CN107729905A - Image information processing method and device - Google Patents
Image information processing method and device Download PDFInfo
- Publication number
- CN107729905A CN107729905A CN201710989393.7A CN201710989393A CN107729905A CN 107729905 A CN107729905 A CN 107729905A CN 201710989393 A CN201710989393 A CN 201710989393A CN 107729905 A CN107729905 A CN 107729905A
- Authority
- CN
- China
- Prior art keywords
- subgraph
- image
- image information
- size
- information
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
- G06V10/449—Biologically inspired filters, e.g. difference of Gaussians [DoG] or Gabor filters
- G06V10/451—Biologically inspired filters, e.g. difference of Gaussians [DoG] or Gabor filters with interaction between the filter responses, e.g. cortical complex cells
- G06V10/454—Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN]
Abstract
The invention discloses a kind of image information processing method and device.Wherein, this method includes：Obtain pending image；The pending image is converted at least one subgraph；According to the size of at least one subgraph, the image information of extraction at least one subgraph.The present invention, which solves, consumes the higher technical problem of hardware cost caused by due to increasing convolution depth.
Description
Technical field
The present invention relates to image processing field, in particular to a kind of image information processing method and device.
Background technology
Now with the burning hot development of artificial intelligence, the process demand increase progressively for image.It is how quick
Processing calculate image, the reduction characteristics of image of smaller error becomes current focus and difficult point.
Current most of manufacturer obtains the more accurate feature of image by increasing convolution depth progressively, however, volume
Product depth is higher, and accuracy is higher, and the hardware cost of consumption is higher.
For it is above-mentioned the problem of, not yet propose effective solution at present.
The content of the invention
It is deep due to increasing convolution at least to solve the embodiments of the invention provide a kind of image information processing method and device
The higher technical problem of hardware cost is consumed caused by degree.
One side according to embodiments of the present invention, there is provided a kind of image information processing method, including：Obtain pending
Image；Above-mentioned pending image is converted at least one subgraph；According to the size of above-mentioned at least one subgraph, in extraction
State the image information of at least one subgraph.
Alternatively, the above-mentioned size according to above-mentioned at least one subgraph, the image of above-mentioned at least one subgraph is extracted
Information, including：Judge whether the size of above-mentioned at least one subgraph is less than predetermined threshold value；If above-mentioned at least one subgraph
Size is less than above-mentioned predetermined threshold value, extracts the profile information of above-mentioned at least one subgraph；If above-mentioned at least one subgraph
Size is more than or equal to above-mentioned predetermined threshold value, extracts the detailed information of above-mentioned at least one subgraph；Wherein, above-mentioned image information bag
Include above-mentioned profile information and above-mentioned detailed information.
Alternatively, the detailed information of the above-mentioned at least one subgraph of said extracted includes：Sampled and extracted using same class two-dimensional
The image information of above-mentioned at least one subgraph；The profile letter of above-mentioned at least one subgraph is rejected from above-mentioned image information
Breath, to obtain detailed information.
Alternatively, according to the size of above-mentioned at least one subgraph, the image information of the above-mentioned at least one subgraph of extraction
Afterwards, the above method also includes：Convolution algorithm is carried out to the image information of above-mentioned at least one subgraph, it is at least one to export
Convolution results；Each convolution results totalling is handled, to export totalling result.
Alternatively, convolution algorithm is carried out to the image information of above-mentioned at least one subgraph, to export at least one convolution
As a result include：Convolution algorithm is carried out to the image information of above-mentioned at least one subgraph respectively by multiple convolution algorithm modules, with
Export above-mentioned at least one convolution results.
Another aspect according to embodiments of the present invention, a kind of image information processing device is additionally provided, including：Obtain single
Member, for obtaining pending image；Converting unit, for above-mentioned pending image to be converted at least one subgraph；Extraction
Unit, for the size according to above-mentioned at least one subgraph, the image information of the above-mentioned at least one subgraph of extraction.
Alternatively, said extracted unit includes：Judge module, for judge above-mentioned at least one subgraph size whether
Less than predetermined threshold value；Extraction module, if the size for above-mentioned at least one subgraph is less than above-mentioned predetermined threshold value, extraction is above-mentioned
The profile information of at least one subgraph；If the size of above-mentioned at least one subgraph is more than or equal to above-mentioned predetermined threshold value, extraction
The detailed information of above-mentioned at least one subgraph；Wherein, above-mentioned image information includes above-mentioned profile information and above-mentioned detailed information.
Another aspect according to embodiments of the present invention, a kind of image information processing device is additionally provided, including：Image generates
Device, for obtaining pending image；Above-mentioned pending image is converted at least one subgraph；Processor, above-mentioned processor
Operation program, wherein, perform following processing step for the data exported from above-mentioned image composer when said procedure is run：Root
According to the size of above-mentioned at least one subgraph, the image information of the above-mentioned at least one subgraph of extraction.
Another aspect according to embodiments of the present invention, additionally provides a kind of storage medium, and above-mentioned storage medium includes storage
Program, wherein, said procedure perform with above-mentioned arbitrary characteristics image information processing method.
Another aspect according to embodiments of the present invention, a kind of processor being additionally provided, above-mentioned processor is used for operation program,
Wherein, said procedure performs the image information processing method with above-mentioned arbitrary characteristics.
In embodiments of the present invention, using the pending image of acquisition；Pending image is converted at least one subgraph；
According to the size of at least one subgraph, the mode of the image information of at least one subgraph is extracted, by by pending image
At least one subgraph that size differs is converted into, and then image information is extracted according to the size of subgraph, has been reached to not
The purpose of different depth convolution is carried out with sized image, it is achieved thereby that reducing the technique effect of hardware consumption cost, and then is solved
Determine and the higher technical problem of hardware cost is consumed caused by due to increasing convolution depth.
Brief description of the drawings
Accompanying drawing described herein is used for providing a further understanding of the present invention, forms the part of the application, this hair
Bright schematic description and description is used to explain the present invention, does not form inappropriate limitation of the present invention.In the accompanying drawings：
Fig. 1 is a kind of schematic flow sheet of optional image information processing method according to embodiments of the present invention；
Fig. 2 is a kind of schematic diagram of optional image information processing method according to embodiments of the present invention；
Fig. 3 is a kind of structural representation of optional image information processing device according to embodiments of the present invention.
Embodiment
In order that those skilled in the art more fully understand the present invention program, below in conjunction with the embodiment of the present invention
Accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is only
The embodiment of a part of the invention, rather than whole embodiments.Based on the embodiment in the present invention, ordinary skill people
The every other embodiment that member is obtained under the premise of creative work is not made, it should all belong to the model that the present invention protects
Enclose.
It should be noted that term " first " in description and claims of this specification and above-mentioned accompanying drawing, "
Two " etc. be for distinguishing similar object, without for describing specific order or precedence.It should be appreciated that so use
Data can exchange in the appropriate case, so as to embodiments of the invention described herein can with except illustrating herein or
Order beyond those of description is implemented.In addition, term " comprising " and " having " and their any deformation, it is intended that cover
Cover it is non-exclusive include, be not necessarily limited to for example, containing the process of series of steps or unit, method, system, product or equipment
Those steps or unit clearly listed, but may include not list clearly or for these processes, method, product
Or the intrinsic other steps of equipment or unit.
Embodiment 1
According to embodiments of the present invention, there is provided a kind of embodiment of the method for image information processing method, it is necessary to explanation,
It can be performed the step of the flow of accompanying drawing illustrates in the computer system of such as one group computer executable instructions, and
And although showing logical order in flow charts, in some cases, can be with different from order execution institute herein
The step of showing or describing.
Fig. 1 is image information processing method according to embodiments of the present invention, as shown in figure 1, this method comprises the following steps：
Step S102, obtain pending image.
In the application above-mentioned steps S102, for neutral net convolution, the input of image can be divided division
For different regions, the extraction to image information is gradually realized by carrying out convolution to different locals, to input maximum
Exemplified by a width of 8 pixel datas, when carrying out 1 × 1 convolution algorithm to 8 pixel datas using 83 × 3 convolution units, respectively
Only 1 multiplier substantially carries out convolution algorithm in 3 × 3 convolution units, therefore other multipliers do not apply to, and this is caused
Hardware resource waste.And the image information processing method of the present embodiment, by the way that the image of different resolution level is entered into row information
Extraction, by the way that an original pyramid by N grades is changed, N number of various sizes of small image can be gradually transformed into,
Then (picture position, image information) is extracted in the sampling for two-dimensional signal being carried out to different small images, lifts making for convolution unit
With rate.
Step S104, pending image is converted at least one subgraph.
In the application above-mentioned steps S104, after pending image is got, pending image is converted at least one
Individual subgraph.Wherein, pending image is converted at least one subgraph, including：Pending image is subjected to N grade contractings
It is small, generate N number of various sizes of subgraph.
Wherein, pending image is carried out to the method for N grade diminutions can include carrying out reducing N on year-on-year basis by processing image
It is secondary, obtain N number of various sizes of subgraph.
Step S106, according to the size of at least one subgraph, extract the image information of at least one subgraph.
In the application above-mentioned steps S106, by the way that the image of different resolution level is carried out into information extraction, by by one
The original pyramid change by N grades is opened, N number of various sizes of subgraph can be gradually transformed into, then to different
Subgraph carries out the sampling extraction (picture position, image information) of two-dimensional signal, due to being taken turns to the subgraph of small size
Wide information extraction, the profile of image can be rejected in for large-sized subgraph, and details is retained, it is right
Large-sized subgraph sampled with class two-dimensional and extracted.In the present embodiment, refer to class two-dimensional sampling to current pixel point
The method that position is recorded, mapped with detailed information.
Specifically, sampled with class two-dimensional：The detailed information or profile information that the image of general different resolution is included be
It is different, it is relatively more for large-sized image (image of i.e. big resolution ratio) detailed information, and for small size image (i.e.
The image of small resolution ratio) general profile information is more comprehensive, such as leaf, the train of thought of the image of big resolution ratio generally for leaf
Details is clearer, and the information that the image of small resolution ratio contains to the profile of leaf is relatively more.For different resolution ratio
Image can be stored by being sampled to image detail to generate a two-dimentional function f (x, y), wherein x, y representative images
Position, f (x, y) represent detailed information.
As a kind of optional implementation, according to the size of at least one subgraph, at least one subgraph is extracted
Image information, including：Judge whether the size of at least one subgraph is less than predetermined threshold value；If the size of at least one subgraph
Less than predetermined threshold value, the profile information of at least one subgraph is extracted；If the size of at least one subgraph is more than or equal to default
Threshold value, extract the detailed information of at least one subgraph；Wherein, image information includes profile information and detailed information.
Alternatively, extracting the detailed information of at least one subgraph includes：Sampled using same class two-dimensional described in extraction at least
The image information of one subgraph；The profile information of at least one subgraph is rejected from image information, to obtain detailed information.
Wherein, the profile information of at least one subgraph is rejected from image information to be included：Believed by wave filter from image
The profile information of at least one subgraph, and each pixel of at least one subgraph to rejecting profile information are rejected in breath
The position of point is recorded with detailed information, mapped.
As shown in Fig. 2 when image information is extracted, for n-th layer subgraph, the figure of N-1 straton images can be deleted
As information, the tomographic image information is obtained.For on various sizes of subgraph, the subgraph of different sizes is taken
The information of band is different, it is however generally that, the subgraph of small size can carry the big profile information of image, and large-sized subgraph
The detailed information of image then can be more carried, difference can be tentatively obtained by carrying out detail extraction to various sizes of subgraph
Details fragment, neutral net convolution then is carried out to different details fragment and realizes image zooming-out.
Alternatively, profile information includes at least one of：Shape facility and locus feature；Detailed information include with
It is at least one lower：Color characteristic, textural characteristics.
By above-mentioned steps, pending image is converted at least one subgraph that size differs, and then according to son
The size extraction image information of image, has reached the purpose that different sized images are carried out with different depth convolution, it is achieved thereby that
The technique effect of hardware consumption cost is reduced, and then it is higher to solve consumption hardware cost caused by due to increasing convolution depth
Technical problem.
As a kind of optional implementation, according to the size of at least one subgraph, at least one subgraph is extracted
After image information, method also includes：Convolution algorithm is carried out to the image information of at least one subgraph, it is at least one to export
Convolution results；Each convolution results totalling is handled, to export totalling result.
Wherein, performed by totalling processing unit and the processing of each convolution results totalling (is e.g. added and added up), with output
The step of adding up result.
Alternatively, convolution algorithm is carried out to the image information of at least one subgraph, to export at least one convolution results
Including：Convolution algorithm is carried out to the image information of at least one subgraph respectively by multiple convolution algorithm modules, to export at least
One convolution results.Multiple convolution algorithm modules can reach raising with the image information of at least one subgraph of parallel processing
Treatment effeciency, the purpose of enhanced performance.The image information of different subgraphs can be by different convolution algorithm resume modules.
It should be noted that can be in such as one group of computer executable instructions the flow of accompanying drawing illustrates the step of
Performed in computer system, although also, show logical order in flow charts, in some cases, can be with not
The order being same as herein performs shown or described step.
Embodiment 2
The embodiment of the present invention additionally provides a kind of image information processing device.It should be noted that the image of the embodiment
Information processor can be used for the image information processing method for performing the embodiment of the present invention.
Fig. 3 is a kind of schematic diagram of image information processing device according to embodiments of the present invention.As shown in figure 3, this is upper
Machine includes：Acquiring unit 20, conversion unit 22 and extraction unit 24.
Acquiring unit 20, for obtaining pending image；
Converting unit 22, for pending image to be converted at least one subgraph；
Extraction unit 24, for the size according at least one subgraph, extract the image information of at least one subgraph.
Alternatively, extraction unit 24 includes：Judge module, it is pre- whether the size for judging at least one subgraph is less than
If threshold value；Extraction module, if the size at least one subgraph is less than predetermined threshold value, extract the wheel of at least one subgraph
Wide information；If the size of at least one subgraph is more than or equal to predetermined threshold value, the detailed information of at least one subgraph is extracted；Its
In, image information includes profile information and detailed information.
Alternatively, extraction module is used to perform the detailed information that following steps extract at least one subgraph：Using similar
The image information of at least one subgraph is extracted in two dimension sampling；The profile of at least one subgraph is rejected from image information
Information, to obtain detailed information.
Alternatively, image information processing device also includes：Processing unit, for the image information at least one subgraph
Convolution algorithm is carried out, to export at least one convolution results；Each convolution results totalling is handled, to export totalling result.
Alternatively, processing unit includes：Multiple convolution algorithm modules, for believing respectively the image of at least one subgraph
Breath carries out convolution algorithm, to export at least one convolution results.
In embodiments of the present invention, using the pending image of acquisition；Pending image is converted at least one subgraph；
According to the size of at least one subgraph, the mode of the image information of at least one subgraph is extracted, by by pending image
At least one subgraph that size differs is converted into, and then image information is extracted according to the size of subgraph, has been reached to not
The purpose of different depth convolution is carried out with sized image, it is achieved thereby that reducing the technique effect of hardware consumption cost, and then is solved
Determine and the higher technical problem of hardware cost is consumed caused by due to increasing convolution depth.
Embodiment 3
The embodiment of the present invention additionally provides a kind of image information processing device.The image information processing device includes：Image
Maker, for obtaining pending image；Pending image is converted at least one subgraph；Processor, processor operation
Program, wherein, perform following processing step for the data exported from image composer when program is run：According at least one son
The size of image, extract the image information of at least one subgraph.
The embodiment of the present invention additionally provides a kind of storage medium, and storage medium includes the program of storage, wherein, program performs
Image information processing method with above-mentioned arbitrary characteristics.
The embodiment of the present invention additionally provides a kind of processor, and processor is used for operation program, wherein, program is performed with upper
State the image information processing method of arbitrary characteristics.
The embodiments of the present invention are for illustration only, do not represent the quality of embodiment.
In the above embodiment of the present invention, the description to each embodiment all emphasizes particularly on different fields, and does not have in some embodiment
The part of detailed description, it may refer to the associated description of other embodiment.
In several embodiments provided herein, it should be understood that disclosed technology contents, others can be passed through
Mode is realized.Wherein, device embodiment described above is only schematical, such as the division of the unit, Ke Yiwei
A kind of division of logic function, can there is an other dividing mode when actually realizing, for example, multiple units or component can combine or
Person is desirably integrated into another system, or some features can be ignored, or does not perform.Another, shown or discussed is mutual
Between coupling or direct-coupling or communication connection can be INDIRECT COUPLING or communication link by some interfaces, unit or module
Connect, can be electrical or other forms.
The unit illustrated as separating component can be or may not be physically separate, show as unit
The part shown can be or may not be physical location, you can with positioned at a place, or can also be distributed to multiple
On unit.Some or all of unit therein can be selected to realize the purpose of this embodiment scheme according to the actual needs.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, can also
That unit is individually physically present, can also two or more units it is integrated in a unit.Above-mentioned integrated list
Member can both be realized in the form of hardware, can also be realized in the form of SFU software functional unit.
If the integrated unit is realized in the form of SFU software functional unit and is used as independent production marketing or use
When, it can be stored in a computer read/write memory medium.Based on such understanding, technical scheme is substantially
The part to be contributed in other words to prior art or all or part of the technical scheme can be in the form of software products
Embody, the computer software product is stored in a storage medium, including some instructions are causing a computer
Equipment (can be personal computer, server or network equipment etc.) perform each embodiment methods described of the present invention whole or
Part steps.And foregoing storage medium includes：USB flash disk, read-only storage (ROM, Read-Only Memory), arbitrary access are deposited
Reservoir (RAM, Random Access Memory), mobile hard disk, magnetic disc or CD etc. are various can be with store program codes
Medium.
Described above is only the preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications also should
It is considered as protection scope of the present invention.
Claims (10)
- A kind of 1. image information processing method, it is characterised in that including：Obtain pending image；The pending image is converted at least one subgraph；According to the size of at least one subgraph, the image information of extraction at least one subgraph.
- 2. according to the method for claim 1, it is characterised in that the size according at least one subgraph, carry The image information of at least one subgraph is taken, including：Judge whether the size of at least one subgraph is less than predetermined threshold value；If the size of at least one subgraph is less than the predetermined threshold value, the profile letter of at least one subgraph is extracted Breath；If the size of at least one subgraph is more than or equal to the predetermined threshold value, the thin of at least one subgraph is extracted Save information；Wherein, described image information includes the profile information and the detailed information.
- 3. according to the method for claim 2, it is characterised in that the detailed information of extraction at least one subgraph Including：The image information of at least one subgraph is extracted using the sampling of same class two-dimensional；The profile information of at least one subgraph is rejected from described image information, to obtain the detailed information.
- 4. according to the method for claim 1, it is characterised in that according to the size of at least one subgraph, extract institute After the image information for stating at least one subgraph, methods described also includes：Convolution algorithm is carried out to the image information of at least one subgraph, to export at least one convolution results；Each convolution results totalling is handled, to export totalling result.
- 5. according to the method for claim 4, it is characterised in that the image information of at least one subgraph is rolled up Product computing, is included with exporting at least one convolution results：Convolution algorithm is carried out to the image information of at least one subgraph respectively by multiple convolution algorithm modules, to export State at least one convolution results.
- A kind of 6. image information processing device, it is characterised in that including：Acquiring unit, for obtaining pending image；Converting unit, for the pending image to be converted at least one subgraph；Extraction unit, for the size according at least one subgraph, the image for extracting at least one subgraph is believed Breath.
- 7. device according to claim 6, it is characterised in that the extraction unit includes：Judge module, for judging whether the size of at least one subgraph is less than predetermined threshold value；Extraction module, if the size at least one subgraph is less than the predetermined threshold value, extraction is described at least one The profile information of subgraph；If the size of at least one subgraph is more than or equal to the predetermined threshold value, described in extraction at least The detailed information of one subgraph；Wherein, described image information includes the profile information and the detailed information.
- A kind of 8. image information processing device, it is characterised in that including：Image composer, for obtaining pending image；The pending image is converted at least one subgraph；Processor, the processor operation program, wherein, number when described program is run for being exported from described image maker According to the following processing step of execution：According to the size of at least one subgraph, the image of extraction at least one subgraph Information.
- A kind of 9. storage medium, it is characterised in that the storage medium includes the program of storage, wherein, described program right of execution Profit requires the image information processing method described in any one in 1 to 5.
- A kind of 10. processor, it is characterised in that the processor is used for operation program, wherein, right of execution when described program is run Profit requires the image information processing method described in any one in 1 to 5.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710989393.7A CN107729905A (en) | 2017-10-19 | 2017-10-19 | Image information processing method and device |
| PCT/CN2018/099598 WO2019076110A1 (en) | 2017-10-19 | 2018-08-09 | Image information processing method and apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710989393.7A CN107729905A (en) | 2017-10-19 | 2017-10-19 | Image information processing method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107729905A true CN107729905A (en) | 2018-02-23 |
Family
ID=61213267
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710989393.7A Pending CN107729905A (en) | 2017-10-19 | 2017-10-19 | Image information processing method and device |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN107729905A (en) |
| WO (1) | WO2019076110A1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019076110A1 (en) * | 2017-10-19 | 2019-04-25 | 格力电器（武汉）有限公司 | Image information processing method and apparatus |
| WO2021018199A1 (en) * | 2019-07-31 | 2021-02-04 | 瀚博半导体(上海)有限公司 | Convolutional neural network-based image processing method and apparatus |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103218831A (en) * | 2013-04-21 | 2013-07-24 | 北京航空航天大学 | Video moving target classification and identification method based on outline constraint |
| CN104835130A (en) * | 2015-04-17 | 2015-08-12 | 北京联合大学 | Multi-exposure image fusion method |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3747737B2 (en) * | 2000-05-10 | 2006-02-22 | 日本電気株式会社 | Wide-area fine image generation method and system, and computer-readable recording medium |
| CN101882305B (en) * | 2010-06-30 | 2014-02-05 | 中山大学 | Method for enhancing image |
| CN106296600B (en) * | 2016-08-01 | 2018-12-21 | 大连理工大学 | A kind of contrast enhancement process decomposed based on wavelet image |
| CN107729905A (en) * | 2017-10-19 | 2018-02-23 | 珠海格力电器股份有限公司 | Image information processing method and device |
-
2017
- 2017-10-19 CN CN201710989393.7A patent/CN107729905A/en active Pending
-
2018
- 2018-08-09 WO PCT/CN2018/099598 patent/WO2019076110A1/en active Application Filing
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103218831A (en) * | 2013-04-21 | 2013-07-24 | 北京航空航天大学 | Video moving target classification and identification method based on outline constraint |
| CN104835130A (en) * | 2015-04-17 | 2015-08-12 | 北京联合大学 | Multi-exposure image fusion method |
Non-Patent Citations (2)
| Title |
|---|
| 刘俊 等: "《高分辨率遥感影像城市道路提取方法研究》", 《海洋测绘》 * |
| 谭贞刚: "《基于多分辨率卷积网络的交通标志检测与识别》", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019076110A1 (en) * | 2017-10-19 | 2019-04-25 | 格力电器（武汉）有限公司 | Image information processing method and apparatus |
| WO2021018199A1 (en) * | 2019-07-31 | 2021-02-04 | 瀚博半导体(上海)有限公司 | Convolutional neural network-based image processing method and apparatus |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2019076110A1 (en) | 2019-04-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104834933B (en) | A kind of detection method and device in saliency region | |
| CN103247036B (en) | Many exposure images fusion method and device | |
| CN108537292A (en) | Semantic segmentation network training method, image, semantic dividing method and device | |
| CN113822209B (en) | Hyperspectral image recognition method and device, electronic equipment and readable storage medium | |
| CN107358586A (en) | A kind of image enchancing method, device and equipment | |
| CN109558806A (en) | The detection method and system of high score Remote Sensing Imagery Change | |
| CN108711137A (en) | A kind of image color expression pattern moving method based on depth convolutional neural networks | |
| CN105118048A (en) | Method and device for identifying copying certificate image | |
| CN108154105A (en) | Aquatic organism detects and recognition methods, device, server and terminal device | |
| CN109684476B (en) | Text classification method, text classification device and terminal equipment | |
| CN108230243B (en) | Background blurring method based on salient region detection model | |
| CN108319894A (en) | Fruit recognition methods based on deep learning and device | |
| CN104285239A (en) | Image processing device, image processing method, program, print medium, and recording medium | |
| CN107766932A (en) | Image processing method and device based on neutral net | |
| CN107749044A (en) | The pond method and device of image information | |
| CN109784372A (en) | A kind of objective classification method based on convolutional neural networks | |
| CN109919160A (en) | Method for recognizing verification code, device, terminal and storage medium | |
| CN109344888A (en) | A kind of image-recognizing method based on convolutional neural networks, device and equipment | |
| CN114529490B (en) | Data processing method, device, equipment and readable storage medium | |
| CN109918204A (en) | Data processing system and method | |
| CN106709431A (en) | Iris recognition method and device | |
| CN106599878A (en) | Face reconstruction correction method and device based on deep learning | |
| CN113919443A (en) | Tobacco maturity state probability calculation method based on image analysis | |
| CN110738660A (en) | Spine CT image segmentation method and device based on improved U-net | |
| CN107729905A (en) | Image information processing 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: 20180223 |
|
| RJ01 | Rejection of invention patent application after publication |