CN115905590A - Image recommendation method and system based on deep hash retrieval and terminal device - Google Patents
Image recommendation method and system based on deep hash retrieval and terminal device Download PDFInfo
- Publication number
- CN115905590A CN115905590A CN202211696082.9A CN202211696082A CN115905590A CN 115905590 A CN115905590 A CN 115905590A CN 202211696082 A CN202211696082 A CN 202211696082A CN 115905590 A CN115905590 A CN 115905590A
- Authority
- CN
- China
- Prior art keywords
- image
- data
- image data
- candidate set
- user
- 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
-
- 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
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses an image recommendation method, system and terminal equipment based on deep hash retrieval, wherein the method comprises the following steps: crawling image data from the Internet through a crawler program, and preprocessing the image data to obtain preprocessed image data; inputting the preprocessed image data into a trained deep hash model to obtain a hash value, and determining an image candidate set based on the hash value; obtaining historical interactive data of a user, inputting the historical interactive data of the user and the image candidate set into a preset sequencing model to obtain an image recommendation list, and feeding back the image recommendation list to the user. According to the image recommendation method and device, the image candidate set is determined through the depth hash algorithm, the image recommendation list is determined based on the sorting model, the image retrieval speed can be greatly improved, and therefore the recommendation process of the image is accelerated.
Description
Technical Field
The invention relates to the technical field of image recommendation, in particular to an image recommendation method, system and terminal device based on deep hash retrieval.
Background
With the development of data diversification of network platforms such as social, multimedia information, e-commerce and the like, especially image data, the number of digital images is increasing at an alarming rate. In the face of this trend, the potential demand for information is gradually shifting from text to images. However, it is difficult for users to quickly find out images of their own interest from a huge and complicated image database. Therefore, image recommendation facing user interest is gradually becoming a popular research direction. Meanwhile, the explosive growth of image data on the internet also brings new opportunities and challenges to the development of recommendation systems.
For a network platform taking a user as a core, the recommendation system can quickly recommend interesting contents to the user by capturing user preferences and carrying out personalized recommendation, and the recommendation system has very important significance for improving the information acquisition efficiency of the user and improving the economic benefit of the network platform. However, for the field of image recommendation, it becomes increasingly difficult to mine a user's interest profile of an image from sparse user-image interaction data. The images have dense semantic information, and the image-rich visual information is used for mining the correlation between the user and the images, so that the image recommendation is a breakthrough. Therefore, the key challenges of image recommendation are how to mine high-level semantic information of images and how to model the images and users for relevant representation, thereby solving the contradiction between user demand and overloaded image data.
Image recommendation is used as a subdivision field of a recommendation system, early image recommendation basically treats an image as an item, recommendation is performed by using description attributes of the image and user interaction records, and a traditional Content-Based algorithm, an ItemCF collaborative filtering algorithm or a matrix decomposition algorithm and the like are adopted in the algorithm. With the continuous development of image feature extraction technology, the algorithm of image recommendation starts to be integrated with the artificially extracted image features, which is most obvious in effect improvement of a recommendation system based on content. Meanwhile, the image retrieval technology also brings new inspiration for image recommendation, in the traditional image retrieval process, characters are labeled on the image manually, and then the image is retrieved by using keywords, so that the method for providing the retrieval result according to the character matching degree described by the image is time-consuming, subjective and ambiguous, and the recommendation effect is not ideal. In a large-scale image application scenario, a certain number of images with similar contents are quickly found out from a million-level image database through image features given a query image, and content-based image retrieval (content-based image retrieval) is a very popular research direction at present, but the retrieval speed is still limited by an image feature vector. Therefore, the conventional image recommendation method is time-consuming and low in recommendation efficiency.
Thus, there is a need for improvements and enhancements in the art.
Disclosure of Invention
The technical problem to be solved by the present invention is to provide an image recommendation method, system and terminal device based on deep hash search for overcoming the above-mentioned defects in the prior art, and to solve the problems of time consumption and low recommendation efficiency of the image recommendation method in the prior art.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows:
in a first aspect, the present invention provides an image recommendation method based on a deep hash search, where the method includes:
crawling image data from the Internet through a crawler program, and preprocessing the image data to obtain preprocessed image data;
inputting the preprocessed image data into a trained deep hash model to obtain a hash value, and determining an image candidate set based on the hash value;
obtaining historical interactive data of a user, inputting the historical interactive data of the user and the image candidate set into a preset sequencing model to obtain an image recommendation list, and feeding back the image recommendation list to the user.
In one implementation, the preprocessing the image data to obtain preprocessed image data includes:
and carrying out format detection, picture cleaning and picture indexing on the image data.
In one implementation, the determining a candidate set of images based on the hash value includes:
acquiring an index database, wherein the index database comprises image characteristics corresponding to the image data;
determining the image candidate set based on the hash value and the image feature.
In one implementation, the determining the candidate set of images based on the hash value and the image feature includes:
calculating the similarity between the image features of the image data based on the hash value, and acquiring the image features with the similarity higher than a preset threshold value to obtain candidate image features;
and obtaining candidate image data based on the candidate image characteristics, and taking the candidate image data as the image candidate set.
In one implementation, the obtaining of the user historical interaction data includes:
and acquiring historical query items generated by browsing, clicking and collecting pictures by a user in a historical time period, and taking the historical query items as the historical interactive data of the user.
In one implementation, the inputting the user history interaction data and the image candidate set into a preset ranking model to obtain an image recommendation list includes:
determining feedback behavior data corresponding to the historical query items according to the historical user interaction data;
and based on the sorting model, sorting each piece of image data in the image candidate set according to the feedback behavior data, and outputting the image recommendation list.
In one implementation, the sorting each piece of image data in the image candidate set according to the feedback behavior data based on the sorting model and outputting the image recommendation list includes:
based on the ranking model, carrying out Bayesian analysis on each image data in the image candidate set to obtain a maximum posterior probability;
and sequencing each piece of image data in the image candidate set based on the maximum posterior probability, and outputting the image recommendation list.
In a second aspect, an embodiment of the present invention further provides an image recommendation system based on deep hash search, where the system includes:
the image data processing module is used for crawling image data from the Internet through a crawler program and preprocessing the image data to obtain preprocessed image data;
the image candidate set determining module is used for inputting the preprocessed image data into a trained deep hash model to obtain a hash value, and determining an image candidate set based on the hash value;
and the image recommendation module is used for acquiring historical interaction data of a user, inputting the historical interaction data of the user and the image candidate set into a preset sequencing model to obtain an image recommendation list, and feeding the image recommendation list back to the user.
In a third aspect, an embodiment of the present invention further provides a terminal device, where the terminal device includes a memory, a processor, and an image recommendation program based on a deep hash search, which is stored in the memory and is executable on the processor, and when the processor executes the image recommendation program based on the deep hash search, the step of implementing the image recommendation method based on the deep hash search in any of the foregoing solutions is implemented.
In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, where an image recommendation program based on a deep hash search is stored on the computer-readable storage medium, and when the image recommendation program based on the deep hash search is executed by a processor, the steps of the image recommendation method based on the deep hash search in any of the above solutions are implemented.
Has the advantages that: compared with the prior art, the invention provides an image recommendation method based on deep hash retrieval. And then, inputting the preprocessed image data into a trained deep hash model to obtain a hash value, and determining an image candidate set based on the hash value. And finally, obtaining historical user interaction data, inputting the historical user interaction data and the image candidate set into a preset sequencing model to obtain an image recommendation list, and feeding the image recommendation list back to the user. According to the image recommendation method and device, the image candidate set is determined through the depth hash algorithm, the image recommendation list is determined based on the sorting model, the image retrieval speed can be greatly increased, and therefore the recommendation process of the image is accelerated.
Drawings
Fig. 1 is a flowchart of a specific implementation of an image recommendation method based on a deep hash search according to an embodiment of the present invention.
Fig. 2 is a schematic overall framework diagram of an image recommendation system based on deep hash search according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of a depth hash model in an image recommendation method based on depth hash retrieval according to an embodiment of the present invention.
Fig. 4 is a schematic diagram illustrating a principle of a ranking model in an image recommendation method based on a deep hash search according to an embodiment of the present invention.
Fig. 5 is a schematic block diagram of an image recommendation system based on deep hash search according to an embodiment of the present invention.
Fig. 6 is a schematic block diagram of an internal structure of a terminal device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and effects of the present invention clearer and clearer, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The embodiment of the invention provides an image recommendation method based on deep hash retrieval. Then, the embodiment inputs the preprocessed image data into a trained deep hash model to obtain a hash value, and determines an image candidate set based on the hash value. Finally, the embodiment acquires the historical interaction data of the user, inputs the historical interaction data of the user and the image candidate set into a preset sequencing model to obtain an image recommendation list, and feeds the image recommendation list back to the user. According to the image recommendation method and device, the image candidate set is determined through the depth hash algorithm, the image recommendation list is determined based on the sorting model, the image retrieval speed can be greatly improved, and therefore the image recommendation process is accelerated.
Exemplary method
The image recommendation method based on the deep hash retrieval can be applied to terminal equipment, and the terminal equipment is an intelligent product terminal such as a computer, an intelligent television, a mobile phone and the like. In this embodiment, as shown in fig. 1, the image recommendation method based on depth hash search includes the following steps:
and S100, crawling image data from the Internet through a crawler program, and preprocessing the image data to obtain preprocessed image data.
Specifically, the embodiment first writes a crawler program for crawling image data. In the embodiment, the decoration style recommendation item is taken as an example, so that firstly, a picture website system can be constructed, the picture website system can collect a plurality of image data, and the image data respectively correspond to different decoration styles. For this reason, the present embodiment may crawl image data by a crawler program and then perform necessary preprocessing on the image data. In this embodiment, as shown in fig. 2, the preprocessing includes performing one or more of format detection, picture cleaning, and picture indexing on the image data. After the image data is preprocessed, the preprocessed image data can be obtained.
And S200, inputting the preprocessed image data into a trained deep hash model to obtain a hash value, and determining an image candidate set based on the hash value.
After the preprocessed image data is obtained, the preprocessed image data may be input into a trained deep hash model, and an image candidate set is obtained through processing of the deep hash model. In the embodiment, the deep convolutional neural network is selected by the deep hash model, because the topology of the network can be better matched with the input image, the feature extraction and the image classification can be simultaneously carried out, and in addition, the training parameters of the network can be greatly reduced in the training process due to the characteristics of CNN sparse connection and weight sharing, so that the whole structure of the convolutional neural network has stronger adaptability and is simpler. The deep hash model may be, as shown in fig. 3, embedding a hash algorithm into the last layer of the CNN model (as shown in fig. 2), and then setting a classification task to perform error back propagation to optimize the model, so that the image features and the hash function may be learned simultaneously, and thus the trained deep hash model may automatically extract the image features based on the image data and calculate the hash value. After the hash value is obtained, the embodiment can determine the image candidate set. The image candidate set is a candidate set in image recommendation.
Specifically, in this embodiment, an index database is first obtained, where the index database includes image features corresponding to the image data. Then, in this embodiment, based on the hash value, the similarity between the image features of the image data is calculated, and the image feature with the similarity higher than a preset threshold is obtained, so as to obtain a candidate image feature. The image data corresponding to the candidate image features have high similarity. Therefore, the present embodiment obtains candidate image data based on the candidate image feature, and uses the candidate image data as the image candidate set. The embodiment adopts a locality sensitive hashing algorithm. The traditional hash algorithm avoids hash collision, and distributes data at different coding positions as much as possible, and the idea of the algorithm is that the probability of mapping high-similarity data to the same position is high, or the mapping positions are close to each other, so that the generated coding space and the image feature space can be kept similar, the similarity measurement calculation speed of the generated binary coding is high, and good retrieval speed performance can be achieved.
Step S300, obtaining user historical interaction data, inputting the user historical interaction data and the image candidate set into a preset sequencing model to obtain an image recommendation list, and feeding the image recommendation list back to a user.
Specifically, after the image candidate set is obtained, the historical interaction data of the user can be obtained, and in specific application, the user generates historical query items by browsing, clicking and collecting pictures in a historical time period, and the historical query items are interaction data, so that the historical query items can be used as the historical interaction data of the user. Next, the present embodiment inputs the historical user interaction data and the candidate image set into a preset ranking model, which is an improved BPR ranking model (bayesian personalized ranking model). Different from other methods based on the user scoring matrix, the BPR ranking model of this embodiment mainly obtains implicit feedback (such as click, collection, etc.) of the user in the interaction process, as shown in fig. 4, is a + number indicate that the user has feedback behavior on item (i.e., historical query item),? The sign then indicates no feedback, and feedback behavior data for each user is generated by a feedback matrix, the + sign in fig. 4 indicates that the user prefers i between i and j-then the opposite, are? The numbers indicate no feedback or the same feedback. Therefore, the present embodiment may determine the feedback behavior data corresponding to the historical query term according to the historical interaction data of the user. Since the feedback behavior data can analyze which image data the user feeds back in the user historical interaction data, the embodiment may perform bayesian analysis on each image data in the image candidate set at this time based on the BPR ranking model to obtain a maximum posterior probability, where the maximum posterior probability is used to reflect a probability that the image data is fed back by the user, such as clicking, collecting, and the like). Then, the present embodiment sorts each piece of image data in the image candidate set based on the maximum posterior probability, and sorts the pieces of image data from high to low in the sorting process, and then outputs the image recommendation list. And feeds back the image recommendation list to the user. Interaction feedback made by a user based on the image recommendation list is recorded, and a corresponding query item is generated, and the query item can be input into the deep hash model to perform parameter optimization on the deep hash model, so that the deep hash model can output a more accurate hash value. In addition, the embodiment also adopts the mixed collaborative filtering of the fused image side information to increase the parameter constraint of the BPR model, thereby greatly strengthening the low-dimensional hidden representation of the User and the image and relieving the data sparsity problem of only using a User-Item matrix training model.
In summary, the recommended images retrieved by using the conventional image hash method depend on features extracted manually, and the visual preference of the user is difficult to capture. The image recommendation method of the embodiment aims to screen out images which meet the visual preference or potential requirement of a user from a large amount of image data and give a group of ranked recommended images according to scores. The embodiment can effectively retrieve similar images by utilizing the traditional image hashing algorithm to calculate the image similarity, but the distance of different images is difficult to measure. According to the method, the image similarity is calculated by adopting a deep hash algorithm, the method effectively combines the high-level semantic features extracted by a deep neural network and the Hash-like encoding operation, not only can the semantic information of the image be used for compression encoding, but also the operation efficiency and the resource consumption are considered. The data volume of the recommendation system of this embodiment is often very huge, so the recommendation system usually adopts some recall strategies to recall candidate data from massive data, and the calculation amount in the sorting stage is reduced. According to the recall strategy of the image recommendation system, the candidate image data set is recalled by utilizing the similarity matrix calculated by the image depth hash result and the user interaction record. In addition, the embodiment also provides an improved BPR sequencing model, the parameter constraint of the BPR model is increased by adopting mixed collaborative filtering of fused image side information, the hidden representation of the low dimension of the User and the image is enhanced, and the data sparsity problem of a model trained only by a User-Item matrix is relieved.
Exemplary System
Based on the foregoing embodiment, the present invention further provides an image recommendation system based on deep hash search, as shown in fig. 5, the system includes: an image data processing module 10, an image candidate set determination module 20, and an image recommendation module 30. Specifically, the image data processing module 10 of this embodiment is configured to crawl image data from the internet through a crawler program, and perform preprocessing on the image data to obtain preprocessed image data. The image candidate set determining module 20 is configured to input the preprocessed image data into a trained deep hash model to obtain a hash value, and determine an image candidate set based on the hash value. The image recommendation module 30 is configured to obtain user history interaction data, input the user history interaction data and the image candidate set into a preset ranking model to obtain an image recommendation list, and feed back the image recommendation list to a user.
In one implementation, the image data processing module 10 of the present embodiment includes:
and the preprocessing unit is used for carrying out format detection, picture cleaning and picture indexing on the image data.
In one implementation, the image candidate set determining module 20 of the present embodiment includes:
a database obtaining unit, configured to obtain an index database, where the index database includes image features corresponding to the image data;
a candidate set determining unit, configured to determine the image candidate set based on the hash value and the image feature.
In one implementation manner, the candidate set determining unit of this embodiment includes:
a candidate image feature obtaining subunit, configured to calculate, based on the hash value, a similarity between image features of the image data, and obtain an image feature with the similarity higher than a preset threshold, so as to obtain a candidate image feature;
and the image candidate set determining unit is used for obtaining candidate image data based on the candidate image characteristics and taking the candidate image data as the image candidate set.
In one implementation, the image recommendation module 30 of the present embodiment includes:
and the historical interactive data acquisition unit is used for acquiring historical query items generated by browsing, clicking and collecting pictures in a historical time period by a user, and taking the historical query items as the historical interactive data of the user.
In one implementation, the image recommendation module 30 of the present embodiment includes:
a feedback behavior data obtaining unit, configured to determine, according to the historical user interaction data, feedback behavior data corresponding to the historical query term;
and the recommendation list output unit is used for sorting each piece of image data in the image candidate set according to the feedback behavior data based on the sorting model and outputting the image recommendation list.
In one implementation manner, the recommendation list output unit of the embodiment includes:
the probability calculating subunit is used for carrying out Bayesian analysis on each piece of image data in the image candidate set based on the ranking model to obtain a maximum posterior probability;
and the list output subunit is used for sorting each piece of image data in the image candidate set based on the maximum posterior probability and outputting the image recommendation list.
The working principle of each module in the image recommendation system based on the deep hash search in this embodiment is the same as the execution process of each step in the above method embodiments, and is not described herein again.
Based on the above embodiment, the present invention further provides a terminal device, and a schematic block diagram of the terminal device may be as shown in fig. 6. The terminal device of this embodiment may include one or more processors 100 (only one shown in fig. 6), a memory 101, and a computer program 102 stored in the memory 101 and executable on the one or more processors 100, for example, a program for image recommendation based on depth hash retrieval. The one or more processors 100, when executing the computer program 102, may implement various steps in an embodiment of a method for image recommendation based on deep hash retrieval. Alternatively, one or more processors 100, when executing computer program 102, may implement the functions of the modules/units in the apparatus embodiment for image recommendation based on deep hash retrieval, which is not limited herein.
In one embodiment, the Processor 100 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
In one embodiment, the storage 101 may be an internal storage unit of the electronic device, such as a hard disk or a memory of the electronic device. The memory 101 may also be an external storage device of the electronic device, such as a plug-in hard disk, a Smart Memory Card (SMC), a Secure Digital (SD) card, a flash memory card (flash card), and the like provided on the electronic device. Further, the memory 101 may also include both an internal storage unit and an external storage device of the electronic device. The memory 101 is used for storing computer programs and other programs and data required by the terminal device. The memory 101 may also be used to temporarily store data that has been output or is to be output.
It will be understood by those skilled in the art that the block diagram of fig. 6 is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation to the terminal device to which the solution of the present invention is applied, and a specific terminal device may include more or less components than those shown in the figure, or may combine some components, or have different arrangements of components.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware related to instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, the computer program can include the processes of the embodiments of the methods described above. Any reference to memory, storage, operational databases, or other media used in embodiments provided herein may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), rambus (Rambus) direct RAM (RDRAM), direct Rambus Dynamic RAM (DRDRAM), and Rambus Dynamic RAM (RDRAM).
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. An image recommendation method based on deep hash retrieval is characterized by comprising the following steps:
crawling image data from the Internet through a crawler program, and preprocessing the image data to obtain preprocessed image data;
inputting the preprocessed image data into a trained deep hash model to obtain a hash value, and determining an image candidate set based on the hash value;
obtaining historical interactive data of a user, inputting the historical interactive data of the user and the image candidate set into a preset sequencing model to obtain an image recommendation list, and feeding back the image recommendation list to the user.
2. The image recommendation method based on the deep hash search as claimed in claim 1, wherein the preprocessing the image data to obtain the preprocessed image data comprises:
and carrying out format detection, picture cleaning and picture indexing on the image data.
3. The image recommendation method based on deep hash search according to claim 1, wherein determining the candidate set of images based on the hash value comprises:
acquiring an index database, wherein the index database comprises image characteristics corresponding to the image data;
determining the image candidate set based on the hash value and the image feature.
4. The image recommendation method based on deep hash search according to claim 3, wherein the determining the image candidate set based on the hash value and the image feature comprises:
calculating the similarity between the image features of the image data based on the hash value, and acquiring the image features with the similarity higher than a preset threshold value to obtain candidate image features;
and obtaining candidate image data based on the candidate image characteristics, and taking the candidate image data as the image candidate set.
5. The image recommendation method based on deep hash retrieval according to claim 1, wherein the obtaining of the user history interaction data comprises:
and acquiring historical query items generated by browsing, clicking and collecting pictures by a user in a historical time period, and taking the historical query items as the historical interactive data of the user.
6. The image recommendation method based on the deep hash search as claimed in claim 1, wherein the inputting the user history interaction data and the image candidate set into a preset ranking model to obtain an image recommendation list comprises:
determining feedback behavior data corresponding to the historical query items according to the historical user interaction data;
and based on the sorting model, sorting each piece of image data in the image candidate set according to the feedback behavior data, and outputting the image recommendation list.
7. The image recommendation method based on deep hash search according to claim 6, wherein said sorting each image data in the image candidate set according to the feedback behavior data based on the sorting model, and outputting the image recommendation list comprises:
based on the ranking model, carrying out Bayesian analysis on each image data in the image candidate set to obtain a maximum posterior probability;
and sequencing each piece of image data in the image candidate set based on the maximum posterior probability, and outputting the image recommendation list.
8. An image recommendation system based on deep hash retrieval, the system comprising:
the image data processing module is used for crawling image data from the Internet through a crawler program and preprocessing the image data to obtain preprocessed image data;
the image candidate set determining module is used for inputting the preprocessed image data into a trained deep hash model to obtain a hash value, and determining an image candidate set based on the hash value;
and the image recommendation module is used for acquiring historical interaction data of a user, inputting the historical interaction data of the user and the image candidate set into a preset sequencing model to obtain an image recommendation list, and feeding the image recommendation list back to the user.
9. A terminal device, characterized in that the terminal device comprises a memory, a processor and an image recommendation program based on deep hash search, which is stored in the memory and can run on the processor, and when the processor executes the image recommendation program based on deep hash search, the steps of the image recommendation method based on deep hash search according to any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium, wherein the computer-readable storage medium stores thereon an image recommendation program based on a depth hash search, and when the image recommendation program based on the depth hash search is executed by a processor, the image recommendation program based on the depth hash search implements the steps of the image recommendation method based on the depth hash search according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211696082.9A CN115905590A (en) | 2022-12-28 | 2022-12-28 | Image recommendation method and system based on deep hash retrieval and terminal device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211696082.9A CN115905590A (en) | 2022-12-28 | 2022-12-28 | Image recommendation method and system based on deep hash retrieval and terminal device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115905590A true CN115905590A (en) | 2023-04-04 |
Family
ID=86471145
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211696082.9A Pending CN115905590A (en) | 2022-12-28 | 2022-12-28 | Image recommendation method and system based on deep hash retrieval and terminal device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115905590A (en) |
-
2022
- 2022-12-28 CN CN202211696082.9A patent/CN115905590A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107391687B (en) | Local log website-oriented hybrid recommendation system | |
| CN109885773B (en) | Personalized article recommendation method, system, medium and equipment | |
| CN107220365B (en) | Accurate recommendation system and method based on collaborative filtering and association rule parallel processing | |
| US10217058B2 (en) | Predicting interesting things and concepts in content | |
| US8762383B2 (en) | Search engine and method for image searching | |
| CN103258000B (en) | Method and device for clustering high-frequency keywords in webpages | |
| CN112819023B (en) | Sample set acquisition method, device, computer equipment and storage medium | |
| CN112307762B (en) | Search result sorting method and device, storage medium and electronic device | |
| CN106708929B (en) | Video program searching method and device | |
| CN103544216A (en) | Information recommendation method and system combining image content and keywords | |
| CN112749330B (en) | Information pushing method, device, computer equipment and storage medium | |
| CN116975615A (en) | Task prediction method and device based on video multi-mode information | |
| US11727051B2 (en) | Personalized image recommendations for areas of interest | |
| CN113268667B (en) | Chinese comment emotion guidance-based sequence recommendation method and system | |
| CN111325030A (en) | Text label construction method and device, computer equipment and storage medium | |
| Tekli | An overview of cluster-based image search result organization: background, techniques, and ongoing challenges | |
| CN106570196B (en) | Video program searching method and device | |
| CN110769288A (en) | Video cold start recommendation method and system | |
| CN116975359A (en) | Resource processing method, resource recommending method, device and computer equipment | |
| US20240004888A1 (en) | Sorting method, apparatus and device, and computer storage medium | |
| CN113139049B (en) | Associated document recommending method, device, computer equipment and storage medium | |
| CN111797765B (en) | Image processing method, device, server and storage medium | |
| Yu et al. | Computer Image Content Retrieval considering K‐Means Clustering Algorithm | |
| CN115905590A (en) | Image recommendation method and system based on deep hash retrieval and terminal device | |
| CN112749313B (en) | Label labeling method, label labeling device, computer equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |