CN110163244A - A kind of ceramic tile texture classifying method and device - Google Patents
A kind of ceramic tile texture classifying method and device Download PDFInfo
- Publication number
- CN110163244A CN110163244A CN201910269271.XA CN201910269271A CN110163244A CN 110163244 A CN110163244 A CN 110163244A CN 201910269271 A CN201910269271 A CN 201910269271A CN 110163244 A CN110163244 A CN 110163244A
- Authority
- CN
- China
- Prior art keywords
- image
- classification
- nonlinear characteristic
- sorted
- linear distance
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24133—Distances to prototypes
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
This application involves technical field of image processing more particularly to a kind of ceramic tile texture classifying methods and device.The described method includes: extracting depth nonlinear characteristic from tile image to be sorted based on feature extraction network；Based on the non-linear distance between depth nonlinear characteristic described in full connection relationship network query function and N number of image classification template, N > 1；Using the classification of image classification template corresponding to the smallest non-linear distance in calculated N number of non-linear distance as the classification of the tile image to be sorted.The embodiment of the present invention is capable of providing the accuracy of ceramic tile Texture classification.
Description
Technical field
This application involves technical field of image processing more particularly to a kind of ceramic tile texture classifying methods and device.
Background technique
With the continuous improvement of scientific and technological level, computer vision is had been to be concerned by more and more people.Computer is regarded
Feel is applied in industrial links to replace artificial, raising production efficiency to have become a kind of trend.In tile typology
In field, since ceramic tile texture is many kinds of, difference is smaller between classification, it is difficult by simple, intuitive standard to ceramic tile texture
Rationally distinguished.
In the currently used method based on computer vision to ceramic tile Texture classification, mainly according to region of interest
The textural characteristics or colour difference information that domain is extracted are classified, due to texture huge number, when the textural characteristics and face of different ceramic tiles
The classification results obtained when color ratio is closer to can have bigger error, so as to cause higher false detection rate.
For example, in a kind of existing ceramic tile texture classifying method, it is different from background reflectivity based on ceramic tile glaze texture
The characteristics of, texture is divided into macro-texture and high reflectance texture, in combination with tone, gray scale to ceramic tile Texture classification, the party
Method solves the problems, such as high reflection texture and background color difference close to being difficult to differentiate between, but multi-class textural characteristics and color more
Ideal classifying quality is difficult to realize when close.
Summary of the invention
In order to solve the above-mentioned technical problem or it at least is partially solved above-mentioned technical problem, this application provides a kind of porcelain
Brick texture classifying method and device.
In a first aspect, the embodiment of the invention provides a kind of ceramic tile texture classifying methods, comprising:
Depth nonlinear characteristic is extracted from tile image to be sorted based on feature extraction network；
Based on non-linear between depth nonlinear characteristic described in full connection relationship network query function and N number of image classification template
Distance, N > 1；
The classification of image classification template corresponding to the smallest non-linear distance in calculated N number of non-linear distance is made
For the classification of the tile image to be sorted.
Optionally, depth nonlinear characteristic is extracted from tile image to be sorted based on feature extraction network, comprising:
Key feature image is intercepted from the tile image to be sorted；
Depth nonlinear characteristic is extracted from the key feature image based on feature extraction network.
Optionally, the feature extraction network includes: multiple convolutional layers and at least two maximum pond layers, be used for from
Depth nonlinear characteristic is extracted in the tile image to be sorted.
Optionally, it is based on described in full connection relationship network query function between depth nonlinear characteristic and N number of image classification template
Non-linear distance, comprising:
The depth nonlinear characteristic is subjected to the extension of N-dimensional degree；
Based on full connection relationship network, the N number of non-linear special feature and N number of figure that the N-dimensional degree extends are calculated
As the non-linear distance between classification model.
Optionally, the depth nonlinear characteristic is subjected to the extension of N-dimensional degree, comprising:
Depth nonlinear characteristic duplication n times are obtained into N number of depth nonlinear characteristic.
Optionally, the full connection relationship network includes: multiple convolutional layers, multiple maximum pond layers and multiple full connections
Layer, for calculating the non-linear distance between the depth nonlinear characteristic and N number of image classification template.
The embodiment of the invention provides a kind of ceramic tile texture classification apparatus for second aspect, comprising:
Characteristic extracting module, for extracting the non-linear spy of depth from tile image to be sorted based on feature extraction network
Sign；
Distance calculation module, for based on depth nonlinear characteristic and N number of image described in full connection relationship network query function point
Non-linear distance between class template, N > 1；
Category determination module, for by figure corresponding to the smallest non-linear distance in calculated N number of non-linear distance
As classification of the classification as the tile image to be sorted of classification model.
Optionally, the characteristic extracting module is specifically used for:
Key feature image is intercepted from the tile image to be sorted；
Depth nonlinear characteristic is extracted from the key feature image based on feature extraction network.
Optionally, the distance calculation module is specifically used for:
The depth nonlinear characteristic is subjected to the extension of N-dimensional degree；
Based on full connection relationship network, the N number of non-linear special feature and N number of figure that the N-dimensional degree extends are calculated
As the non-linear distance between classification model.
The third aspect, the embodiment of the invention provides a kind of ceramic tile texture classification apparatus, comprising: memory, processor,
In:
For the memory for storing one or more computer instruction, one or more computer instruction is described
Processor realizes above-mentioned ceramic tile texture classifying method when executing.
Fourth aspect, the embodiment of the invention provides a kind of computer storage mediums, which is characterized in that the storage medium
For storing computer program, the computer program is for realizing above-mentioned ceramic tile Texture classification side when executing computer
Method.
Above-mentioned technical proposal provided by the embodiments of the present application has the advantages that compared with prior art
In the ceramic tile Texture classification scheme of the embodiment of the present invention, the non-linear spy of depth is extracted from tile image to be sorted
Sign, and it is non-thread between the depth nonlinear characteristic and N number of image classification template extracted based on full connection relationship network query function
Property distance, using the non-linear classification apart from corresponding image classification template minimum in N number of non-linear distance as ceramic tile to be sorted
The classification of image.The program is not limited by extracted feature is applicable to multi-class ceramic tile Texture classification, can be improved ceramic tile
The efficiency and accuracy of Texture classification, accuracy of identification height, strong robustness.
Detailed description of the invention
The drawings herein are incorporated into the specification and forms part of this specification, and shows and meets implementation of the invention
Example, and be used to explain the principle of the present invention together with specification.
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, for those of ordinary skill in the art
Speech, without any creative labor, is also possible to obtain other drawings based on these drawings.
Fig. 1 is the flow chart for the ceramic tile texture classifying method that the embodiment of the present invention one provides；
Fig. 2 is the flow chart of ceramic tile texture classifying method provided by Embodiment 2 of the present invention；
Fig. 3 is a kind of structural schematic diagram of feature extraction network provided in an embodiment of the present invention；
Fig. 4 is a kind of structural schematic diagram of full connection relationship network provided in an embodiment of the present invention；
Fig. 5 is a kind of ceramic tile texture color lump classification schematic diagram provided in an embodiment of the present invention；
Fig. 6 is the structural schematic diagram of ceramic tile texture classification apparatus provided in an embodiment of the present invention；
Fig. 7 is the structural schematic diagram of electronic equipment corresponding to Fig. 6 shown device.
Specific embodiment
To keep the purposes, technical schemes and advantages of the embodiment of the present application clearer, below in conjunction with the embodiment of the present application
In attached drawing, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the application, instead of all the embodiments.Based on the embodiment in the application, ordinary skill people
Member's every other embodiment obtained without making creative work, shall fall in the protection scope of this application.
The term used in embodiments of the present invention is only to be not intended to be limiting merely for for the purpose of describing particular embodiments
The present invention.In the embodiment of the present invention and the "an" of singular used in the attached claims, " described " and "the"
It is also intended to including most forms, unless the context clearly indicates other meaning, " a variety of " generally comprise at least two.
It should be appreciated that term "and/or" used herein is only a kind of incidence relation for describing affiliated partner, indicate
There may be three kinds of relationships, for example, A and/or B, can indicate: individualism A, exist simultaneously A and B, individualism B these three
Situation.In addition, character "/" herein, typicallys represent the relationship that forward-backward correlation object is a kind of "or".
In order to solve the problems, such as that ceramic tile Texture classification error is larger in the prior art, efficiency is lower, the embodiment of the present invention is mentioned
A kind of ceramic tile texture classifying method is supplied, this method is based on computer vision recognition technology, and not limited by extracted feature can
Suitable for multi-class ceramic tile Texture classification, the efficiency and accuracy of ceramic tile Texture classification, accuracy of identification height, robustness can be improved
By force.
Fig. 1 is the flow chart for the ceramic tile texture classifying method that the embodiment of the present invention one provides, and this method can be by supporting to count
The terminal device of calculation machine Visual identification technology executes.As shown in Figure 1, this method comprises the following steps:
Step S101: depth nonlinear characteristic is extracted from tile image to be sorted based on feature extraction network.
Step S102: based on depth nonlinear characteristic described in full connection relationship network query function and N number of image classification template it
Between non-linear distance, N > 1.
Step S103: by image classification mould corresponding to the smallest non-linear distance in calculated N number of non-linear distance
Classification of the classification of plate as the tile image to be sorted.
In the embodiment of the present invention, depth nonlinear characteristic is extracted from tile image to be sorted, later based on complete
Non-linear distance between the depth nonlinear characteristic of connection relationship network query function ceramic tile and various image classification templates, and root
Classification belonging to tile image to be sorted is determined according to calculated non-linear distance.Wherein, by calculated minimum it is non-linear away from
Classification of the classification as tile image to be sorted from corresponding image classification template.
Fig. 2 is the flow chart of ceramic tile texture classifying method provided by Embodiment 2 of the present invention, as shown in Fig. 2, this method packet
Include following steps:
Step S201: key feature image is intercepted from tile image to be sorted.
Before extracting depth nonlinear characteristic in tile image to be sorted, tile image to be sorted is carried out first pre-
Processing.Specifically, key feature image can be intercepted from tile image to be sorted, for example, with the ratio interception of length and width 1:1 to
Classification tile image best embodies the image of ceramic tile feature as key feature image, optionally, can also be by ceramic tile to be sorted
The image at position is as key feature image among image.It, can be with after intercepting key feature image in tile image to be sorted
Key feature image is zoomed in and out, so that key feature image meets subsequent calculation processing requirement, such as to key feature figure
The scaled size of picture is 84*84.
Step S202: depth nonlinear characteristic is extracted from the key feature image based on feature extraction network.
Optionally, the feature extraction network includes: multiple convolutional layers and at least two maximum pond layers, be used for from
Depth nonlinear characteristic is extracted in the tile image to be sorted.
As shown in figure 3, the structure of the feature extraction network includes: 4 convolutional layers and 2 maximum pond layers.
Step S203: the depth nonlinear characteristic is subjected to the extension of N-dimensional degree.
It is non-thread to the depth after the depth nonlinear characteristic for obtaining tile image to be sorted based on feature extraction network query function
Property feature carry out the extension of N-dimensional degree.
Optionally, the extension of N-dimensional degree is carried out to depth nonlinear characteristic, comprising: the depth nonlinear characteristic is replicated into n times
Obtain N number of depth nonlinear characteristic.
Step S204: being based on full connection relationship network, calculate N number of non-linear special feature that the N-dimensional degree extends with
Non-linear distance between N number of image classification template.
Wherein, the structure of full connection relationship network includes: multiple convolutional layers, multiple maximum pond layers and multiple full connections
Layer, for calculating the non-linear distance between the depth nonlinear characteristic and N number of image classification template.
As shown in figure 4, full connection relationship network includes 2 convolutional layers, 2 maximum ponds in a specific implementation
Change layer and 2 full articulamentums.
Step S205: by image classification mould corresponding to the smallest non-linear distance in calculated N number of non-linear distance
Classification of the classification of plate as the tile image to be sorted.
As shown in figure 5, sample characteristic is extracted from test sample (tile image i.e. to be sorted) based on feature extraction network,
And by the sample characteristic carry out the extension of N-dimensional degree, obtain N number of sample characteristic, by N number of sample especially with N number of template characteristic (i.e.
Image classification template) as input feature vector be input in full connection relationship network with calculate separately to obtain N number of sample characteristic with it is N number of
Non-linear distance between template characteristic, it is non-linear apart from representative sample feature and template between sample characteristic and template characteristic
Similarity between feature, the similarity between non-linear distance minimum representative sample feature and template characteristic is higher, takes similar
Spend classification of the highest template characteristic as tile image to be sorted.
The ceramic tile texture classification apparatus of one or more embodiments of the invention described in detail below.Those skilled in the art
Member is it is appreciated that described device is configured the step of can be used commercially available hardware component instructed by this programme to constitute.
Fig. 6 is the structural schematic diagram of ceramic tile texture classification apparatus provided in an embodiment of the present invention, comprising: characteristic extracting module
11, distance calculation module 12 and category determination module 13, in which:
Characteristic extracting module 11, for extracting the non-linear spy of depth from tile image to be sorted based on feature extraction network
Sign；
Distance calculation module 12, for based on depth nonlinear characteristic and N number of image described in full connection relationship network query function
Non-linear distance between classification model, N > 1；
Category determination module 13, for will be corresponding to the smallest non-linear distance in calculated N number of non-linear distance
Classification of the classification of image classification template as the tile image to be sorted.
Optionally, the characteristic extracting module 11 is specifically used for: intercepting key feature from the tile image to be sorted
Image；Depth nonlinear characteristic is extracted from the key feature image based on feature extraction network.
Optionally, the distance calculation module 12 is specifically used for: the depth nonlinear characteristic is carried out the extension of N-dimensional degree；
Based on full connection relationship network, the N number of non-linear special feature and N number of image classification mould that the N-dimensional degree extends are calculated
Non-linear distance between plate.
Fig. 6 shown device can execute the operation that ceramic tile texture classification apparatus executes in above method embodiment, this implementation
The part that example is not described in detail can refer to the related description of above-mentioned shown embodiment of the method.The implementation procedure of the technical solution and
Technical effect is referring to the description in above-mentioned shown embodiment of the method, and details are not described herein.
The foregoing describe the built-in function of ceramic tile texture classification apparatus and structures, in a possible design, ceramic tile line
The structure of reason sorter can be realized as an electronic equipment, such as computer, as shown in fig. 7, the electronic equipment may include: place
Manage device 21 and memory 22.Wherein, the memory 22 supports the electronic equipment to execute in above method embodiment for storing
The program of the ceramic tile texture classifying method of offer, the processor 21 is configurable for executing to be stored in the memory 22
Program.It wherein, can also include communication interface 23 in the structure of the electronic equipment, for the electronic equipment and other equipment ratio
Such as memory node or communication.
In addition, the embodiment of the invention provides a kind of computer storage medium, for storing above-mentioned electronic equipments
Computer software instructions, it includes for executing program involved in ceramic tile texture classifying method in above method embodiment.
The apparatus embodiments described above are merely exemplary, wherein described, unit can as illustrated by the separation member
It is physically separated with being or may not be, component shown as a unit may or may not be physics list
Member, it can it is in one place, or may be distributed over multiple network units.It can be selected according to the actual needs
In some or all of the modules achieve the purpose of the solution of this embodiment.Those of ordinary skill in the art are not paying creativeness
Labour in the case where, it can understand and implement.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It is realized by the mode of required general hardware platform is added, naturally it is also possible to which reality is come in conjunction with by way of hardware and software
It is existing.Based on this understanding, substantially the part that contributes to existing technology can be to calculate in other words for above-mentioned technical proposal
The form of machine product embodies, and it wherein includes the meter of computer usable program code that the present invention, which can be used in one or more,
The computer journey implemented in calculation machine usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of sequence product.
In a typical configuration, calculating equipment includes one or more processors (CPU), input/output interface, net
Network interface and memory.
Memory may include the non-volatile memory in computer-readable medium, random access memory (RAM) and/or
The forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flash RAM).Memory is computer-readable medium
Example.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology come realize information store.Information can be computer readable instructions, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase change memory (PRAM), static random access memory (SRAM), moves
State random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electric erasable
Programmable read only memory (EEPROM), flash memory or other memory techniques, read-only disc read only memory (CD-ROM) (CD-ROM),
Digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape magnetic disk storage or other magnetic storage devices
Or any other non-transmission medium, can be used for storage can be accessed by a computing device information.As defined in this article, it calculates
Machine readable medium does not include temporary computer readable media (transitory media), such as the data-signal and carrier wave of modulation.
The above is only a specific embodiment of the invention, is made skilled artisans appreciate that or realizing this hair
It is bright.Various modifications to these embodiments will be apparent to one skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, of the invention
It is not intended to be limited to the embodiments shown herein, and is to fit to and applied principle and features of novelty phase one herein
The widest scope of cause.
Claims (10)
1. a kind of ceramic tile texture classifying method characterized by comprising
Depth nonlinear characteristic is extracted from tile image to be sorted based on feature extraction network；
Based between depth nonlinear characteristic described in full connection relationship network query function and N number of image classification template it is non-linear away from
From N > 1；
Using the classification of image classification template corresponding to the smallest non-linear distance in calculated N number of non-linear distance as institute
State the classification of tile image to be sorted.
2. the method according to claim 1, wherein being mentioned from tile image to be sorted based on feature extraction network
Take depth nonlinear characteristic, comprising:
Key feature image is intercepted from the tile image to be sorted；
Depth nonlinear characteristic is extracted from the key feature image based on feature extraction network.
3. method according to claim 1 or 2, which is characterized in that the feature extraction network include: multiple convolutional layers with
And at least two maximum pond layers, for the extraction depth nonlinear characteristic from the tile image to be sorted.
4. the method according to claim 1, wherein non-linear based on depth described in full connection relationship network query function
Non-linear distance between feature and N number of image classification template, comprising:
The depth nonlinear characteristic is subjected to the extension of N-dimensional degree；
Based on full connection relationship network, the N number of non-linear special feature and N number of image point that the N-dimensional degree extends are calculated
Non-linear distance between class template.
5. according to the method described in claim 4, it is characterized in that, the depth nonlinear characteristic is carried out the extension of N-dimensional degree, packet
It includes:
Depth nonlinear characteristic duplication n times are obtained into N number of depth nonlinear characteristic.
6. the method according to claim 1, wherein the full connection relationship network include: multiple convolutional layers, it is more
A maximum pond layer and multiple full articulamentums, with for calculate the depth nonlinear characteristic and N number of image classification template it
Between non-linear distance.
7. a kind of ceramic tile texture classification apparatus characterized by comprising
Characteristic extracting module, for extracting depth nonlinear characteristic from tile image to be sorted based on feature extraction network；
Distance calculation module, for based on depth nonlinear characteristic and N number of image classification mould described in full connection relationship network query function
Non-linear distance between plate, N > 1；
Category determination module, for dividing image corresponding to the smallest non-linear distance in calculated N number of non-linear distance
Classification of the classification of class template as the tile image to be sorted.
8. device according to claim 7, which is characterized in that the characteristic extracting module is specifically used for:
Key feature image is intercepted from the tile image to be sorted；
Depth nonlinear characteristic is extracted from the key feature image based on feature extraction network.
9. device according to claim 7, which is characterized in that the distance calculation module is specifically used for:
The depth nonlinear characteristic is subjected to the extension of N-dimensional degree；
Based on full connection relationship network, the N number of non-linear special feature and N number of image point that the N-dimensional degree extends are calculated
Non-linear distance between class template.
10. a kind of ceramic tile texture classification apparatus characterized by comprising memory, processor, in which:
The memory is for storing one or more computer instruction, and one or more computer instruction is by the processing
Such as ceramic tile texture classifying method described in any one of claims 1 to 6 is realized when device executes.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910269271.XA CN110163244A (en) | 2019-04-04 | 2019-04-04 | A kind of ceramic tile texture classifying method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910269271.XA CN110163244A (en) | 2019-04-04 | 2019-04-04 | A kind of ceramic tile texture classifying method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110163244A true CN110163244A (en) | 2019-08-23 |
Family
ID=67639066
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910269271.XA Pending CN110163244A (en) | 2019-04-04 | 2019-04-04 | A kind of ceramic tile texture classifying method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110163244A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112560926A (en) * | 2020-12-07 | 2021-03-26 | 杭州聚玻科技有限公司 | Method for automatically determining glass type |
| CN112633393A (en) * | 2020-12-29 | 2021-04-09 | 北京理工大学重庆创新中心 | Automatic classification method and device for ceramic tile textures |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103324759A (en) * | 2013-07-05 | 2013-09-25 | 易翔 | Ceramic tile intelligent recognition device |
| CN106971174A (en) * | 2017-04-24 | 2017-07-21 | 华南理工大学 | A kind of CNN models, CNN training methods and the vein identification method based on CNN |
| CN107730500A (en) * | 2017-11-01 | 2018-02-23 | 广东工业大学 | A kind of ceramic tile texture detection, system, device and readable storage medium storing program for executing |
| CN109064502A (en) * | 2018-07-11 | 2018-12-21 | 西北工业大学 | The multi-source image method for registering combined based on deep learning and artificial design features |
| CN109522493A (en) * | 2018-09-04 | 2019-03-26 | 西安艾润物联网技术服务有限责任公司 | Information-pushing method and Related product |
-
2019
- 2019-04-04 CN CN201910269271.XA patent/CN110163244A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103324759A (en) * | 2013-07-05 | 2013-09-25 | 易翔 | Ceramic tile intelligent recognition device |
| CN106971174A (en) * | 2017-04-24 | 2017-07-21 | 华南理工大学 | A kind of CNN models, CNN training methods and the vein identification method based on CNN |
| CN107730500A (en) * | 2017-11-01 | 2018-02-23 | 广东工业大学 | A kind of ceramic tile texture detection, system, device and readable storage medium storing program for executing |
| CN109064502A (en) * | 2018-07-11 | 2018-12-21 | 西北工业大学 | The multi-source image method for registering combined based on deep learning and artificial design features |
| CN109522493A (en) * | 2018-09-04 | 2019-03-26 | 西安艾润物联网技术服务有限责任公司 | Information-pushing method and Related product |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112560926A (en) * | 2020-12-07 | 2021-03-26 | 杭州聚玻科技有限公司 | Method for automatically determining glass type |
| CN112560926B (en) * | 2020-12-07 | 2023-04-18 | 杭州聚玻科技有限公司 | Method for automatically determining glass type |
| CN112633393A (en) * | 2020-12-29 | 2021-04-09 | 北京理工大学重庆创新中心 | Automatic classification method and device for ceramic tile textures |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110473141A (en) | Image processing method, device, storage medium and electronic equipment | |
| TWI543083B (en) | Method of creating two dimensional code | |
| CN110378235A (en) | A kind of fuzzy facial image recognition method, device and terminal device | |
| JP6731529B1 (en) | Single-pixel attack sample generation method, device, equipment and storage medium | |
| CN105608459B (en) | The dividing method and its device of commodity picture | |
| CN110795976A (en) | Method, device and equipment for training object detection model | |
| CN108682007B (en) | JPEG image resampling automatic detection method based on depth random forest | |
| US11816946B2 (en) | Image based novelty detection of material samples | |
| CN110363091A (en) | Face identification method, device, equipment and storage medium in the case of side face | |
| CN110059677A (en) | Digital table recognition methods and equipment based on deep learning | |
| CN110163095B (en) | Loop detection method, loop detection device and terminal equipment | |
| CN110287806A (en) | A kind of traffic sign recognition method based on improvement SSD network | |
| WO2021248932A1 (en) | Image data processing method and apparatus, device and readable storage medium | |
| CN108492294A (en) | A kind of appraisal procedure and device of image color harmony degree | |
| CN110163244A (en) | A kind of ceramic tile texture classifying method and device | |
| CN110349167A (en) | A kind of image instance dividing method and device | |
| CN112001373A (en) | Article identification method and device and storage medium | |
| CN108764100A (en) | A kind of goal behavior detection method and server | |
| CN115063861A (en) | Model training method, image background similarity judgment method and device | |
| US20210166073A1 (en) | Image generation method and computing device | |
| CN110599487A (en) | Article detection method, apparatus and storage medium | |
| CN112053287B (en) | Image super-resolution method, device and equipment | |
| CN111369489A (en) | Image identification method and device and terminal equipment | |
| CN111726592B (en) | Method and apparatus for obtaining architecture of image signal processor | |
| CN105095889B (en) | Feature extraction, character recognition, engine generates, information determines method and device |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190823 |
|
| RJ01 | Rejection of invention patent application after publication |