CN109063701A - Labeling method, device, equipment and the storage medium of target in a kind of infrared image - Google Patents
Labeling method, device, equipment and the storage medium of target in a kind of infrared image Download PDFInfo
- Publication number
- CN109063701A CN109063701A CN201810897531.3A CN201810897531A CN109063701A CN 109063701 A CN109063701 A CN 109063701A CN 201810897531 A CN201810897531 A CN 201810897531A CN 109063701 A CN109063701 A CN 109063701A
- Authority
- CN
- China
- Prior art keywords
- target
- infrared image
- algorithm
- indicia framing
- visible images
- 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
- 238000002372 labelling Methods 0.000 title claims abstract description 38
- 238000001514 detection method Methods 0.000 claims abstract description 46
- 238000009432 framing Methods 0.000 claims abstract description 45
- 238000007500 overflow downdraw method Methods 0.000 claims abstract description 10
- 230000001360 synchronised effect Effects 0.000 claims abstract description 10
- 238000013528 artificial neural network Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 7
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 25
- 241000282326 Felis catus Species 0.000 description 5
- 230000006870 function Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 239000012141 concentrate Substances 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000004297 night vision Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 238000001931 thermography Methods 0.000 description 1
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
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/143—Sensing or illuminating at different wavelengths
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/255—Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/09—Recognition of logos
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
This application discloses labeling method, device, equipment and the storage mediums of target in a kind of infrared image, this method comprises: collecting and recording synchronous visible images and infrared image using double light fusion methods；The target in the visible images is marked by indicia framing using trained algorithm of target detection；After the visible images and the infrared image are registrated, the indicia framing in the visible images is transformed in the infrared image, to complete the label to target in the infrared image.The time of a large amount of handmarking's infrared images is saved in this way, while is avoided to a certain extent because the influence of infrared image itself causes the problem of being easy marked erroneous or deviation, and increases the type that can mark target in infrared image.
Description
Technical field
The present invention relates to field of target recognition, more particularly to the labeling method of target in a kind of infrared image, device, set
Standby and storage medium.
Background technique
Currently, it is all that artificial mode carries out and concentrates on visible regime that image object, which is marked, both at home and abroad.Example
Such as famous ImageNet, is employed by way of crowdsourcing close to 50,000 people and about 1,000,000,000 photos are marked.Target is known
Other technology, the effect of algorithm progress target identification is better after the database of tag image is bigger, and manually carries out image tagged
Again be an especially time consuming process, at present for infrared image label database it is considerably less.
Visible regime is all concentrated on further for the research overwhelming majority of target identification.Relative to visible light, infrared heat
Imaging technique can play the role of night vision in the case where not having light at such as night, such as in the auxiliary field of vehicle, install infrared
Thermal imaging can detect the targets such as pedestrian and vehicle on pitch-dark road, increase the safety of driving.However visible regime
The numerous studies of target identification not can be used directly in infrared regime yet.
The existing mark mode for infrared image be all with handmarking's mode as visible images marking class, together
The contrast of Shi Yinwei infrared image is lower relative to visible images, and image texture details does not have visible light to enrich, image side
Edge does not have that visible light is precipitous, relatively unobvious, therefore the label of infrared image is more difficult relative to visible light, expends the time more
Mostly and it is easy to generate deviation because edge is not seen.Some temperature are with the much the same object of environment in infrared image simultaneously
Because gray scale is not that very high reason is difficult to be gone out by artificial accurate marker, the label of object is also mainly collected in infrared image
In the object relatively high in temperature.
Therefore, how to realize the unartificial label to target in infrared image, be that those skilled in the art are urgently to be resolved
Technical problem.
Summary of the invention
In view of this, the purpose of the present invention is to provide the labeling method of target in a kind of infrared image, device, equipment and
Storage medium can save the time of a large amount of handmarking's infrared images, improve label accuracy.Its concrete scheme is as follows:
The labeling method of target in a kind of infrared image, comprising:
Synchronous visible images and infrared image are collected and recorded using double light fusion methods；
The target in the visible images is marked by indicia framing using trained algorithm of target detection；
After the visible images and the infrared image are registrated, the indicia framing in the visible images is become
It changes in the infrared image, to complete the label to target in the infrared image.
Preferably, in above-mentioned infrared image provided in an embodiment of the present invention in the labeling method of target, using training
Before the target in the visible images is marked in good algorithm of target detection, further includes:
Using a variety of different visible light sample image databases as training set；
It is trained to different neural network structures is input to after training set progress target labels, trains difference
Algorithm of target detection.
Preferably, in above-mentioned infrared image provided in an embodiment of the present invention in the labeling method of target, using training
While the target in the visible images is marked by indicia framing for good algorithm of target detection, further includes:
The indicia framing of debug by way of intersecting and comparing.
Preferably, in above-mentioned infrared image provided in an embodiment of the present invention in the labeling method of target, by intersecting ratio
Pair mode debug indicia framing, specifically include:
Calculate the friendship for the indicia framing framework that each algorithm of target detection marks and ratio；
Judge the friendship and than whether being greater than given threshold；If so, determining that each algorithm of target detection detected
The same target；If it is not, then determining that each algorithm of target detection does not detect the same target, give up the target；
For the same target detected, the indicia framing framework that each algorithm of target detection marks is done non-very big
Value inhibits, and determines unique indicia framing.
The embodiment of the invention also provides a kind of labelling apparatus of target in infrared image, comprising:
Image capture module, for collecting and recording synchronous visible images and infrared image using double light fusion methods；
Target label module, for being passed through using trained algorithm of target detection to the target in the visible images
Indicia framing is marked；
Indicia framing conversion module can described in after being registrated the visible images and the infrared image
Indicia framing in light-exposed image transforms in the infrared image, to complete the label to target in the infrared image.
Preferably, in above-mentioned infrared image provided in an embodiment of the present invention in the labelling apparatus of target, further includes:
Algorithm training module, for using a variety of different visible light sample image databases as training set；To the instruction
Different neural network structures is input to after white silk collection progress target labels to be trained, and trains different algorithm of target detection.
Preferably, in above-mentioned infrared image provided in an embodiment of the present invention in the labelling apparatus of target, further includes:
Error exception module, the indicia framing for the debug by way of intersecting and comparing.
The embodiment of the invention also provides a kind of marking arrangement of target in infrared image, including processor and memory,
Wherein, it realizes when the processor executes the computer program saved in the memory as provided in an embodiment of the present invention above-mentioned
The labeling method of target in infrared image.
The embodiment of the invention also provides a kind of computer readable storage mediums, for storing computer program, wherein institute
It states and realizes the label side such as target in above-mentioned infrared image provided in an embodiment of the present invention when computer program is executed by processor
Method.
Labeling method, device, equipment and the storage medium of target, this method in a kind of infrared image provided by the present invention
It include: to collect and record synchronous visible images and infrared image using double light fusion methods；Using trained target detection
The target in the visible images is marked by indicia framing for algorithm；By the visible images and the infrared image
After being registrated, the indicia framing in the visible images is transformed in the infrared image, to complete to the infrared figure
The label of target as in.
The present invention utilizes the algorithm of target detection under visible light, to mark the target under infrared image.Especially by simultaneously
Visible images and infrared image are acquired, and target detection label is carried out using existing algorithm of target detection under visible light,
It completes to the automatic label of infrared picture, does not need that manpower is dynamic to be gone to mark a sheet by a sheet infrared picture, save so a large amount of artificial
The time of infrared image is marked, while is avoided to a certain extent because the influence of infrared image itself causes to be easy marked erroneous
Or the problem of deviation, and increase the type that target can be marked in infrared image.
Detailed description of the invention
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, the accompanying drawings in the following description is only this
The embodiment of invention for those of ordinary skill in the art without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is the flow chart of the labeling method of target in infrared image provided in an embodiment of the present invention；
Fig. 2 is the structural schematic diagram of the labelling apparatus of target in infrared image provided in an embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The present invention provides a kind of labeling method of target in infrared image, as shown in Figure 1, comprising the following steps:
S101, synchronous visible images and infrared image are collected and recorded using double light fusion methods；
S102, the target in visible images is marked by indicia framing using trained algorithm of target detection；
S103, after will be seen that light image and infrared image are registrated, the indicia framing in visible images is transformed to red
In outer image, to complete the label to target in infrared image.
In above-mentioned infrared image provided in an embodiment of the present invention in the labeling method of target, first using double light fusion methods
Collect and record synchronous visible images and infrared image；Then using trained algorithm of target detection to visible images
In target be marked by indicia framing；Finally it will be seen that after light image and infrared image be registrated, visible images
In indicia framing transform in infrared image, to complete to the label of target in infrared image.The mesh under visible light is utilized in this way
Detection algorithm is marked, to mark the target under infrared image, especially by acquiring visible images and infrared image simultaneously, and can
It is light-exposed lower using existing algorithm of target detection progress target detection label, it completes not needing the automatic label of infrared picture
Manpower is dynamic to be gone to mark a sheet by a sheet infrared picture, saves the time of a large amount of handmarking's infrared images, while keeping away to a certain extent
Exempt from because the influence of infrared image itself causes the problem of being easy marked erroneous or deviation, and increases in infrared image
The type of target can be marked.
Further, in the specific implementation, in above-mentioned infrared image provided in an embodiment of the present invention target label side
In method, it is being marked to the target in visible images using trained algorithm of target detection executing step S102
Before, it can also include: using a variety of different visible light sample image databases as training set；Target labels are carried out to training set
It is input to different neural network structures afterwards to be trained, trains different algorithm of target detection.
Specifically, the different target detection algorithm trained with different training set and different neural network structures is
Detect the algorithms of different of specific type target in the picture, algorithm of target detection include traditional algorithm (such as HOG+SVM) or
Deep learning algorithm (such as RCNN series, YOLO, SSD).It in the training process, can be for example, by the data of 10,000 cat pictures
Focusing study goes out the feature of cat, makes it possible to detect the cat in new picture；The source of this ten thousand cat pictures has very much, such as
The set of data samples such as ImageNet, MS-COCO, PASCAL-VOC, the neural network structure use when target detection are also
Different, such as AlexNet, ResNet, DarkNet etc., therefore have many different detection algorithms.Different detection algorithms
Effect be also different, such as the cat in same picture, some algorithms can detect that some can't detect, for example detect
When the indicia framing that marks, the label of some labels accurately having has deviation.
Further, in the specific implementation, in above-mentioned infrared image provided in an embodiment of the present invention target label
In method, indicia framing has deviation, mistake in order to prevent, is executing step S102 using trained algorithm of target detection to visible
It can also include: the debug by way of intersecting and comparing while target in light image is marked by indicia framing
Indicia framing.
Specifically, intersect compare by way of debug indicia framing, can specifically include: calculating each target first
The friendship for the indicia framing framework that detection algorithm marks and than (IoU, Intersection-over-Union)；Judgement is handed over and ratio is
It is no to be greater than given threshold；If so, determining that each algorithm of target detection detected the same target；If it is not, then determining each target
Detection algorithm does not detect the same target, give up target (because would rather without label target, can not marked erroneous mesh
Mark)；Then, for the same target detected, non-maximum is done to the indicia framing framework that each algorithm of target detection marks
Inhibit (Non-Maximum Suppression, NMS), determines unique indicia framing.
Based on the same inventive concept, the embodiment of the invention also provides a kind of labelling apparatus of target in infrared image, by
The labeling method of target in the labelling apparatus of target solves the problems, such as in the infrared image principle and a kind of aforementioned infrared image
It is similar, therefore the implementation of the labelling apparatus of target may refer to the reality of the labeling method of target in infrared image in the infrared image
It applies, overlaps will not be repeated.
In the specific implementation, in infrared image provided in an embodiment of the present invention target labelling apparatus, as shown in Fig. 2, tool
Body includes:
Image capture module 11, for collecting and recording synchronous visible images and infrared figure using double light fusion methods
Picture；
Target label module 12, for passing through mark to the target in visible images using trained algorithm of target detection
Note frame is marked；
Indicia framing conversion module 13, for it will be seen that light image and infrared image are registrated after, in visible images
Indicia framing transform in infrared image, to complete to the label of target in infrared image.
In above-mentioned infrared image provided in an embodiment of the present invention in the labelling apparatus of target, above three mould can be passed through
The interaction of block marks the target under infrared image using the algorithm of target detection under visible light, saves a large amount of artificial
The time of infrared image is marked, while is avoided to a certain extent because the influence of infrared image itself causes to be easy marked erroneous
Or the problem of deviation, and increase the type that target can be marked in infrared image.
Further, in the specific implementation, in above-mentioned infrared image provided in an embodiment of the present invention target label side
In method, further includes: algorithm training module, for using a variety of different visible light sample image databases as training set；To instruction
Different neural network structures is input to after white silk collection progress target labels to be trained, and trains different algorithm of target detection.
Further, in the specific implementation, in above-mentioned infrared image provided in an embodiment of the present invention target label
In method, indicia framing has deviation, mistake in order to prevent, further includes: error exception module, for being arranged by way of intersecting and comparing
Except the indicia framing of mistake.
Corresponding contents disclosed in previous embodiment can be referred to about the more specifical course of work of above-mentioned modules,
This is no longer repeated.
Correspondingly, the embodiment of the invention also discloses a kind of marking arrangement of target in infrared image, including processor and
Memory；Wherein, infrared image disclosed in previous embodiment is realized when processor executes the computer program saved in memory
The labeling method of middle target.
It can be with reference to corresponding contents disclosed in previous embodiment, herein no longer about the more specifical process of the above method
It is repeated.
Further, the invention also discloses a kind of computer readable storage mediums, for storing computer program；It calculates
Machine program realizes the labeling method of target in aforementioned disclosed infrared image when being executed by processor.
It can be with reference to corresponding contents disclosed in previous embodiment, herein no longer about the more specifical process of the above method
It is repeated.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with it is other
The difference of embodiment, same or similar part may refer to each other between each embodiment.For being filled disclosed in embodiment
It sets, for equipment, storage medium, since it is corresponded to the methods disclosed in the examples, so be described relatively simple, correlation
Place is referring to method part illustration.
Professional further appreciates that, unit described in conjunction with the examples disclosed in the embodiments of the present disclosure
And algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware and
The interchangeability of software generally describes each exemplary composition and step according to function in the above description.These
Function is implemented in hardware or software actually, the specific application and design constraint depending on technical solution.Profession
Technical staff can use different methods to achieve the described function each specific application, but this realization is not answered
Think beyond scope of the present application.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can directly be held with hardware, processor
The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
Labeling method, device, equipment and the storage medium of target in a kind of infrared image provided in an embodiment of the present invention, should
Method includes: to collect and record synchronous visible images and infrared image using double light fusion methods；Using trained target
The target in visible images is marked by indicia framing for detection algorithm；It will be seen that light image and infrared image are registrated
Afterwards, the indicia framing in visible images is transformed in infrared image, to complete the label to target in infrared image.It saves in this way
The time of a large amount of handmarking's infrared images has been saved, while has been avoided to a certain extent because the influence of infrared image itself causes
The problem of being easy marked erroneous or deviation, and increase the type that target can be marked in infrared image.
Finally, it is to be noted that, herein, relational terms be used merely to by an entity or operation with it is another
A entity or operation distinguish, and without necessarily requiring or implying between these entities or operation, there are any this actual
Relationship or sequence.Moreover, the terms "include", "comprise" or any other variant thereof is intended to cover non-exclusive inclusion,
So that the process, method, article or equipment for including a series of elements not only includes those elements, but also including not having
The other element being expressly recited, or further include for elements inherent to such a process, method, article, or device.Do not having
There is the element limited in the case where more limiting by sentence "including a ...", it is not excluded that in the mistake including the element
There is also other identical elements in journey, method, article or equipment.
The labeling method of target, device, equipment and storage medium in infrared image provided by the present invention are carried out above
It is discussed in detail, used herein a specific example illustrates the principle and implementation of the invention, above embodiments
Explanation be merely used to help understand method and its core concept of the invention；At the same time, for those skilled in the art,
According to the thought of the present invention, there will be changes in the specific implementation manner and application range, in conclusion in this specification
Appearance should not be construed as limiting the invention.
Claims (9)
1. the labeling method of target in a kind of infrared image characterized by comprising
Synchronous visible images and infrared image are collected and recorded using double light fusion methods；
The target in the visible images is marked by indicia framing using trained algorithm of target detection；
After the visible images and the infrared image are registrated, the indicia framing in the visible images is transformed to
In the infrared image, to complete the label to target in the infrared image.
2. the labeling method of target in infrared image according to claim 1, which is characterized in that using trained mesh
Before the target in the visible images is marked in mark detection algorithm, further includes:
Using a variety of different visible light sample image databases as training set；
It is trained to different neural network structures is input to after training set progress target labels, trains different mesh
Mark detection algorithm.
3. the labeling method of target in infrared image according to claim 2, which is characterized in that using trained mesh
While the target in the visible images is marked by indicia framing for mark detection algorithm, further includes:
The indicia framing of debug by way of intersecting and comparing.
4. the labeling method of target in infrared image according to claim 3, which is characterized in that by intersecting the side compared
The indicia framing of formula debug, specifically includes:
Calculate the friendship for the indicia framing framework that each algorithm of target detection marks and ratio；
Judge the friendship and than whether being greater than given threshold；If so, it is same to determine that each algorithm of target detection detected
A target；If it is not, then determining that each algorithm of target detection does not detect the same target, give up the target；
For the same target detected, non-maximum is done to the indicia framing framework that each algorithm of target detection marks and is pressed down
System, determines unique indicia framing.
5. the labelling apparatus of target in a kind of infrared image characterized by comprising
Image capture module, for collecting and recording synchronous visible images and infrared image using double light fusion methods；
Target label module, for passing through label to the target in the visible images using trained algorithm of target detection
Frame is marked；
Indicia framing conversion module, after being registrated the visible images and the infrared image, the visible light
Indicia framing in image transforms in the infrared image, to complete the label to target in the infrared image.
6. the labelling apparatus of target in infrared image according to claim 5, which is characterized in that further include:
Algorithm training module, for using a variety of different visible light sample image databases as training set；To the training set
Different neural network structures is input to after progress target labels to be trained, and trains different algorithm of target detection.
7. the labelling apparatus of target in infrared image according to claim 6, which is characterized in that further include:
Error exception module, the indicia framing for the debug by way of intersecting and comparing.
8. the marking arrangement of target in a kind of infrared image, which is characterized in that including processor and memory, wherein the place
Reason device realizes such as the described in any item infrared images of Claims 1-4 when executing the computer program saved in the memory
The labeling method of middle target.
9. a kind of computer readable storage medium, which is characterized in that for storing computer program, wherein the computer journey
The labeling method such as target in the described in any item infrared images of Claims 1-4 is realized when sequence is executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810897531.3A CN109063701A (en) | 2018-08-08 | 2018-08-08 | Labeling method, device, equipment and the storage medium of target in a kind of infrared image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810897531.3A CN109063701A (en) | 2018-08-08 | 2018-08-08 | Labeling method, device, equipment and the storage medium of target in a kind of infrared image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109063701A true CN109063701A (en) | 2018-12-21 |
Family
ID=64678751
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810897531.3A Pending CN109063701A (en) | 2018-08-08 | 2018-08-08 | Labeling method, device, equipment and the storage medium of target in a kind of infrared image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109063701A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110570454A (en) * | 2019-07-19 | 2019-12-13 | 华瑞新智科技（北京）有限公司 | Method and device for detecting foreign matter invasion |
| CN111738180A (en) * | 2020-06-28 | 2020-10-02 | 浙江大华技术股份有限公司 | Key point marking method and device, storage medium and electronic device |
| WO2024002186A1 (en) * | 2022-06-28 | 2024-01-04 | 中兴通讯股份有限公司 | Image fusion method and apparatus, and storage medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101793562A (en) * | 2010-01-29 | 2010-08-04 | 中山大学 | Face detection and tracking algorithm of infrared thermal image sequence |
| CN103700101A (en) * | 2013-12-19 | 2014-04-02 | 华东师范大学 | Non-rigid brain image registration method |
| CN104134208A (en) * | 2014-07-17 | 2014-11-05 | 北京航空航天大学 | Coarse-to-fine infrared and visible light image registration method by adopting geometric construction characteristics |
| CN104268853A (en) * | 2014-03-06 | 2015-01-07 | 上海大学 | Infrared image and visible image registering method |
| CN105203159A (en) * | 2015-10-14 | 2015-12-30 | 武汉三江中电科技有限责任公司 | Single channel visible light and infrared image collecting, fusing and monitoring system |
| CN105371957A (en) * | 2015-10-23 | 2016-03-02 | 国家电网公司 | Transformer station equipment infrared temperature registration positioning and method |
| CN105654067A (en) * | 2016-02-02 | 2016-06-08 | 北京格灵深瞳信息技术有限公司 | Vehicle detection method and device |
| CN107491781A (en) * | 2017-07-21 | 2017-12-19 | 国家电网公司 | A kind of crusing robot visible ray and infrared sensor data fusion method |
| CN107578432A (en) * | 2017-08-16 | 2018-01-12 | 南京航空航天大学 | Merge visible ray and the target identification method of infrared two band images target signature |
| CN108008259A (en) * | 2017-11-14 | 2018-05-08 | 国网江西省电力有限公司电力科学研究院 | Based on infrared, the integrated detection method of Uv and visible light image co-registration and device |
-
2018
- 2018-08-08 CN CN201810897531.3A patent/CN109063701A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101793562A (en) * | 2010-01-29 | 2010-08-04 | 中山大学 | Face detection and tracking algorithm of infrared thermal image sequence |
| CN103700101A (en) * | 2013-12-19 | 2014-04-02 | 华东师范大学 | Non-rigid brain image registration method |
| CN104268853A (en) * | 2014-03-06 | 2015-01-07 | 上海大学 | Infrared image and visible image registering method |
| CN104134208A (en) * | 2014-07-17 | 2014-11-05 | 北京航空航天大学 | Coarse-to-fine infrared and visible light image registration method by adopting geometric construction characteristics |
| CN105203159A (en) * | 2015-10-14 | 2015-12-30 | 武汉三江中电科技有限责任公司 | Single channel visible light and infrared image collecting, fusing and monitoring system |
| CN105371957A (en) * | 2015-10-23 | 2016-03-02 | 国家电网公司 | Transformer station equipment infrared temperature registration positioning and method |
| CN105654067A (en) * | 2016-02-02 | 2016-06-08 | 北京格灵深瞳信息技术有限公司 | Vehicle detection method and device |
| CN107491781A (en) * | 2017-07-21 | 2017-12-19 | 国家电网公司 | A kind of crusing robot visible ray and infrared sensor data fusion method |
| CN107578432A (en) * | 2017-08-16 | 2018-01-12 | 南京航空航天大学 | Merge visible ray and the target identification method of infrared two band images target signature |
| CN108008259A (en) * | 2017-11-14 | 2018-05-08 | 国网江西省电力有限公司电力科学研究院 | Based on infrared, the integrated detection method of Uv and visible light image co-registration and device |
Non-Patent Citations (4)
| Title |
|---|
| 曾文锋: "红外与可见光图像融合中的快速配准方法", 《红外与激光工程》 * |
| 杜军平 等: "《多源运动图像的跨尺度融合研究》", 30 June 2018 * |
| 王艳翔 等: "可见光与近红外医学图像融合算法及软件", 《科学技术与工程》 * |
| 裴璐乾: "SAR、红外、可见光图像配准及融合算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110570454A (en) * | 2019-07-19 | 2019-12-13 | 华瑞新智科技（北京）有限公司 | Method and device for detecting foreign matter invasion |
| CN110570454B (en) * | 2019-07-19 | 2022-03-22 | 华瑞新智科技（北京）有限公司 | Method and device for detecting foreign matter invasion |
| CN111738180A (en) * | 2020-06-28 | 2020-10-02 | 浙江大华技术股份有限公司 | Key point marking method and device, storage medium and electronic device |
| CN111738180B (en) * | 2020-06-28 | 2023-03-24 | 浙江大华技术股份有限公司 | Key point marking method and device, storage medium and electronic device |
| WO2024002186A1 (en) * | 2022-06-28 | 2024-01-04 | 中兴通讯股份有限公司 | Image fusion method and apparatus, and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6821762B2 (en) | Systems and methods for detecting POI changes using convolutional neural networks | |
| CN109063701A (en) | Labeling method, device, equipment and the storage medium of target in a kind of infrared image | |
| TWI716012B (en) | Sample labeling method, device, storage medium and computing equipment, damage category identification method and device | |
| CN103996036B (en) | A kind of map data collecting method and device | |
| CN105354548A (en) | Surveillance video pedestrian re-recognition method based on ImageNet retrieval | |
| CN110309768B (en) | Method and equipment for detecting staff at vehicle inspection station | |
| US11403766B2 (en) | Method and device for labeling point of interest | |
| CN102982332A (en) | Retail terminal goods shelf image intelligent analyzing system based on cloud processing method | |
| CN106296814A (en) | Highway maintenance detection and virtual interactive interface method and system | |
| JP2016151967A (en) | Information processor, road structure management system, and road structure management method | |
| CN103065520A (en) | Detection system for backing car into storage and detection method thereof | |
| CN113255578B (en) | Traffic identification recognition method and device, electronic equipment and storage medium | |
| CN106251695A (en) | Destination's parking stall intelligent recommendation system and method based on parking space state monitoring | |
| CN109886928A (en) | A kind of target cell labeling method, device, storage medium and terminal device | |
| JP2022514891A (en) | Systems and methods for automatic image labeling for supervised machine learning | |
| CN101848377A (en) | Device and method for intelligent linkage of multi-video recording device based on cloud computing and mass video searching | |
| JP7389787B2 (en) | Domain adaptive object detection device and method based on multi-level transition region | |
| CN109710705A (en) | Map point of interest treating method and apparatus | |
| KR20200001455A (en) | Method, device and system for processing image tagging information | |
| CN103076007A (en) | Side parking detecting system and method | |
| CN114202003A (en) | System and method for road sign ground truth construction using knowledge graphs and machine learning | |
| Kronprasert et al. | Intersection safety assessment using video-based traffic conflict analysis: The case study of Thailand | |
| CN104766293A (en) | Method and device for detecting blood vessel in image | |
| CN108416769A (en) | Based on pretreated IVOCT images vulnerable plaque automatic testing method | |
| CN112016542A (en) | Urban waterlogging intelligent detection method and system |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181221 |