CN110502663A - A kind of rds data processing method and processing device based on artificial intelligence - Google Patents
A kind of rds data processing method and processing device based on artificial intelligence Download PDFInfo
- Publication number
- CN110502663A CN110502663A CN201910790317.2A CN201910790317A CN110502663A CN 110502663 A CN110502663 A CN 110502663A CN 201910790317 A CN201910790317 A CN 201910790317A CN 110502663 A CN110502663 A CN 110502663A
- Authority
- CN
- China
- Prior art keywords
- video
- videograph
- camera
- record
- interest
- 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
- 238000012545 processing Methods 0.000 title claims abstract description 44
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 16
- 238000003672 processing method Methods 0.000 title claims abstract description 11
- 230000011218 segmentation Effects 0.000 claims abstract description 44
- 238000000034 method Methods 0.000 claims abstract description 36
- 238000013507 mapping Methods 0.000 claims abstract description 17
- 238000001914 filtration Methods 0.000 claims description 44
- 238000000605 extraction Methods 0.000 claims description 9
- 239000012634 fragment Substances 0.000 claims description 4
- 238000010191 image analysis Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 239000012141 concentrate Substances 0.000 claims description 3
- 238000007405 data analysis Methods 0.000 abstract description 5
- 238000012549 training Methods 0.000 description 12
- 230000000694 effects Effects 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000013135 deep learning Methods 0.000 description 3
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000002203 pretreatment Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 229910052742 iron Inorganic materials 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/71—Indexing; Data structures therefor; Storage structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/75—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Business, Economics & Management (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Databases & Information Systems (AREA)
- Signal Processing (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides a kind of rds data processing method and processing device based on artificial intelligence, the method includes multiple cameras are laid along rail, each camera is directed toward a camera record, the camera record includes camera identification, rail position where camera, the rail mark of the rail where camera;Construct video management mapping table, the video management mapping table record has the corresponding relationship between the video storage file folder that camera records and camera record is directed toward, and each video storage file folder record has at least one videograph of its corresponding camera shooting;Rds data process instruction is obtained, the Video segmentation instruction in the data processing instructions direction includes that camera record and the camera record relevant videograph and number;Videograph collection is extracted according to the data processing instructions and is handled.The present invention realizes the primary dcreening operation of railway information, and can provide convenience for subsequent further data analysis.
Description
Technical field
The present invention relates to railway territory more particularly to a kind of rds data processing method and processing devices based on artificial intelligence.
Background technique
Railway is one of key pillars of the national economy, and the requirement of railway construction and management is also increasingly being promoted, at video
The development of reason technology is that the monitoring link of railway administration brings many traversals, also thereby produces mass image data, these seas
Railway information and garbage are contained in amount image data, and how the image-region comprising railway information is extracted,
Become the problem of getting worse to reduce the interference of garbage.
Summary of the invention
In order to solve to extract in the prior art to the image-region comprising railway information, to reduce garbage
The technical issues of interference, the present invention provide a kind of rds data processing method and processing device based on artificial intelligence.
The present invention is realized with following technical solution:
A kind of rds data processing method based on artificial intelligence, which comprises
Multiple cameras are laid along rail, each camera is directed toward a camera record, the camera record
Including camera identification, rail position where camera, the rail mark of the rail where camera;
Video management mapping table is constructed, the video management mapping table record has camera record and camera record to be directed toward
Video storage file folder between corresponding relationship, each video storage file folder record has its corresponding camera shooting extremely
A few videograph;
Rds data process instruction is obtained, the data processing instructions are directed toward the instruction of at least one Video segmentation, each view
Frequency split order includes that a camera record and the camera record relevant at least one videograph number;
Videograph collection is extracted according to the data processing instructions;
The record concentrated to the videograph is handled.
It is preferably, described to extract videograph collection according to the data processing instructions, comprising:
Obtain each Video segmentation instruction;
It instructs acquisition camera to record corresponding video storage file folder according to each Video segmentation, and is deposited from the video
Selection target video in file is stored up, the number of the target video is by the Video segmentation instruction hit;
Videograph collection is constituted by the target video instructed according to each Video segmentation.
Preferably, video pre-filtering rule is for pre-processing videograph, in order to be obtained based on videograph
Area-of-interest can satisfy the basic demand for carrying out image analysis in Target Photo, and the basic demand includes area-of-interest
The plentiful degree of railway information meet the requirements and/or area-of-interest fragment requirement.
Preferably, the record concentrated to the videograph is handled, comprising:
It obtains each Video segmentation and instructs corresponding video pre-filtering rule;
Sorted out according to the videograph that the video pre-filtering rule concentrates the videograph, each classification knot
Fruit is directed toward unique video pre-filtering rule;
According to the videograph in its corresponding categorization results of video pre-filtering rule process.
Preferably, the videograph according in its corresponding categorization results of video pre-filtering rule process, comprising:
Video intercepting is carried out to the videograph in the categorization results according to the interception rule in video pre-filtering rule
To obtain pictures to be processed;
It is described to obtain that each picture to be processed in the pictures to be processed is inputted into preset image processing model
The area-of-interest recommendation results of picture to be processed, the area-of-interest recommendation results include interested in picture to be processed
Region and the corresponding index interested of area-of-interest;
It is regular interested with each picture to be processed according to the area-of-interest processing in the video pre-filtering rule
Region recommendation results obtain the corresponding Target Photo of each picture to be processed;
The corresponding Target Photo collection of the pictures to be processed is obtained according to the corresponding Target Photo of each picture to be processed.
A kind of rds data processing unit based on artificial intelligence, described device include:
Module is laid, for laying multiple cameras along rail, each camera is directed toward a camera record, institute
Stating camera record includes camera identification, the rail position where camera, the rail mark of the rail where camera;
Video management mapping table constructs module, for constructing video management mapping table, the video management mapping table record
There are the corresponding relationship between the video storage file folder that camera records and camera record is directed toward, each video storage file folder
Record has at least one videograph of its corresponding camera shooting;
Rds data process instruction obtains module, and for obtaining rds data process instruction, the data processing instructions refer to
It is instructed at least one Video segmentation, each Video segmentation instruction includes that a camera record is related to camera record
At least one videograph number;
Videograph collection extraction module, for extracting videograph collection according to the data processing instructions;
Processing module, the record for concentrating to the videograph are handled.
Preferably, the videograph collection extraction module, comprising:
Video segmentation command unit, for obtaining each Video segmentation instruction;
Videograph selecting unit is stored for instructing acquisition camera to record corresponding video according to each Video segmentation
File, and the selection target video from video storage file folder, the number of the target video is by the Video segmentation
Instruction hit;
Videograph collection generation unit, for being made of video note the target video instructed according to each Video segmentation
Record collection.
Preferably, the processing module, comprising:
Video pre-filtering Rule unit instructs corresponding video pre-filtering rule for obtaining each Video segmentation;
Sort out unit, the videograph for concentrating according to the video pre-filtering rule to the videograph is returned
Class, each categorization results are directed toward unique video pre-filtering rule;
Videograph acquiring unit, for being remembered according to the video in its corresponding categorization results of video pre-filtering rule process
Record.
The beneficial effects of the present invention are:
A kind of rds data processing method and processing device based on artificial intelligence, can be right according to various video pre-filtering rules
The relevant video data of railway carries out a variety of pretreatments to obtain meeting the Target Photo collection of video analysis requirement, so that mesh
The enrichment degree and/or number of tiles of marking this railway information of pictures meet the requirement of data analysis, realize railway information
Primary dcreening operation, and convenience can be provided for subsequent further data analysis.
Detailed description of the invention
Fig. 1 is a kind of rds data processing method flow chart based on artificial intelligence provided in this embodiment;
Fig. 2 is that the record provided in this embodiment concentrated to the videograph carries out process flow diagram;
Fig. 3 is the videograph provided in this embodiment according in its corresponding categorization results of video pre-filtering rule process
Flow chart;
Fig. 4 is a kind of rds data processing unit block diagram based on artificial intelligence provided in this embodiment.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, the present invention is made into one below in conjunction with attached drawing
Step ground detailed description.
Embodiment 1:
The embodiment of the present invention discloses a kind of rds data processing method based on artificial intelligence, as shown in Figure 1, the method
Include:
S101. multiple cameras are laid along rail, each camera is directed toward a camera record, the camera
Record includes camera identification, the rail position where camera, the rail mark of the rail where camera.
S103. video management mapping table is constructed, the video management mapping table record has camera record and camera note
The corresponding relationship between video storage file folder that record is directed toward, each video storage file folder record have its corresponding camera to clap
At least one videograph taken the photograph.
S105. rds data process instruction is obtained, the data processing instructions are directed toward the instruction of at least one Video segmentation, often
A Video segmentation instruction includes that a camera record and the camera record relevant at least one videograph number.
S107. videograph collection is extracted according to the data processing instructions.
Specifically, described to include: according to data processing instructions extraction videograph collection
Obtain each Video segmentation instruction.
It instructs acquisition camera to record corresponding video storage file folder according to each Video segmentation, and is deposited from the video
Selection target video in file is stored up, the number of the target video is by the Video segmentation instruction hit.
Videograph collection is constituted by the target video instructed according to each Video segmentation.
S109. the record concentrated to the videograph is handled.
Specifically, the record concentrated to the videograph is handled, as shown in Figure 2, comprising:
S1091. it obtains each Video segmentation and instructs corresponding video pre-filtering rule.
Video pre-filtering rule is for pre-processing videograph, in order to obtain Target Photo based on videograph
Middle area-of-interest can satisfy carry out image analysis basic demand, such as have area-of-interest railway information it is plentiful
Degree is met the requirements, that is, includes more railway information;The quantity of area-of-interest cannot be excessive in Target Photo for another example,
There should not be more area-of-interest fragment, in order to avoid cause taking long time for image analysis link.
S1093. sorted out according to the videograph that the video pre-filtering rule concentrates the videograph, each
Categorization results are directed toward unique video pre-filtering rule.
S1095. according to the videograph in its corresponding categorization results of video pre-filtering rule process.
Specifically, the videograph according in its corresponding categorization results of video pre-filtering rule process, such as Fig. 3 institute
Show, comprising:
S10951. the videograph in the categorization results is regarded according to the interception rule in video pre-filtering rule
Frequency interception is to obtain pictures to be processed.
Specifically, it can be intercepted according to the preset time interval.
S10953. each picture to be processed in the pictures to be processed is inputted into preset image processing model to obtain
To the area-of-interest recommendation results of the picture to be processed, the area-of-interest recommendation results include in picture to be processed
Area-of-interest and the corresponding index interested of area-of-interest.
Specifically, the preset image processing model can be used existing machine mould and be obtained by training.This
Inventive embodiments disclose the training method of one of machine mould, which comprises
S1. training sample set is constructed, the training sample set includes and positive example sample set and negative example sample set the positive example
Sample set is by the sample set including area-of-interest annotation results, and the negative example sample set is to be laid in taking the photograph around rail
The sample set for not being marked area-of-interest, the area-of-interest annotation results packet are randomly selected in the video flowing shot as head
Include the index interested of area-of-interest and area-of-interest.
S3. each training sample that the training sample is concentrated is carried out described in feature extraction acquisition based on feature extractor
The corresponding sample characteristics collection of training sample set.
Specifically, the feature extractor includes sequentially connected picture input layer, the first convolution process layer, the first pond
Layer, the second convolution process layer, the second pond layer, full articulamentum and normalization layer.Wherein, the feature extractor uses existing
Machine learning algorithm training obtains, and the embodiment of the present invention does not repeat them here.The embodiment of the present invention is, it is emphasized that in order to which Enhanced feature mentions
Effect is taken, the embodiment of the present invention improves the loss function of the feature extractor, increases to reach for one
The purpose of the difference of each characteristic pattern of the output of feature extractor described in the picture of input.
The loss function of the embodiment of the present invention can be obtained by adding selection item on primary loss functional foundations, institute
It states the loss that each characteristic pattern that selection item is used to export feature extractor generates to be adjusted, so that it is big to increase generation loss
Characteristic pattern influence power.
In a feasible embodiment, the loss function of the embodiment of the present invention can be represented asWherein α, β, M, δiThe weight of control selection degree is respectively indicated,
It can be set according to actual needs;Parameter is selected, can be a lesser numerical value, such as 0.025;At second convolution
Manage the number of the characteristic pattern of layer output;The characteristic pattern in i-th of channel that the second convolution process layer generates is being completely used for training institute
State the average activity in the sample of feature extractor.Wherein LlostFor the loss function of script, the concrete form of loss function
The prior art can be referred to, the embodiment of the present invention does not repeat them here, for example, it can beWherein
sum,Out(s1),T(s1) it is total sample number, the output label corresponding with some sample that some sample generates, wherein for true
Label is given value for fixed training set and machine mould.
Further, the characteristic pattern that the embodiment of the present invention provides i-th of channel is being completely used for training the feature extraction
The calculation method of average activity in the sample of device, which comprises
S10. each sample is obtained in whole feature set of graphs of each channel output of the second convolution process layer.
S30. the feature atlas in i-th of channel is obtained according to whole feature set of graphs, the feature atlas is described
Source is the set of the characteristic pattern composition in i-th of channel of the second convolution process layer in All Eigenvalues set.
S50. the pixel of each characteristic pattern is concentrated to be integrated respectively along X direction and y direction to obtain the characteristic pattern
To activity.
S70. it is being completely used for training the spy divided by the characteristic pattern that total sample number obtains i-th of channel by the activity
Levy the average activity in the sample of extractor.
S5. the sample characteristics collection is inputted into preset deep learning network model, according to the deep learning network mould
The area-of-interest recommendation results of each sample of type output, according to the area-of-interest recommendation results of each sample
The penalty values between area-of-interest for being labeled out in sample set adjust the parameter of the deep learning network model to obtain
To image processing model.
S10955. rule and each picture to be processed are handled according to the area-of-interest in the video pre-filtering rule
Area-of-interest recommendation results obtain the corresponding Target Photo of each picture to be processed.
In a feasible embodiment, the area-of-interest according in the video pre-filtering rule handles rule
The corresponding Target Photo of each picture to be processed is obtained with the area-of-interest recommendation results of each picture to be processed, comprising:
S100. each area-of-interest and the corresponding index interested of the area-of-interest of picture to be processed are obtained.
S200. the quantity of the area-of-interest is counted.
If S300. the quantity is greater than preset first threshold, according to the index interested of each area-of-interest and
The different degree of the areal calculation area-of-interest of area-of-interest.
Specifically, the different degree of the area-of-interest can by the weighted value of index interested and area-of-interest come
Characterization.
S400. processing is successively carried out to area-of-interest according to different degree descending order interested until area-of-interest is total
Until number is equal to preset first threshold.
In a feasible embodiment, carrying out processing to area-of-interest includes:
If the index interested of the area-of-interest is lower than preset third threshold value, the area-of-interest is deleted.
If the index interested of the area-of-interest is not less than preset third threshold value, the area-of-interest is obtained
Adjacent other area-of-interests, and select the highest region of index interested as target in other area-of-interests
Region.
The area-of-interest is merged into the target area.
In said embodiment, the iron of video pre-filtering rule area-of-interest suitable for for Target Photo
Road information enrichment degree has among the scene of relatively high requirement.
In another feasible embodiment, carrying out processing to area-of-interest includes:
If the index interested of the area-of-interest is lower than preset third threshold value, the area-of-interest is deleted.
If the area of the area-of-interest is not less than preset third threshold value, it is adjacent to obtain the area-of-interest
Other area-of-interests, and select the maximum region of area as target area in other area-of-interests.
The area-of-interest is merged into the target area.
In said embodiment, video pre-filtering rule fragment suitable for for Target Photo requires more severe
Among the scene at quarter.
S10957. the corresponding target of the pictures to be processed is obtained according to the corresponding Target Photo of each picture to be processed
Pictures.
The rds data processing method based on artificial intelligence that the embodiment of the invention discloses a kind of, can be according to various videos
Preprocessing rule carries out a variety of pretreatments to the relevant video data of railway to obtain meeting the Target Photo of video analysis requirement
Collection, so that the enrichment degree of this railway information of Target Photo collection and/or number of tiles meet the requirement of data analysis, it is real
The primary dcreening operation of railway information is showed, and convenience can be provided for subsequent further data analysis.
The present invention provides a kind of rds data processing unit based on artificial intelligence, as shown in figure 4, described device includes:
Module 201 is laid, for laying multiple cameras along rail, each camera is directed toward a camera note
Record, the camera record includes camera identification, the rail position where camera, the rail mark of the rail where camera
Know;
Video management mapping table constructs module 203, for constructing video management mapping table, the video management mapping table note
Record has the corresponding relationship between the video storage file folder that camera records and camera record is directed toward, each video storage file
Folder record has at least one videograph of its corresponding camera shooting;
Rds data process instruction obtains module 205, for obtaining rds data process instruction, the data processing instructions
It is directed toward the instruction of at least one Video segmentation, each Video segmentation instruction includes that a camera record and the camera record phase
At least one videograph number closed;
Videograph collection extraction module 207, for extracting videograph collection according to the data processing instructions;
Processing module 209, the record for concentrating to the videograph are handled.
Further, the videograph collection extraction module, comprising:
Video segmentation command unit, for obtaining each Video segmentation instruction;
Videograph selecting unit is stored for instructing acquisition camera to record corresponding video according to each Video segmentation
File, and the selection target video from video storage file folder, the number of the target video is by the Video segmentation
Instruction hit;
Videograph collection generation unit, for being made of video note the target video instructed according to each Video segmentation
Record collection.
Further, the processing module, comprising:
Video pre-filtering Rule unit instructs corresponding video pre-filtering rule for obtaining each Video segmentation;
Sort out unit, the videograph for concentrating according to the video pre-filtering rule to the videograph is returned
Class, each categorization results are directed toward unique video pre-filtering rule;
Videograph acquiring unit, for being remembered according to the video in its corresponding categorization results of video pre-filtering rule process
Record.
A kind of rds data processing unit based on artificial intelligence disclosed by the embodiments of the present invention is based on embodiment of the method
Identical inventive concept.
The above disclosure is only the preferred embodiments of the present invention, cannot limit the right model of the present invention with this certainly
It encloses, therefore equivalent changes made in accordance with the claims of the present invention, is still within the scope of the present invention.
Claims (8)
1. a kind of rds data processing method based on artificial intelligence, which is characterized in that the described method includes:
Multiple cameras are laid along rail, each camera is directed toward a camera record, and the camera record includes
Camera identification, the rail position where camera, the rail mark of the rail where camera;
Construct video management mapping table, the view that the video management mapping table record has camera record and camera record to be directed toward
Corresponding relationship between frequency storage folder, each video storage file folder record have at least the one of its corresponding camera shooting
A videograph;
Rds data process instruction is obtained, the data processing instructions are directed toward the instruction of at least one Video segmentation, each video point
Cutting instruction includes that a camera record and the camera record relevant at least one videograph number;
Videograph collection is extracted according to the data processing instructions;
The record concentrated to the videograph is handled.
2. the method according to claim 1, wherein described extract videograph according to the data processing instructions
Collection, comprising:
Obtain each Video segmentation instruction;
It instructs acquisition camera to record corresponding video storage file folder according to each Video segmentation, and stores text from the video
Selection target video in part folder, the number of the target video is by the Video segmentation instruction hit;
Videograph collection is constituted by the target video instructed according to each Video segmentation.
3. the method according to claim 1, wherein at the record concentrated to the videograph
Reason, comprising:
It obtains each Video segmentation and instructs corresponding video pre-filtering rule;
Sorted out according to the videograph that the video pre-filtering rule concentrates the videograph, each categorization results refer to
To unique video pre-filtering rule;
According to the videograph in its corresponding categorization results of video pre-filtering rule process.
4. according to the method described in claim 1, it is characterized by:
Video pre-filtering rule is for pre-processing videograph, in order to obtain feeling in Target Photo based on videograph
Interest region can satisfy the basic demand for carrying out image analysis, and the basic demand includes the railway information of area-of-interest
Plentiful degree is met the requirements and/or area-of-interest fragment requirement.
5. according to the method described in claim 4, it is characterized in that, described according to video pre-filtering rule process, its corresponding is returned
Videograph in class result, comprising:
Video intercepting is carried out to obtain to the videograph in the categorization results according to the interception rule in video pre-filtering rule
To pictures to be processed;
It is described wait locate to obtain that each picture to be processed in the pictures to be processed is inputted into preset image processing model
The area-of-interest recommendation results of picture are managed, the area-of-interest recommendation results include the area-of-interest in picture to be processed
And the corresponding index interested of area-of-interest;
According to the area-of-interest of area-of-interest processing rule and each picture to be processed in the video pre-filtering rule
Recommendation results obtain the corresponding Target Photo of each picture to be processed;
The corresponding Target Photo collection of the pictures to be processed is obtained according to the corresponding Target Photo of each picture to be processed.
6. a kind of rds data processing unit based on artificial intelligence, which is characterized in that described device includes:
Module is laid, for laying multiple cameras along rail, each camera is directed toward a camera record, described to take the photograph
As head record includes camera identification, the rail position where camera, the rail mark of the rail where camera;
Video management mapping table constructs module, and for constructing video management mapping table, the video management mapping table record is taken the photograph
The corresponding relationship between video storage file folder being directed toward as head record and camera record, each video storage file folder record
There is at least one videograph of its corresponding camera shooting;
Rds data process instruction obtains module, and for obtaining rds data process instruction, the data processing instructions are pointed to
Few Video segmentation instruction, each Video segmentation instruction includes a camera record and the camera record it is relevant extremely
Few videograph number;
Videograph collection extraction module, for extracting videograph collection according to the data processing instructions;
Processing module, the record for concentrating to the videograph are handled.
7. device according to claim 6, which is characterized in that the videograph collection extraction module, comprising:
Video segmentation command unit, for obtaining each Video segmentation instruction;
Videograph selecting unit, for instructing acquisition camera to record corresponding video storage file according to each Video segmentation
Folder, and the selection target video from video storage file folder, the number of the target video are instructed by the Video segmentation
Hit;
Videograph collection generation unit, for constituting videograph by the target video instructed according to each Video segmentation
Collection.
8. device according to claim 6, which is characterized in that the processing module, comprising:
Video pre-filtering Rule unit instructs corresponding video pre-filtering rule for obtaining each Video segmentation;
Sort out unit, the videograph for being concentrated according to the video pre-filtering rule to the videograph is sorted out,
Each categorization results are directed toward unique video pre-filtering rule;
Videograph acquiring unit, for according to the videograph in its corresponding categorization results of video pre-filtering rule process.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910790317.2A CN110502663A (en) | 2019-08-26 | 2019-08-26 | A kind of rds data processing method and processing device based on artificial intelligence |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910790317.2A CN110502663A (en) | 2019-08-26 | 2019-08-26 | A kind of rds data processing method and processing device based on artificial intelligence |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110502663A true CN110502663A (en) | 2019-11-26 |
Family
ID=68589466
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910790317.2A Pending CN110502663A (en) | 2019-08-26 | 2019-08-26 | A kind of rds data processing method and processing device based on artificial intelligence |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110502663A (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070083617A1 (en) * | 2005-10-07 | 2007-04-12 | Infosys Technologies | Video on demand system and methods thereof |
CN104867161A (en) * | 2015-05-14 | 2015-08-26 | 国家电网公司 | Video-processing method and device |
CN105893510A (en) * | 2016-03-30 | 2016-08-24 | 北京格灵深瞳信息技术有限公司 | Video structurization system and target search method thereof |
CN105976392A (en) * | 2016-05-30 | 2016-09-28 | 北京智芯原动科技有限公司 | Maximum-output-probability-based vehicle tyre detection method and apparatus |
CN107493458A (en) * | 2017-09-13 | 2017-12-19 | 郑州云海信息技术有限公司 | A kind of method and system for accessing monitor video |
CN107562888A (en) * | 2017-09-05 | 2018-01-09 | 北京瑞凯软件科技开发有限公司 | The data sheet generation method and system of a kind of track traffic synthetic monitoring system |
CN109697424A (en) * | 2018-12-19 | 2019-04-30 | 浙江大学 | A kind of high-speed railway impurity intrusion detection device and method based on FPGA and deep learning |
-
2019
- 2019-08-26 CN CN201910790317.2A patent/CN110502663A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070083617A1 (en) * | 2005-10-07 | 2007-04-12 | Infosys Technologies | Video on demand system and methods thereof |
CN104867161A (en) * | 2015-05-14 | 2015-08-26 | 国家电网公司 | Video-processing method and device |
CN105893510A (en) * | 2016-03-30 | 2016-08-24 | 北京格灵深瞳信息技术有限公司 | Video structurization system and target search method thereof |
CN105976392A (en) * | 2016-05-30 | 2016-09-28 | 北京智芯原动科技有限公司 | Maximum-output-probability-based vehicle tyre detection method and apparatus |
CN107562888A (en) * | 2017-09-05 | 2018-01-09 | 北京瑞凯软件科技开发有限公司 | The data sheet generation method and system of a kind of track traffic synthetic monitoring system |
CN107493458A (en) * | 2017-09-13 | 2017-12-19 | 郑州云海信息技术有限公司 | A kind of method and system for accessing monitor video |
CN109697424A (en) * | 2018-12-19 | 2019-04-30 | 浙江大学 | A kind of high-speed railway impurity intrusion detection device and method based on FPGA and deep learning |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Mechrez et al. | Photorealistic style transfer with screened poisson equation | |
CN110188635A (en) | A kind of plant pest recognition methods based on attention mechanism and multi-level convolution feature | |
DE112016005006T5 (en) | AUTOMATIC VIDEO EXECUTIVE SUMMARY | |
CN106295584A (en) | Depth migration study is in the recognition methods of crowd's attribute | |
CN106021406B (en) | A kind of online mask method of iterative image of data-driven | |
CN110321956B (en) | Grass pest control method and device based on artificial intelligence | |
CN107784118A (en) | A kind of Video Key information extracting system semantic for user interest | |
CN110503047A (en) | A kind of rds data processing method and processing device based on machine learning | |
CN112132197A (en) | Model training method, image processing method, device, computer equipment and storage medium | |
CN106127247A (en) | Image classification method based on multitask many examples support vector machine | |
Wang et al. | Multiscale deep alternative neural network for large-scale video classification | |
CN105956631A (en) | On-line progressive image classification method facing electronic image base | |
CN109583507A (en) | A kind of pig body recognition methods based on depth convolutional neural networks | |
CN109389555A (en) | A kind of Panorama Mosaic method and device | |
Rong et al. | Pest Identification and Counting of Yellow Plate in Field Based on Improved Mask R‐CNN | |
CN110363233A (en) | A kind of the fine granularity image-recognizing method and system of the convolutional neural networks based on block detector and Fusion Features | |
CN104680193A (en) | Online target classification method and system based on fast similarity network fusion algorithm | |
CN105957124A (en) | Method and device for color editing of natural image with repetitive scene elements | |
CN109523499A (en) | A kind of multi-source fusion full-view modeling method based on crowdsourcing | |
CN107992937A (en) | Unstructured data decision method and device based on deep learning | |
CN111191531A (en) | Rapid pedestrian detection method and system | |
CN114898359B (en) | Litchi plant diseases and insect pests detection method based on improvement EFFICIENTDET | |
CN114170137B (en) | Pepper disease identification method, identification system and computer readable storage medium | |
CN110188780A (en) | Method and device for constructing deep learning model for positioning multi-target feature points | |
CN107506362A (en) | Image classification based on customer group optimization imitates brain storage method |
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: 20191126 |