CN112199572B - Beijing pattern collecting and arranging system - Google Patents
Beijing pattern collecting and arranging system Download PDFInfo
- Publication number
- CN112199572B CN112199572B CN202011239453.1A CN202011239453A CN112199572B CN 112199572 B CN112199572 B CN 112199572B CN 202011239453 A CN202011239453 A CN 202011239453A CN 112199572 B CN112199572 B CN 112199572B
- Authority
- CN
- China
- Prior art keywords
- beijing
- pattern
- patterns
- module
- kyoto
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
-
- 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
- G06F16/51—Indexing; Data structures therefor; Storage structures
-
- 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
- G06F16/55—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Abstract
The invention discloses a Beijing pattern collection and arrangement system, which comprises: the Beijing pattern factor summarizing and classifying module is used for summarizing and classifying the Beijing pattern factors and generating a Beijing pattern factor set; the recognition model construction module is used for respectively training and obtaining corresponding Beijing pattern crawling models based on the Beijing pattern spectrum factor sets to generate a Beijing pattern recognition model set; the Beijing pattern crawler module is used for synchronously running the Beijing pattern recognition model sets based on Hadoop to respectively dig corresponding Beijing pattern data on each network base station to generate a Beijing pattern data set; and the Beijing pattern finishing module is used for finishing the Beijing patterns, eliminating the embedded repeated Beijing patterns and constructing the association relation among the Beijing patterns. The invention realizes the automatic collection and classification of the Beijing patterns, can greatly reduce the artificial workload and can well avoid the errors and omission of data.
Description
Technical Field
The invention relates to the field of Beijing patterns, in particular to a Beijing pattern collecting and arranging system.
Background
At present, the arrangement of the Beijing patterns is generally carried out manually, which is time-consuming and labor-consuming and easy to cause errors and omission of data.
Disclosure of Invention
In order to solve the problems, the invention provides a Beijing pattern collecting and sorting system, which realizes automatic collection, sorting and classified storage of the Beijing patterns, can greatly reduce the manual workload and can well avoid errors and omission of data.
In order to achieve the above purpose, the technical scheme adopted by the invention is as follows:
a kyoto pattern collection and finishing system, comprising:
the Beijing pattern factor summarizing and classifying module is used for summarizing and classifying the Beijing pattern factors and generating a Beijing pattern factor set;
the recognition model construction module is used for respectively training and obtaining corresponding Beijing pattern crawling models based on the Beijing pattern spectrum factor sets to generate a Beijing pattern recognition model set;
the Beijing pattern crawler module is used for synchronously running the Beijing pattern recognition model sets based on Hadoop to respectively dig corresponding Beijing pattern data on each network base station to generate a Beijing pattern data set;
and the Beijing pattern finishing module is used for finishing the Beijing patterns, eliminating the embedded repeated Beijing patterns and constructing the association relation among the Beijing patterns.
Furthermore, the Beijing pattern factor summarization and classification module realizes the summarization of the Beijing pattern factors in a manual circling mode, extracts the characteristic parameters of the Beijing pattern factors, and classifies the Beijing pattern factors based on the similarity of the characteristic parameters.
Furthermore, the Beijing pattern recognition model adopts a Faster R-CNN model, and is obtained based on training of the Beijing pattern factor set, and one Beijing pattern recognition model corresponds to an independent data storage folder.
Further, each update of each set of Beijing pattern factors automatically generates a new Beijing pattern recognition model, which is built based on the new Beijing pattern factors.
Further, each Beijing pattern in the Beijing pattern data set carries a invisible Beijing pattern factor marking frame, and the display of the Beijing pattern factor marking frame can be realized by clicking the Beijing pattern.
Furthermore, the Beijing pattern arrangement module adopts an image classification algorithm based on a Lightweight Group Attention Module (LGAM) to construct the association relation among the Beijing patterns according to the Beijing pattern factors embedded in the Beijing patterns, and eliminates repeated Beijing pattern spectrum data.
Further, the method further comprises the following steps:
and the Beijing pattern detection module is used for detecting the target image and the video-on-Beijing pattern and feeding back the detection result in the form of an image carrying the Beijing pattern factor mark frame.
The invention has the following beneficial effects:
1) Through the internet technology and the Faster R-CNN model set, the automatic collection and classification of the Beijing patterns are realized, the artificial workload can be greatly reduced, and meanwhile, the errors and omission of data can be well avoided.
2) Different Faster R-CNN models are constructed based on the classification of the Beijing patterns, then the Beijing patterns are synchronously operated based on Hadoop to respectively dig corresponding Beijing pattern data on each network base station, and the Beijing patterns can be classified while the collection of the Beijing patterns is realized, so that the collection and classification efficiency of the Beijing patterns is greatly improved.
3) Each update of each Beijing pattern factor set automatically generates a new Beijing pattern recognition model, so that the automatic update of the Beijing pattern database can be realized.
Drawings
FIG. 1 is a block diagram of a system for collecting and sorting patterns in the Beijing opera according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to examples in order to make the objects and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Examples
As shown in fig. 1, an embodiment of the present invention provides a kyoto pattern collection and arrangement system, including:
the Beijing pattern factor summarizing and classifying module is used for summarizing and classifying the Beijing pattern factors and generating a Beijing pattern factor set;
the recognition model construction module is used for respectively training and obtaining corresponding Beijing pattern crawling models based on the Beijing pattern spectrum factor sets to generate a Beijing pattern recognition model set;
the Beijing pattern crawler module is used for synchronously running the Beijing pattern recognition model sets based on Hadoop to respectively dig corresponding Beijing pattern data on each network base station to generate a Beijing pattern data set;
the Beijing pattern finishing module is used for finishing the Beijing patterns, removing the embedded repeated Beijing patterns and constructing the association relation among the Beijing patterns;
and the Beijing pattern detection module is used for detecting the target image and the video-on-Beijing pattern and feeding back the detection result in the form of an image carrying the Beijing pattern factor mark frame. When the system is used, a user uploads a target image and a target video in an uploading mode, after the uploading is finished, the user clicks 'detection', the Beijing pattern detection module is started, and the Beijing pattern recognition model set is awakened to recognize the factors of the patterns of the target image and the target video, wherein a video frame taking script is configured for the target video, and an image is acquired every certain frame number. The module can also be used for pre-detection of the Beijing pattern, namely, when a user finds a new Beijing pattern factor, the module can confirm whether the Beijing pattern factor exists in the current system, and if the identification result is blank, the Beijing pattern factor is proved to be a new factor.
In this embodiment, the kyoto pattern factor summarization classification module realizes the summarization of the kyoto pattern factors by a manual defining mode, and realizes the classification of the kyoto pattern factors based on the similarity of the characteristic parameters by extracting the characteristic parameters of the kyoto pattern factors. Specifically, the present Beijing pattern set is uploaded through the Beijing pattern/video uploading module, and then the Beijing pattern factors on each Beijing pattern are defined in a way of traversing each Beijing pattern, and the defined Beijing pattern areas automatically generate the Beijing pattern factor images, so that the collection of the Beijing pattern factors is realized; the extraction of the characteristic parameters of the Beijing pattern factor image is realized based on the BP neural network model, and the classification of the Beijing pattern factor is realized based on the similarity of the characteristic parameters.
In this embodiment, the Beijing pattern recognition model is a Faster R-CNN model, and is trained based on the Beijing pattern factor set, and one Beijing pattern recognition model corresponds to an independent data storage folder.
In this embodiment, each update of each of the set of kyoto pattern factors automatically generates a new kyoto pattern recognition model, the new kyoto pattern recognition model is built based on the new kyoto pattern factors, after the new kyoto pattern recognition model is generated, the kyoto pattern crawler module is automatically started, and corresponding kyoto pattern data is respectively extracted from each network base station based on the new kyoto pattern recognition model; it is noted that the new Beijing pattern recognition model automatically organizes the collected Beijing patterns into the data storage file corresponding to the Beijing pattern recognition model constructed by the original Beijing pattern factor set.
In the implementation, each Beijing pattern in the Beijing pattern data set carries a invisible Beijing pattern factor marking frame, and the display of the Beijing pattern factor marking frame can be realized by clicking the Beijing pattern, so that the direct check of the Beijing pattern factors on each Beijing pattern is convenient for a worker.
In this embodiment, the kyoto pattern arrangement module uses an image classification algorithm based on a Lightweight Group Attention Module (LGAM) to construct an association relationship between the kyoto patterns according to the kyoto pattern factors embedded in the kyoto patterns, and eliminates repeated kyoto pattern spectrum data.
The foregoing is merely a preferred embodiment of the present invention and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present invention, which are intended to be comprehended within the scope of the present invention.
Claims (7)
1. A kyoto pattern collection and finishing system, comprising:
the Beijing pattern factor summarizing and classifying module is used for summarizing and classifying the Beijing pattern factors and generating a Beijing pattern factor set;
the recognition model construction module is used for respectively training and obtaining corresponding Beijing pattern recognition models based on the Beijing pattern factor sets and generating a Beijing pattern recognition model set;
the Beijing pattern crawler module is used for synchronously running the Beijing pattern recognition model sets based on Hadoop to respectively dig corresponding Beijing pattern data on each network base station to generate a Beijing pattern data set;
and the Beijing pattern finishing module is used for finishing the Beijing patterns, eliminating the embedded repeated Beijing patterns and constructing the association relation among the Beijing patterns.
2. The system according to claim 1, wherein the classification module is configured to collect the pattern factors of the Beijing opera-tion by manual definition, and to classify the pattern factors of the Beijing opera-tion based on the similarity of the characteristic parameters by extracting the characteristic parameters of the pattern factors of the Beijing opera-tion.
3. The system of claim 1, wherein the kyoto pattern recognition model is a fast R-CNN model, and is trained based on a kyoto pattern factor set, and wherein one kyoto pattern recognition model corresponds to one independent data storage folder.
4. The system of claim 1, wherein each update of the set of kyoto pattern factors automatically generates a new kyoto pattern recognition model, the new kyoto pattern recognition model being constructed based on the new kyoto pattern factors.
5. The system of claim 1, wherein each of the plurality of patterns in the plurality of patterns carries a mark frame for a pattern factor of the plurality of patterns, wherein the mark frame for the pattern factor of the plurality of patterns is invisible.
6. The system according to claim 1, wherein the Beijing pattern arrangement module uses an image classification algorithm based on a lightweight group attention module to construct the association relationship between the Beijing patterns according to the Beijing pattern factors loaded in the Beijing patterns, and eliminates the repeated Beijing pattern spectrum data.
7. The kyoto pattern collection finishing system of claim 1, further comprising:
and the Beijing pattern detection module is used for detecting the target image and the video-on-Beijing pattern and feeding back the detection result in the form of an image carrying the Beijing pattern factor mark frame.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011239453.1A CN112199572B (en) | 2020-11-09 | 2020-11-09 | Beijing pattern collecting and arranging system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011239453.1A CN112199572B (en) | 2020-11-09 | 2020-11-09 | Beijing pattern collecting and arranging system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112199572A CN112199572A (en) | 2021-01-08 |
| CN112199572B true CN112199572B (en) | 2023-06-06 |
Family
ID=74033103
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011239453.1A Active CN112199572B (en) | 2020-11-09 | 2020-11-09 | Beijing pattern collecting and arranging system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112199572B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106529564A (en) * | 2016-09-26 | 2017-03-22 | 浙江工业大学 | Food image automatic classification method based on convolutional neural networks |
| CN107437099A (en) * | 2017-08-03 | 2017-12-05 | 哈尔滨工业大学 | A kind of specific dress ornament image recognition and detection method based on machine learning |
| CN110020674A (en) * | 2019-03-13 | 2019-07-16 | 清华大学 | A kind of cross-cutting adapting to image classification method promoting topic categories discrimination |
| CN110059672A (en) * | 2019-04-30 | 2019-07-26 | 福州大学 | A method of microscopic cell image detection model is carried out using incremental learning to increase class study |
| CN110853052A (en) * | 2019-10-25 | 2020-02-28 | 湖北民族大学 | Tujia brocade pattern primitive segmentation method based on deep learning |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2013186793A (en) * | 2012-03-09 | 2013-09-19 | Toshiba Corp | Information processing device, image file generation method and program |
| IL241863A0 (en) * | 2015-10-06 | 2016-11-30 | Agent Video Intelligence Ltd | Method and system for classifying objects from a stream of images |
-
2020
- 2020-11-09 CN CN202011239453.1A patent/CN112199572B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106529564A (en) * | 2016-09-26 | 2017-03-22 | 浙江工业大学 | Food image automatic classification method based on convolutional neural networks |
| CN107437099A (en) * | 2017-08-03 | 2017-12-05 | 哈尔滨工业大学 | A kind of specific dress ornament image recognition and detection method based on machine learning |
| CN110020674A (en) * | 2019-03-13 | 2019-07-16 | 清华大学 | A kind of cross-cutting adapting to image classification method promoting topic categories discrimination |
| CN110059672A (en) * | 2019-04-30 | 2019-07-26 | 福州大学 | A method of microscopic cell image detection model is carried out using incremental learning to increase class study |
| CN110853052A (en) * | 2019-10-25 | 2020-02-28 | 湖北民族大学 | Tujia brocade pattern primitive segmentation method based on deep learning |
Non-Patent Citations (2)
| Title |
|---|
| Vehicle Detection Based on Drone Images with the Improved Faster R-CNN;Lixin Wang等;《Proceedings of the 2019 11th International Conference on Machine Learning and Computing》;第466–471页 * |
| 融合优选图案的深度学习目标识别及定位技术;王立鹏等;《哈尔滨工程大学学报》;第549-555页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112199572A (en) | 2021-01-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109389161B (en) | Garbage identification evolutionary learning method, device, system and medium based on deep learning | |
| CN103824053B (en) | The sex mask method and face gender detection method of a kind of facial image | |
| CN111444848A (en) | Specific scene model upgrading method and system based on federal learning | |
| CN102346847B (en) | License plate character recognizing method of support vector machine | |
| CN109063649B (en) | Pedestrian re-identification method based on twin pedestrian alignment residual error network | |
| CN108090406B (en) | Face recognition method and system | |
| Huang et al. | Detection of human faces using decision trees | |
| CN101187927A (en) | Criminal case joint investigation intelligent analysis method | |
| CN110929760A (en) | Garbage classification software based on computer vision | |
| CN111488917A (en) | Garbage image fine-grained classification method based on incremental learning | |
| CN110348505B (en) | Vehicle color classification model training method and device and vehicle color identification method | |
| CN112367273B (en) | Flow classification method and device of deep neural network model based on knowledge distillation | |
| CN106407369A (en) | Photo management method and system based on deep learning face recognition | |
| CN112733936A (en) | Recyclable garbage classification method based on image recognition | |
| CN113158891B (en) | Cross-camera pedestrian re-identification method based on global feature matching | |
| CN109635805A (en) | Image text location method and device, image text recognition methods and device | |
| CN113627266A (en) | Video pedestrian re-identification method based on Transformer space-time modeling | |
| CN109583375A (en) | A kind of the facial image illumination recognition methods and system of multiple features fusion | |
| CN107577994A (en) | A kind of pedestrian based on deep learning, the identification of vehicle auxiliary product and search method | |
| CN110443181A (en) | Face identification method and device | |
| CN113761242A (en) | Big data image recognition system and method based on artificial intelligence | |
| CN116228319A (en) | Intelligent advertisement putting method and system based on buried points | |
| CN111461000A (en) | Intelligent office garbage classification method based on CNN and wavelet analysis | |
| CN110826449A (en) | Non-motor vehicle re-identification target retrieval method based on light convolutional neural network | |
| CN114140696A (en) | Commodity identification system optimization method, commodity identification system optimization device, commodity identification 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |