.
Optionally, the feature matching module is specifically configured to: calculating the similarity of the local correlation characteristics of the images to be searched and the local correlation characteristics of each image in the search database to obtain a plurality of similarities; and sequencing the multiple similarities, and extracting images corresponding to the preset number of similarities as retrieval results belonging to the same subcategory as the images to be retrieved.
Optionally, the similarity is calculated using a cosine similarity function.
The embodiment of the present invention also provides a storage medium, as shown in fig. 4, on which a computer program 601 is stored, which when executed by a processor, implements the steps of the image retrieval method based on association rule learning in the above embodiment. The storage medium also stores audio and video stream data, characteristic frame data, interactive request signaling, encrypted data, preset data size and the like. The storage medium may be a magnetic Disk, an optical Disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a Flash Memory (Flash Memory), a Hard Disk (HDD), or a Solid State Drive (SSD); the storage medium may also comprise a combination of memories of the kind described above.
It will be appreciated by those skilled in the art that implementing all or part of the above-described embodiment method may be implemented by a computer program to instruct related hardware, where the program may be stored in a computer readable storage medium, and the program may include the above-described embodiment method when executed. Wherein the storage medium may be a magnetic Disk, an optical Disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a Flash Memory (Flash Memory), a Hard Disk (HDD), or a Solid State Drive (SSD); the storage medium may also comprise a combination of memories of the kind described above.
The embodiment of the present invention further provides an electronic device, as shown in fig. 5, where the electronic device may include a processor 51 and a memory 52, where the processor 51 and the memory 52 may be connected by a bus or other means, and in fig. 5, the connection is exemplified by a bus.
The processor 51 may be a central processing unit (Central Processing Unit, CPU). The processor 51 may also be other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or combinations thereof.
The memory 52 serves as a non-transitory computer readable storage medium that may be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as corresponding program instructions/modules in embodiments of the present invention. The processor 51 executes various functional applications of the processor and data processing by running non-transitory software programs, instructions, and modules stored in the memory 52, that is, implements the association rule learning-based image retrieval method in the above-described method embodiment.
The memory 52 may include a memory program area that may store an operating device, an application program required for at least one function, and a memory data area; the storage data area may store data created by the processor 51, etc. In addition, memory 52 may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, memory 52 may optionally include memory located remotely from processor 51, which may be connected to processor 51 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The one or more modules are stored in the memory 52, which when executed by the processor 51, perform the image retrieval method based on association rule learning in the embodiment shown in fig. 1-2.
The specific details of the electronic device may be understood correspondingly with reference to the corresponding related descriptions and effects in the embodiments shown in fig. 1 to 2, which are not repeated here.
Although embodiments of the present invention have been described in connection with the accompanying drawings, various modifications and variations may be made by those skilled in the art without departing from the spirit and scope of the invention, and such modifications and variations fall within the scope of the invention as defined by the appended claims.
Claims (10)
1. An image retrieval method based on association rule learning, comprising the following steps:
acquiring an image to be searched and an image in a search database;
extracting local correlation characteristics of the images to be searched and the images in the search database based on an association rule learning algorithm;
and carrying out feature matching on the local correlation characteristics of the images to be searched and the images in the search database, and determining the search results of the same subcategory as the images to be searched in the search database.
2. The image retrieval method based on association rule learning according to claim 1, wherein extracting local correlation features of an image to be retrieved and an image in a retrieval database based on an association rule learning algorithm comprises:
extracting local correlation characteristics of the image to be retrieved based on a correlation rule learning algorithm;
and extracting local correlation characteristics of each image in the retrieval database based on an association rule learning algorithm.
3. The image retrieval method based on association rule learning according to claim 2, wherein extracting local correlation features of an image to be retrieved based on an association rule learning algorithm comprises:
extracting a feature map of an image to be retrieved and activated pixels in each feature map;
and calculating local correlation characteristics of the image to be searched by adopting an association rule learning algorithm based on the feature map and the activated pixels in each feature map.
4. An image retrieval method based on association rule learning according to claim 3, wherein extracting feature images of images to be retrieved and activated pixels in each feature image comprises:
mapping the image to be retrieved into a group of feature images by adopting a pre-training model;
calculating the average value of the response values of all the pixel points in each feature map;
and comparing the response values of all the pixel points in each feature map with the corresponding average value, and taking the pixel points with the response values larger than the corresponding average value as activated pixels in the feature map.
5. The image retrieval method based on association rule learning as claimed in claim 3, wherein calculating local correlation characteristics of the image to be retrieved using an association rule learning algorithm based on the feature map and the activated pixels in each feature map comprises:
generating a feature map set and an activated pixel set based on the feature map and the activated pixels in each feature map;
calculating a frequent mode of the image to be retrieved according to the feature map set and the activated pixel set;
and extracting the features in the frequent mode by adopting pooling operation as local correlation features of the images to be retrieved.
6. The image retrieval method based on association rule learning according to claim 1, wherein feature matching is performed on local correlation features of an image to be retrieved and an image in a retrieval database, and a retrieval result in the retrieval database and the image to be retrieved belonging to the same subcategory is determined, comprising:
calculating the similarity of the local correlation characteristics of the images to be searched and the local correlation characteristics of each image in the search database to obtain a plurality of similarities;
and sequencing the multiple similarities, and extracting images corresponding to the preset number of similarities as retrieval results belonging to the same subcategory as the images to be retrieved.
7. The method for image retrieval based on association rule learning as claimed in claim 6, wherein the similarity is calculated using a cosine similarity function.
8. An image retrieval apparatus based on association rule learning, comprising:
the image acquisition module is used for acquiring the image to be searched and the image in the search database;
the feature extraction module is used for extracting local correlation features of the images to be searched and the images in the search database based on the association rule learning algorithm;
the feature matching module is used for carrying out feature matching on the local correlation features of the images to be searched and the images in the search database, and determining the search results of the same subcategory as the images to be searched in the search database.
9. A computer-readable storage medium storing computer instructions for causing the computer to execute the association rule learning-based image retrieval method according to any one of claims 1 to 7.
10. An electronic device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, the processor executing the computer instructions to perform the association rule learning-based image retrieval method as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310446006.0A CN116186318B (en) | 2023-04-24 | 2023-04-24 | Image retrieval method, device and storage medium based on association rule learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310446006.0A CN116186318B (en) | 2023-04-24 | 2023-04-24 | Image retrieval method, device and storage medium based on association rule learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116186318A true CN116186318A (en) | 2023-05-30 |
| CN116186318B CN116186318B (en) | 2023-07-14 |
Family
ID=86452447
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310446006.0A Active CN116186318B (en) | 2023-04-24 | 2023-04-24 | Image retrieval method, device and storage medium based on association rule learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116186318B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114581956A (en) * | 2022-03-17 | 2022-06-03 | 重庆邮电大学 | Multi-branch fine-grained feature fusion pedestrian re-identification method |
| CN115129915A (en) * | 2021-03-26 | 2022-09-30 | 北京新氧科技有限公司 | Repeated image retrieval method, device, equipment and storage medium |
| CN115311544A (en) * | 2022-08-18 | 2022-11-08 | 中国农业大学 | Underwater fish target detection method and device |
-
2023
- 2023-04-24 CN CN202310446006.0A patent/CN116186318B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115129915A (en) * | 2021-03-26 | 2022-09-30 | 北京新氧科技有限公司 | Repeated image retrieval method, device, equipment and storage medium |
| CN114581956A (en) * | 2022-03-17 | 2022-06-03 | 重庆邮电大学 | Multi-branch fine-grained feature fusion pedestrian re-identification method |
| CN115311544A (en) * | 2022-08-18 | 2022-11-08 | 中国农业大学 | Underwater fish target detection method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116186318B (en) | 2023-07-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101856120B1 (en) | Discovery of merchants from images | |
| CN109816482B (en) | Knowledge graph construction method, device and equipment of e-commerce platform and storage medium | |
| US9535930B2 (en) | System and method for using an image to provide search results | |
| CN103608826B (en) | Annotated using product in the video of Web information mining | |
| US10482146B2 (en) | Systems and methods for automatic customization of content filtering | |
| WO2011054002A2 (en) | Content-based image search | |
| CN111581423B (en) | Target retrieval method and device | |
| CN113987119A (en) | Data retrieval method, cross-modal data matching model processing method and device | |
| CN111582913A (en) | Advertisement recommendation method and device | |
| US9613283B2 (en) | System and method for using an image to provide search results | |
| CN110083731B (en) | Image retrieval method, device, computer equipment and storage medium | |
| CN115062200A (en) | User behavior mining method and system based on artificial intelligence | |
| CN111292153B (en) | Information recommendation method, device, system and storage medium | |
| CN111259274A (en) | Information processing method, device, equipment and information display device | |
| EP3564833B1 (en) | Method and device for identifying main picture in web page | |
| CN116186318B (en) | Image retrieval method, device and storage medium based on association rule learning | |
| CN115687756A (en) | Search recommendation method and device | |
| CN110853115B (en) | Creation method and device of development flow page | |
| Jayaprabha et al. | Content based image retrieval methods using self supporting retrieval map algorithm | |
| Choi et al. | Object-aware image thumbnailing using image classification and enhanced detection of ROI | |
| CN103164504A (en) | Smartphone refined picture searching system and method | |
| Horiuchi et al. | JupySim: Jupyter Notebook Similarity Search System. | |
| CN115587140B (en) | Electronic engineering project data visual management method and device based on big data | |
| Li | Deep-learning-based research on detection algorithms for marine fish | |
| CN117708366A (en) | Image retrieval method, device, equipment and storage medium based on artificial intelligence |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20231023 Address after: 102209 18 Riverside Avenue, Changping District science and Technology City, Beijing Patentee after: State Grid Smart Grid Research Institute Co.,Ltd. Patentee after: STATE GRID CORPORATION OF CHINA Address before: 102209 18 Riverside Avenue, Changping District science and Technology City, Beijing Patentee before: State Grid Smart Grid Research Institute Co.,Ltd. |