CN108647574A - Floating material image detection model generating method, recognition methods and equipment - Google Patents
Floating material image detection model generating method, recognition methods and equipment Download PDFInfo
- Publication number
- CN108647574A CN108647574A CN201810314088.2A CN201810314088A CN108647574A CN 108647574 A CN108647574 A CN 108647574A CN 201810314088 A CN201810314088 A CN 201810314088A CN 108647574 A CN108647574 A CN 108647574A
- Authority
- CN
- China
- Prior art keywords
- image
- floating material
- floating
- detection model
- grade
- 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
- G06V20/00—Scenes; Scene-specific elements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides a kind of floating material image detection model generating method, recognition methods and equipment.The floating material image detection model generating method, including：Obtain training data；The training data includes the floating object image floating material grade corresponding with each floating object image for setting quantity；Initial neural network is trained using the training data, obtains floating material image detection model.Using the floating material image detection Model Identification floating material of the present invention, the light by floating material region can be avoided to be influenced, to improve the accuracy of floating material image recognition.
Description
Technical field
The present invention relates to image identification technical field more particularly to a kind of floating material image detection model generating method, drifts
Image-recognizing method, computer readable storage medium and the computer equipment of float.
Background technology
With the development of information age, image data is more huge, and is being increased with a speed quickly.Pass through this
A little images can excavate out the thinking of people, and foundation is provided for further decision.Before carrying out data mining, need pair
These huge image datas carry out effective Classification Management.In the Classification Management to image data, image recognition is one
Element task has important meaning and value in terms of information extraction, information identification and information retrieval.Allow computer picture
People equally carries out image classification or image recognition to image data, and there is prodigious difficulty, these difficulties to be that image data has
Scrambling, the representation method of different images data, order of magnitude difference of image etc..Deep learning (Deep
Learning), especially convolutional neural networks (CNN), are the research emphasis in Image Processing and Pattern Recognition in recent years, by
It is more and more paid close attention to people.It is a kind of neural network established, simulate human brain progress analytic learning, can imitate human brain
Mechanism explain data, such as image, sound and text.
Floating material digital image recognition system is a towards water industry and energy industry in numerous image recognition products
River drifting substances image automatic identification product can float in video pictures with the presence or absence of rubbish before automatic identification river or reservoir
Object, calculates the size and occupied area of floating material, and the achievement of identification is carried out to the service application of profession, substitutes traditional artificial
Video monitor improves user job efficiency, reduces manual operation.
From the point of view of the hierarchical structure of machine learning model, the development of machine learning probably experienced to be changed twice：Shallow-layer
Habit and deep learning.When in face of foreign lands' problem of multiple variables, shallow-layer learning model is difficult to express it, needs multilayer
The network of implicit node could be effectively indicated, and deep learning model succinctly can effectively indicate complicated function.With
The development of big data, huge data are faced, shallow-layer learning structure is even more to have shown short slab in terms of model descriptive power,
It is difficult to the intrinsic representation of abundant mining data, only the stronger model of ability to express could be excavated out from big data more more has
The information of value, this also excites the learning motivation that people explore deep learning model to Complex Function Modeling.Image recognition
Goal in research is exactly to be divided into pre-defined different classifications according to certain possessed attribute in image.How to build
Image feature representation and disaggregated model are the key that solve the problems, such as image understanding, and Many researchers are conducted extensive research and carried
Some effective methods are gone out.Traditional method is largely view-based access control model code book model, which is utilized manually well
The iamge description of ingehious design and effective machine learning model, but its expression to image media layer damage and high-layer semantic information
Power is limited, can not break through semantic gap.In recent years, the breakthrough development of deep learning provides new think of to solve this problem
Road, and application of succeeding in many pattern recognition problems.
However, the existing rubbish floating object recognition methods based on traditional images color treatments technology, by extracting image
Color histogram is gone forward side by side row threshold division, and calculates the ratio of rubbish floating object to judge the severity of pollution.This method by
It is limited to the variation of the specific diversity and environment of floating material, when floating material color is various, being easy will be close with water body color
Rubbish misidentify into normal water body, be more when illumination difference, the variation of water body color is also very big, causes identification wrong
Accidentally.The existing pollution level recognition methods based on floating analyte detection is trained very using support vector machines as grader
Limited rubbish floating object type, and the quantity for calculating rubbish floating object is used as according to identification pollution level.This method equally by
It is limited to the variation of the specific diversity and environment of rubbish floating object, under true environment, rubbish is random to stack, is rambling
Floating, light is poor, so that being difficult to identify that corresponding floating species and its quantity, pollution level is caused to identify mistake.
Invention content
The present invention provides a kind of floating material image detection model generating method, recognition methods and equipment, to improve floating
The accuracy of object image identification.
An embodiment of the present invention provides a kind of floating material image detection model generating methods, including：Obtain training data；Institute
It includes the floating object image floating material grade corresponding with each floating object image for setting quantity to state training data；Using institute
It states training data and trains initial neural network, obtain floating material image detection model.
In one embodiment, training data is obtained, including：Obtain image data set；Each of described image data set
Image corresponds to a floating material grade；Data extending is carried out to described image data set by image transformation, obtains the training
Data；Described image transformation includes one or more of rotation, mirror image, flexible and distortion.
In one embodiment, the floating material grade is normal, slight pollution, intermediate pollution or serious pollution；It is described first
Beginning neural network is CaffeNet neural networks.
In one embodiment, initial neural network is trained using the training data, obtains floating material image detection model,
Including：Based on ImageNet models, training is finely adjusted to initial neural network, obtains floating material image detection model.
The embodiment of the present invention also provides a kind of image-recognizing method of floating material, including：Obtain the monitoring of floating object area
Image；Using monitoring image described in floating material image detection model inspection, at least one floating material grade is obtained；The floating material
Image detection model trains initial neural network to obtain using training data；The training data includes the drift for setting quantity
Float image floating material grade corresponding with each floating object image.
In one embodiment, using monitoring image described in floating material image detection model inspection, at least one floating is obtained
Object grade, including：Using the multiple monitoring images of floating material image detection model inspection, the monitoring image is obtained at least
One floating material grade and corresponding probability.
In one embodiment, the image-recognizing method of the floating material further includes：When the corresponding floating material of the monitoring image
Grade is setting grade, and the probability of the floating material grade is more than to send warning information to terminal device when setting probability value.
In one embodiment, the image-recognizing method of the floating material further includes：It is corresponding each to send the monitoring image
Floating material grade and corresponding probability to storage device or show equipment.
The embodiment of the present invention also provides a kind of computer readable storage medium, is stored thereon with computer program, the program
The step of the various embodiments described above the method is realized when being executed by processor.
The embodiment of the present invention also provides a kind of computer equipment, including memory, processor and storage are on a memory simultaneously
The computer program that can be run on a processor, the processor realize the various embodiments described above the method when executing described program
The step of.
Floating material image detection model generating method, the image-recognizing method of floating material, the computer of the embodiment of the present invention
Readable storage medium storing program for executing and computer equipment, by making the training data include setting the floating object image of quantity and each described
The corresponding floating material grade of object image is floated, and trains initial neural network to obtain floating object image using the training data and examines
Model is surveyed, the light by floating material region can be avoided to be influenced, the floating material image recognition based on color or quantity is overcome
The shortcomings that method, improves the accuracy of floating material image recognition.
Description of the drawings
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 technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
Obtain other attached drawings according to these attached drawings.In the accompanying drawings：
Fig. 1 is the flow diagram of the floating material image detection model generating method of one embodiment of the invention.
Fig. 2 is the method flow schematic diagram that training data is obtained in one embodiment of the invention.
Fig. 3 is the flow diagram of the image-recognizing method of the floating material of one embodiment of the invention.
Fig. 4 is the flow diagram of the image-recognizing method of the floating material of further embodiment of this invention.
Specific implementation mode
Understand in order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the accompanying drawings to this hair
Bright embodiment is described in further details.Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but simultaneously
It is not as a limitation of the invention.
Fig. 1 is the flow diagram of the floating material image detection model generating method of one embodiment of the invention.Such as Fig. 1 institutes
Show, the floating material image detection model generating method, it may include：
Step S110：Obtain training data；The training data includes setting the floating object image of quantity and each described
Float the corresponding floating material grade of object image；
Step S120：Initial neural network is trained using the training data, obtains floating material image detection model.
The floating object image is to include the image of floating material, such as the image of reservoir surface, may include floating material in image.
Include the floating object image of setting quantity in training data.Each floating object image corresponds to one or more floating material grades.Drift
Float grade can be by manually evaluating.After training data is inputted initial neural network, the drift of the training data can be exported
For float grade forecast as a result, for example, if belonging to a certain floating material grade, output result can be 1, if it is not, output result
It can be 0, output result is compared with the correct floating material grade known in advance, due to the sample size of training data
(setting quantity) is certain, it is possible to which the accuracy for calculating prediction is known that prediction error according to prediction accuracy, passes through
Iterative calculation can reduce prediction error, reach given threshold.The floating material image detection model can be directed to certain network
The parameter model of structure.
In the present embodiment, by making the training data include the floating object image for setting quantity and each floating material
The corresponding floating material grade of image, and train initial neural network to obtain floating material image detection mould using the training data
Type can avoid the light by floating material region from being influenced, and overcome the floating material image-recognizing method based on color or quantity
The shortcomings that, improve the accuracy of floating material image recognition.
Fig. 2 is the method flow schematic diagram that training data is obtained in one embodiment of the invention.As shown in Fig. 2, in above-mentioned step
In rapid S110, training data is obtained, it may include：
Step S111：Obtain image data set；Each image in described image data set corresponds to a floating material grade；
Step S112：Data extending is carried out to described image data set by image transformation, obtains the training data；Institute
It includes one or more of rotation, mirror image, flexible and distortion to state image transformation.
In the present embodiment, by being rotated to described image data set, mirror image, the transformation such as flexible, distortion, carry out data
Expand, training data total amount can be increased in the case where not increasing image collection amount, improves the efficiency of training data preparation.
In some embodiments, the floating material grade is normal, slight pollution, intermediate pollution or serious pollution.This is normal,
E.g. normal water body, normal air etc..In some embodiments, the initial neural network is CaffeNet neural networks.
In some embodiments, in above-mentioned steps S120, initial neural network is trained using the training data, is floated
Float image detection model, it may include：Based on ImageNet models, training is finely adjusted to initial neural network, is floated
Object image detection model.ImageNet models are a parameter models, and training, energy are finely adjusted based on ImageNet models
The calculation amount for reducing training, reduces the quantitative requirement of training sample.Such as iterations be 50,000 times, learning rate be 0.005~
0.05, such as take 0.01.
Floating material image detection model based on above-described embodiment, the present invention also provides a kind of image recognition sides of floating material
Method.
Fig. 3 is the flow diagram of the image-recognizing method of the floating material of one embodiment of the invention.As shown in figure 3, the drift
The image-recognizing method of float, it may include：
Step S210：Obtain the monitoring image of floating object area；
Step S220：Using monitoring image described in floating material image detection model inspection, at least one floating material etc. is obtained
Grade；The floating material image detection model trains initial neural network to obtain using training data；The training data packet
Include the floating object image floating material grade corresponding with each floating object image of setting quantity.
In the present embodiment, floating material image detection model is by making the training data include the floating material for setting quantity
Image floating material grade corresponding with each floating object image, and train initial neural network to obtain using the training data
It arrives, the image recognition of floating material is carried out using the floating material image detection model, the light by floating material region can be avoided
Line influences, and the shortcomings that overcoming the floating material image-recognizing method based on color or quantity improves the accurate of floating material image recognition
Degree.
In some embodiments, in the case where the monitoring image is multiple, above-mentioned steps S210 utilizes floating object image
Detection model detects the monitoring image, obtains at least one floating material grade, it may include：Utilize floating material image detection model
Multiple monitoring images are detected, at least one floating material grade of the monitoring image and corresponding probability are obtained.
Pass through the corresponding floating material grade of a certain monitoring image ratio shared in the floating material grade of multiple monitoring images
Example can obtain probability.One monitoring image can correspond to one or more floating material grades, and each floating material grade can be right
Answer a probability.When monitoring image corresponds to multiple floating material grades, if the probability of one of floating material grade obviously compared with
Greatly, it is believed that recognition result is that this larger floating material grade of probability can if each floating material grade probability is similar
To think normal or think that identification is invalid.
Fig. 4 is the flow diagram of the image-recognizing method of the floating material of further embodiment of this invention.As shown in figure 4, should
The image-recognizing method of floating material, may also include：
Step S230：When the corresponding floating material grade of the monitoring image is to set grade, and the floating material grade is general
When rate is more than setting probability value, warning information is sent to terminal device.
Floating material grade can be normal water body, slight pollution, intermediate pollution or serious pollution, for example, when floating material etc.
When grade is serious pollution (setting grade), warning information can be sent to terminal device, to notify related personnel to handle.It should
Terminal device is, for example, the equipment such as computer, mobile phone.It (is answered further, it is possible to which warning information is sent to the APP on terminal device
With) in, such as short message, wechat etc..
In some embodiments, the image-recognizing method of floating material may also include：It is corresponding each to send the monitoring image
Floating material grade and corresponding probability to storage device or show equipment.Such as floating material grade and corresponding probability can be sent out
It send to computer, and checks or inquire on the web page of computer.
In some embodiments, above-mentioned steps S210 obtains the monitoring image of floating object area, it may include：Obtain floating material
The video image in region；The video image is converted into monitoring image.Video image can be obtained from the monitoring point of object to be measured
.
In one specific embodiment, by taking the rubbish floating object identification in water body as an example, the rubbish floating based on deep learning
Object image automatic identification algorithm, it may include：
Step S1：Rubbish floating object image recognition network structure is defined, training network based on CaffeNet is used.
Step S2：Data extending is carried out to data set, increases data set Random-Rotation, mirror image, flexible, Skewed transformation figure
Picture, and data set is divided into four grades：Normal water body, slight pollution, intermediate pollution, serious pollution.
Step S3：According to network structure and parameter that S1 is configured, setting iterations are 50000 times, learning rate 0.01,
Batch sizes are 50, and are finely adjusted training on ImageNet models, obtain rubbish floating object identification model.
Step S4：The rubbish floating object identification model obtained according to S3 is sent into network with input picture I and carries out forward calculation,
The level categories and its probability of rubbish floating object prediction are exported when prediction, so far algorithm description terminates.
The method of the present embodiment utilizes video pictures before computer picture automatic identification technology automatic identification river or reservoir
In whether there is rubbish floating object, calculate the size and occupied area of floating material, with replace traditional artificial naked eyes judge reading, and
Early-warning conditions are set, early warning is automatically analyzed, the throwing of the artificial energy of river reservoir rubbish floating object video monitoring can be gradually reduced
Enter, really promote working efficiency, rubbish floating object video monitoring ability and dynamics will be substantially improved, and note abnormalities accurate much sooner
Really, the effect at rubbish floating object video monitoring station will be played utmostly.
In some embodiments, floating material digital image recognition system mainly may include the access of video image, image recognition
Computation model, result of calculation analysis and early warning, result of calculation displaying.The access of video image refers to floating material digital image recognition system
System is connect with video monitoring point device, obtains video image information in real time, and video image conversion picture is preserved, used by system automatically
It is identified in image content.Image recognition computation model is to establish floating material for video frequency graphic monitoring before river or reservoir automatically to know
Other computation model allows computer that can automatically identify picture and whether there is rubbish floating object, calculates size and the institute of floating material
Area, and constantly self-teaching optimization are accounted for, recognition accuracy is improved.Result of calculation analysis and early warning is to be directed to the achievement identified,
By setting threshold value of warning, automatic decision whether superthreshold, can automatically generate warning information after superthreshold, and by wechat,
Short message sending related personnel.Result of calculation displaying is the achievement information calculated for system automatic identification, is carried out by web intuitive
Displaying and inquiry.
The embodiment of the present invention also provides a kind of computer readable storage medium, is stored thereon with computer program, the program
The step of the various embodiments described above the method is realized when being executed by processor.
The embodiment of the present invention also provides a kind of computer equipment, including memory, processor and storage are on a memory simultaneously
The computer program that can be run on a processor, the processor realize the various embodiments described above the method when executing described program
The step of.
In conclusion the image recognition side of the floating material image detection model generating method of the embodiment of the present invention, floating material
Method, computer readable storage medium and computer equipment, by making the training data include the floating object image for setting quantity
Floating material grade corresponding with each floating object image, and train initial neural network to be floated using the training data
Float image detection model can avoid the light by floating material region from being influenced, and overcome the floating based on color or quantity
The shortcomings that object image recognition methods, improves the accuracy of floating material image recognition.
In the description of this specification, reference term " one embodiment ", " specific embodiment ", " some implementations
Example ", " such as ", the description of " example ", " specific example " or " some examples " etc. mean it is described in conjunction with this embodiment or example
Particular features, structures, materials, or characteristics are included at least one embodiment or example of the invention.In the present specification,
Schematic expression of the above terms may not refer to the same embodiment or example.Moreover, the specific features of description, knot
Structure, material or feature can be combined in any suitable manner in any one or more of the embodiments or examples.Each embodiment
Involved in the step of implementation of the sequence for schematically illustrating the present invention, sequence of steps therein is not construed as limiting, can be as needed
It appropriately adjusts.
It should be understood by those skilled in the art that, the embodiment of the present invention can be provided as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the present invention
Apply the form of example.Moreover, the present invention can be used in one or more wherein include computer usable program code computer
The computer program production implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of product.
The present invention be with reference to according to the method for the embodiment of the present invention, the flow of equipment (system) and computer program product
Figure and/or block diagram describe.It should be understood that can be realized by computer program instructions every first-class in flowchart and/or the block diagram
The combination of flow and/or box in journey and/or box and flowchart and/or the block diagram.These computer programs can be provided
Instruct the processor of all-purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine so that the instruction executed by computer or the processor of other programmable data processing devices is generated for real
The device for the function of being specified in present one flow of flow chart or one box of multiple flows and/or block diagram or multiple boxes.
These computer program instructions, which may also be stored in, can guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works so that instruction generation stored in the computer readable memory includes referring to
Enable the manufacture of device, the command device realize in one flow of flow chart or multiple flows and/or one box of block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device so that count
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, in computer or
The instruction executed on other programmable devices is provided for realizing in one flow of flow chart or multiple flows and/or block diagram one
The step of function of being specified in a box or multiple boxes.
Particular embodiments described above has carried out further in detail the purpose of the present invention, technical solution and advantageous effect
Describe in detail it is bright, it should be understood that the above is only a specific embodiment of the present invention, the guarantor being not intended to limit the present invention
Range is protected, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should be included in this
Within the protection domain of invention.
Claims (10)
1. a kind of floating material image detection model generating method, which is characterized in that including：
Obtain training data；The training data includes that the floating object image of setting quantity is corresponding with each floating object image
Floating material grade；
Initial neural network is trained using the training data, obtains floating material image detection model.
2. floating material image detection model generating method as described in claim 1, which is characterized in that obtain training data, packet
It includes：
Obtain image data set；Each image in described image data set corresponds to a floating material grade；
Data extending is carried out to described image data set by image transformation, obtains the training data；Described image transformation packet
Include one or more of rotation, mirror image, flexible and distortion.
3. floating material image detection model generating method as described in claim 1, which is characterized in that the floating material grade is
Normally, slight pollution, intermediate pollution or serious pollution；The initial neural network is CaffeNet neural networks.
4. floating material image detection model generating method as described in claim 1, which is characterized in that utilize the training data
The initial neural network of training, obtains floating material image detection model, including：
Based on ImageNet models, training is finely adjusted to initial neural network, obtains floating material image detection model.
5. a kind of image-recognizing method of floating material, which is characterized in that including：
Obtain the monitoring image of floating object area；
Using monitoring image described in floating material image detection model inspection, at least one floating material grade is obtained；The floating material
Image detection model trains initial neural network to obtain using training data；The training data includes the drift for setting quantity
Float image floating material grade corresponding with each floating object image.
6. the image-recognizing method of floating material as claimed in claim 5, which is characterized in that utilize floating material image detection model
The monitoring image is detected, at least one floating material grade is obtained, including：
Using the multiple monitoring images of floating material image detection model inspection, at least one floating of the monitoring image is obtained
Object grade and corresponding probability.
7. the image-recognizing method of floating material as claimed in claim 6, which is characterized in that further include：
When the corresponding floating material grade of the monitoring image is to set grade, and the probability of the floating material grade is more than setting probability
When value, warning information is sent to terminal device.
8. the image-recognizing method of floating material as claimed in claim 6, which is characterized in that further include：
The corresponding each floating material grade of the monitoring image and corresponding probability are sent to storage device or display equipment.
9. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is held by processor
The step of claim 1 to 8 the method is realized when row.
10. a kind of computer equipment, including memory, processor and storage are on a memory and the meter that can run on a processor
Calculation machine program, which is characterized in that the step of processor realizes claim 1 to 8 the method when executing described program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810314088.2A CN108647574A (en) | 2018-04-10 | 2018-04-10 | Floating material image detection model generating method, recognition methods and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810314088.2A CN108647574A (en) | 2018-04-10 | 2018-04-10 | Floating material image detection model generating method, recognition methods and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108647574A true CN108647574A (en) | 2018-10-12 |
Family
ID=63745606
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810314088.2A Pending CN108647574A (en) | 2018-04-10 | 2018-04-10 | Floating material image detection model generating method, recognition methods and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108647574A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109614924A (en) * | 2018-12-07 | 2019-04-12 | 江西洪都航空工业集团有限责任公司 | A kind of garbage on water detection method based on deep learning algorithm |
| CN110674740A (en) * | 2019-09-24 | 2020-01-10 | 北京海益同展信息科技有限公司 | Underwater garbage cleaning method, device and system and computer readable storage medium |
| CN110866900A (en) * | 2019-11-05 | 2020-03-06 | 江河瑞通（北京）技术有限公司 | Water body color identification method and device |
| CN111104860A (en) * | 2019-11-19 | 2020-05-05 | 浙江工业大学 | Unmanned aerial vehicle water quality chromaticity monitoring method based on machine vision |
| CN111223262A (en) * | 2020-01-19 | 2020-06-02 | 西石(厦门)科技股份有限公司 | Flue fire hazard early warning method based on intelligent image identification |
| CN112418128A (en) * | 2020-11-30 | 2021-02-26 | 重庆市生态环境大数据应用中心 | Surface water monitoring and management system and method |
| CN112668591A (en) * | 2020-08-25 | 2021-04-16 | 广东生太修复科技有限公司 | Intelligent river water burst treatment method and related device |
| CN113344885A (en) * | 2021-06-15 | 2021-09-03 | 温州大学 | River floating object detection method based on cascade convolution neural network |
| CN113850482A (en) * | 2021-09-09 | 2021-12-28 | 中国铁道科学研究院集团有限公司 | Risk early warning method and device for floaters along railway |
| CN115249339A (en) * | 2022-06-10 | 2022-10-28 | 广州中科云图智能科技有限公司 | River floating object identification system, method, equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007135659A3 (en) * | 2006-05-23 | 2008-07-31 | Elbit Sys Electro Optics Elop | Clustering - based image registration |
| CN105631482A (en) * | 2016-03-03 | 2016-06-01 | 中国民航大学 | Convolutional neural network model-based dangerous object image classification method |
| CN105630882A (en) * | 2015-12-18 | 2016-06-01 | 哈尔滨工业大学深圳研究生院 | Remote sensing data deep learning based offshore pollutant identifying and tracking method |
| CN106874948A (en) * | 2017-02-08 | 2017-06-20 | 武汉海卓科科技有限公司 | A kind of black smelly water automatic identification and appraisal procedure |
-
2018
- 2018-04-10 CN CN201810314088.2A patent/CN108647574A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007135659A3 (en) * | 2006-05-23 | 2008-07-31 | Elbit Sys Electro Optics Elop | Clustering - based image registration |
| CN105630882A (en) * | 2015-12-18 | 2016-06-01 | 哈尔滨工业大学深圳研究生院 | Remote sensing data deep learning based offshore pollutant identifying and tracking method |
| CN105631482A (en) * | 2016-03-03 | 2016-06-01 | 中国民航大学 | Convolutional neural network model-based dangerous object image classification method |
| CN106874948A (en) * | 2017-02-08 | 2017-06-20 | 武汉海卓科科技有限公司 | A kind of black smelly water automatic identification and appraisal procedure |
Non-Patent Citations (1)
| Title |
|---|
| 吴有富 等: "一种基于SOM的水面污染物智能监控***", 《环保科技》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109614924A (en) * | 2018-12-07 | 2019-04-12 | 江西洪都航空工业集团有限责任公司 | A kind of garbage on water detection method based on deep learning algorithm |
| CN110674740A (en) * | 2019-09-24 | 2020-01-10 | 北京海益同展信息科技有限公司 | Underwater garbage cleaning method, device and system and computer readable storage medium |
| CN110866900A (en) * | 2019-11-05 | 2020-03-06 | 江河瑞通（北京）技术有限公司 | Water body color identification method and device |
| CN111104860A (en) * | 2019-11-19 | 2020-05-05 | 浙江工业大学 | Unmanned aerial vehicle water quality chromaticity monitoring method based on machine vision |
| CN111104860B (en) * | 2019-11-19 | 2022-02-15 | 浙江工业大学 | Unmanned aerial vehicle water quality chromaticity monitoring method based on machine vision |
| CN111223262A (en) * | 2020-01-19 | 2020-06-02 | 西石(厦门)科技股份有限公司 | Flue fire hazard early warning method based on intelligent image identification |
| CN112668591A (en) * | 2020-08-25 | 2021-04-16 | 广东生太修复科技有限公司 | Intelligent river water burst treatment method and related device |
| CN112418128B (en) * | 2020-11-30 | 2022-01-21 | 重庆市生态环境大数据应用中心 | Surface water monitoring and management system and method |
| CN112418128A (en) * | 2020-11-30 | 2021-02-26 | 重庆市生态环境大数据应用中心 | Surface water monitoring and management system and method |
| CN113344885A (en) * | 2021-06-15 | 2021-09-03 | 温州大学 | River floating object detection method based on cascade convolution neural network |
| CN113850482A (en) * | 2021-09-09 | 2021-12-28 | 中国铁道科学研究院集团有限公司 | Risk early warning method and device for floaters along railway |
| CN115249339A (en) * | 2022-06-10 | 2022-10-28 | 广州中科云图智能科技有限公司 | River floating object identification system, method, equipment and storage medium |
| CN115249339B (en) * | 2022-06-10 | 2024-05-28 | 广州中科云图智能科技有限公司 | River float recognition system, method, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108647574A (en) | Floating material image detection model generating method, recognition methods and equipment | |
| EP3913542A2 (en) | Method and apparatus of training model, device, medium, and program product | |
| CN110796031A (en) | Table identification method and device based on artificial intelligence and electronic equipment | |
| CN110276264B (en) | Crowd density estimation method based on foreground segmentation graph | |
| CN108596221A (en) | The image-recognizing method and equipment of rod reading | |
| CN104077447B (en) | Urban three-dimensional space vector modeling method based on paper plane data | |
| CN110516539A (en) | Remote sensing image building extracting method, system, storage medium and equipment based on confrontation network | |
| CN113486726A (en) | Rail transit obstacle detection method based on improved convolutional neural network | |
| CN113362299B (en) | X-ray security inspection image detection method based on improved YOLOv4 | |
| CN116863274A (en) | Semi-supervised learning-based steel plate surface defect detection method and system | |
| CN113642474A (en) | Hazardous area personnel monitoring method based on YOLOV5 | |
| CN106445915A (en) | New word discovery method and device | |
| JPWO2020240808A1 (en) | Learning device, classification device, learning method, classification method, learning program, and classification program | |
| CN111461121A (en) | Electric meter number identification method based on YO L OV3 network | |
| CN108961358A (en) | A kind of method, apparatus and electronic equipment obtaining samples pictures | |
| CN111611933A (en) | Information extraction method and system for document image | |
| CN116861258B (en) | Model processing method, device, equipment and storage medium | |
| CN114463643A (en) | Multi-model decision-level fusion landslide identification method and device | |
| CN114283315A (en) | RGB-D significance target detection method based on interactive guidance attention and trapezoidal pyramid fusion | |
| CN109829469A (en) | A kind of vehicle checking method based on deep learning | |
| CN112257496A (en) | Deep learning-based power transmission channel surrounding environment classification method and system | |
| CN113537173B (en) | Face image authenticity identification method based on face patch mapping | |
| CN114359269A (en) | Virtual food box defect generation method and system based on neural network | |
| CN113870196A (en) | Image processing method, device, equipment and medium based on anchor point cutting graph | |
| CN117253188A (en) | Transformer substation grounding wire state target detection method based on improved YOLOv5 |
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 |