CN109063669A - A kind of bridge zone ship's navigation Situation analysis method and device based on image recognition - Google Patents
A kind of bridge zone ship's navigation Situation analysis method and device based on image recognition Download PDFInfo
- Publication number
- CN109063669A CN109063669A CN201810934480.7A CN201810934480A CN109063669A CN 109063669 A CN109063669 A CN 109063669A CN 201810934480 A CN201810934480 A CN 201810934480A CN 109063669 A CN109063669 A CN 109063669A
- Authority
- CN
- China
- Prior art keywords
- ship
- navigation
- situation
- feature
- image
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- 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
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
- G06V10/446—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering using Haar-like filters, e.g. using integral image techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
- G06V10/507—Summing image-intensity values; Histogram projection analysis
-
- 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
- G06V2201/08—Detecting or categorising vehicles
Abstract
The invention discloses a kind of bridge zone ship's navigation Situation analysis method based on image recognition, comprising the following steps: obtain the underway sailing situation feature of ship history, according to the underway sailing situation feature of ship history, construct ship's navigation situation database；To ship to be detected, the vessel position in the monitor video image of navigation channel is detected using Viola-Jones frame；The AAM positioning feature point model merged based on diverse characteristics is established, the sailing situation feature of the ship to be detected in navigation channel monitor video image is identified by AAM positioning feature point model；The multiple features fusion model based on support vector machines is established, the ship situation feature to be detected identified is matched with ship's navigation situation database by multiple features fusion model；According to matching result, the sailing situation during ship's navigation to be detected is measured in real time and is alarmed.Present invention reduces the risks that ship is navigated by water in bridge zone, provide reference frame and safety guarantee for ship's navigation manipulation.
Description
Technical field
The present invention relates to the monitoring of ship's navigation situation and field of image recognition more particularly to a kind of bridges based on image recognition
Area's ship's navigation Situation analysis method and device.
Background technique
Ship can be met in bridge zone, mutually the sailing situations hair such as position, speed, course and posture of ship during evacuation etc.
It is raw to change, and then ship is caused easily to be in jeopardy and complicated operational configuration when the navigation of bridge zone.Bridge approach is littoral
Picture pick-up device carries out real time monitoring to ship and obtains video image with video recording, then carries out image knowledge using computer software model
It not with analysis, and is compared with ship's navigation situation database, judges whether ship is in the hole.This method can be
It is offered an opinion and is referred to by bridge zone to safety of ship from now on.
Image recognition technology includes infrared technique, sensor technology etc., is registrated with landform in navigation, map, is provided naturally
The important application value in many fields such as source analysis, weather forecast, environmental monitoring, physiology studies of lesions.The technology is suitable for water
The ship's navigation situation identification that face is run at a low speed.
GPS positioning, track analysis instrument and Track Plotter are utilized currently, focusing primarily upon to the research of ship's navigation Study on Trend
It records and analyzes, electronic Chart Display etc..But these study limitations obtain related data in needing ship independently to issue signal.
Image recognition technology is utilized on basis herein, in conjunction with flow direction, waterway width, the environmental informations such as bridge width realize ship course
It automatically tracking, detects accommodation of the various ships by bridge zone, the speed of a ship or plane, the sailing situations such as course, it is ensured that safety of ship passes through bridge zone,
Touching bridge is avoided to touch pier.
Summary of the invention
How the technical problem to be solved by the present invention is to realize positioning to by the navigating ship of bridge zone, and identify ship
Sailing situation, reduce the risk navigated by water in bridge zone of ship；A kind of bridge zone ship's navigation situation based on image recognition point is provided
Analyse method and device.
The technical solution adopted by the present invention to solve the technical problems is:
The present invention provides a kind of bridge zone ship's navigation Situation analysis method based on image recognition, comprising the following steps:
Step 1: the underway sailing situation feature of ship history is obtained, according to the underway sailing situation of ship history
Feature constructs ship's navigation situation database；
Step 2: navigation channel monitor video image is obtained, to ship to be detected, using Viola-Jones frame to navigation channel
Vessel position in monitor video image is detected；
Step 3: establishing the AAM positioning feature point model merged based on diverse characteristics, pass through AAM positioning feature point model
Identify the sailing situation feature of the ship to be detected in navigation channel monitor video image；
Step 4: the multiple features fusion model based on support vector machines is established, by multiple features fusion model to identifying
The ship situation feature to be detected come is matched with ship's navigation situation database；
Step 5: the sailing situation during ship's navigation to be detected is measured in real time and is reported according to matching result
It is alert.
Further, ship boat is constructed according to the underway sailing situation feature of ship history in step one of the invention
Row situation database method particularly includes:
Ship's navigation situation refers to the practical fortune of ship that ship is influenced under objective environment by various interior external force and is formed
Dynamic trend and state and the subsequent variation for including；By collecting ship sailing situation relevant historical data, according to ship and week
The relative motion relation of ship, shoal, harbour as the determining ship of comparison of object of reference is enclosed, relative motion relation includes ship
Transversely or longitudinally motion information, azimuth-range information；According to vessel position, navigation course and with object of reference or
The range information of barrier constructs ship's navigation situation database.
Further, utilize Viola-Jones frame to the ship in the monitor video image of navigation channel in step two of the invention
Berth, which is set, to be detected, method particularly includes:
Vessel position in the monitor video image of navigation channel is detected, to navigation channel monitor video image carry out gray processing and
Pros' figure equalization pretreatment, carries out image enhancement processing to determining ship image, carries out gray processing and ship image
The blurring and enhancing of characteristic portion are handled, and are carried out the operation of illumination weighing apparatusization to image, are removed background, the influence of illumination factor；
The Haar characteristic value of image is extracted with Viola-Jones algorithm, Haar characteristic value uses feature templates, every kind of mould
Plate includes black-and-white two color rectangle, the pixel in white rectangle frame and subtracts pixel in grey rectangle frame and obtains characteristic value, really
Determine the adjacent rectangular area of ship specific position, ship's particulars is quantified to distinguish ship and non-ship's particulars；
Then vessel position is further accurately positioned using integral image techniques and acceleration classifier.
Further, step three of the invention kind is identified in navigation channel monitor video image by AAM positioning feature point model
Ship to be detected sailing situation feature, method particularly includes:
Ship's navigation situation feature is further analyzed and extracted on the basis of vessel position detection, is calculated in conjunction with Canny
Son, histograms of oriented gradients HOG and local binary patterns LBP establish the ship's navigation situation identification model of diverse characteristics fusion,
Ship's navigation situation feature is accurately positioned finally by AAM positioning feature point algorithm.
Further, step three of the invention method particularly includes:
(1) noise is removed come smoothed image based on Canny operator application gaussian filtering；
(2) intensity gradient that image is looked for based on Canny operator, using first difference operator calculating horizontal direction and vertically
The gradient magnitude component in direction, the maximum point of range value is image border point on gradient direction；
(3) ship's navigation situation feature is extracted using local binary patterns LBP, with the strong LBP operator counterweight of description power
The textural characteristics in region are wanted to extract；LBP algorithm carries out the image feature value after feature extraction to be indicated by histogram；
(4) by calculating with statistical picture gradient orientation histogram come construction feature, ship's navigation is depicted by method
Bright spot, the subtle mode in part and its distribution situation of dim spot, edge of situation image；
(5) the AAM positioning feature point algorithm based on diverse characteristics fusion, extracts ship important area feature, navigates in ship
Characteristic point is demarcated on row image, is classified to realize to ship's navigation situation.
Further, pass through multiple features fusion model in step four of the invention to the ship situation to be detected identified
Feature is matched with ship's navigation situation database, method particularly includes:
On the basis of ship's navigation situation identification, ship's navigation situation feature codomain is drawn using support vector machines
Point, the feature that Classification and Identification extracts finally matches sailing situation feature with the feature of lane database, examines ship
Sailing situation accurately identifies the validity and reliability of model.
Further, according to matching result in step five of the invention, to the navigation state during ship's navigation to be detected
Gesture is measured in real time and alarms, method particularly includes:
It alarms ship's navigation situation precarious position information, alarms if identifying that ship belongs to dangerous operational configuration
System will issue alarm signal.
Further, the alarm signal in step five of the invention includes: super drinking water, course is dangerous and navigates beyond safety
Speed.
The present invention provides a kind of bridge zone ship's navigation Study on Trend device based on image recognition, which includes:
Ship's navigation situation database, for obtaining the underway sailing situation feature of ship history, according to ship history
Underway sailing situation feature constructs ship's navigation situation database；
Image collecting device, for acquiring the navigation channel monitor video image during ship's navigation；
Pattern recognition device utilizes Viola-Jones to ship to be detected for obtaining navigation channel monitor video image
Frame detects the vessel position in the monitor video image of navigation channel；It is fixed to establish the AAM characteristic point merged based on diverse characteristics
Bit model identifies the sailing situation feature of the ship to be detected in navigation channel monitor video image by AAM positioning feature point model；
Image matching apparatus passes through multiple features fusion mould for establishing the multiple features fusion model based on support vector machines
Type matches the ship situation feature to be detected identified with ship's navigation situation database；
Warning device, for being examined in real time to the sailing situation during ship's navigation to be detected according to matching result
It surveys and alarms.
The beneficial effect comprise that: the bridge zone ship's navigation Situation analysis method of the invention based on image recognition
And device, the identification new method of bridge zone ship's navigation situation is provided, by monitoring using Viola-Jones frame to navigation channel
Vessel position in video image is detected, and the accuracy of identification of vessel position can be greatly improved；It is merged by diverse characteristics
AAM positioning feature point model, can effectively identify the sailing situation feature of ship to be detected；Pass through multiple features fusion mould again
Type matches the ship situation feature to be detected identified with ship's navigation situation database, can identify ship's navigation
Precarious position in the process reduces the risk that ship is navigated by water in bridge zone, provides reference frame and safety for ship's navigation manipulation
It ensures.
Detailed description of the invention
Present invention will be further explained below with reference to the attached drawings and examples, in attached drawing:
Fig. 1 is ship's navigation Study on Trend process flow diagram flow chart in the present invention；
Ship's navigation Study on Trend equipment drawing in the present invention of the position Fig. 2；
Fig. 3 is ship's navigation situation fixation and recognition figure in the present invention；
Fig. 4 is ship's navigation situation recognition methods figure in the present invention.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that described herein, specific examples are only used to explain the present invention, not
For limiting the present invention.
As shown in Figure 1, the invention discloses the bridge zone ship's navigation Situation analysis methods based on image recognition to set with analysis
It is standby, the identification new method of bridge zone ship's navigation situation is proposed, the risk that ship is navigated by water in bridge zone is reduced, is grasped for ship's navigation
It is vertical that reference frame and safety guarantee are provided.The specific steps are that:
Step 1: according to the sailing situation feature of ship, ship's navigation situation database is constructed.To by bridge zone peace
The full depth of water, the type of safety lane width and ship, safe draft, the data such as safety speed are for statistical analysis, establish logical
Cross the ship's navigation situation database of bridge zone.Ship Types identification is not only contributed in this way, additionally it is possible to be convenient for bridge zone shipping transport
Management.Fig. 2 is ship's navigation Study on Trend equipment drawing.
Step 2: the vessel position in the monitor video image of navigation channel is detected using Viola-Jones frame.Pass through
Image collecting device carries out camera shooting to the ship by bridge zone and takes pictures if navigation channel monitors laser night vision fog-penetrating camera.Then right
The picture of interception is pre-processed.Since shooting photo is influenced by factors such as illumination, photo size is not also identical, but is
Mitigate the influence due to many factors, avoids ship's fix from malfunctioning, need to pre-process the picture of acquisition first.Pre- place
The content of reason includes picture gray processing and equalization.The image of various formats is generally first transformed into gray scale by Digital Image Processing kind
Image is so that the calculation amount of subsequent image becomes less, and only the variation of gray scale power is without the variation of color.Histogram
Figure equalization refers to by constructing tonal gradation, the histogram of original image is transformed, the region for concentrating original image tonal gradation is drawn
It opens or keeps intensity profile uniform, to increase contrast, keep the details of image clear, achieve the purpose that enhancing.Next it just uses
The Haar characteristic value of Viola-Jones algorithm extraction picture.Then with Viola-Jones algorithm integral image techniques into
Row accelerates feature calculation, the position of ship in picture is accurately positioned.
In formula, ii indicates integrogram, and i is original image, and ii (x, y) means that two angle steel joints of i (1,1) and i (x, y) are encircled a city
Rectangular area size.
By the coordinate on four vertex of the calculated vessel position histogram of previous section, come with simple geometric knowledge
The central point for further obtaining vessel area frame, is then extended out on the basis of central point according to a certain percentage around again,
Frame is selected to be adjusted ship, to realize that ship's fix picture intercepts.Unitized processing is carried out to the picture of interception, all
Being converted into pixel is 160 × 250, the picture that size is 16 × 16.
Step 3: it establishes in the AAM positioning feature point model identification navigation channel monitor video image merged based on diverse characteristics
Ship's navigation situation feature.
1. removing noise based on Canny operator application gaussian filtering come smoothed image.Here consider picture denoising effect and
The indexs such as computational efficiency realize the noise processing of acquisition image by the Wiener filtering of airspace median filtering and frequency domain is used.Frequently
The basic principle of the Wiener filtering in domain is exactly the mean square deviation minimum for making original image f (x, y) and replying original image g (x, y).It is public
Formula is as follows:
First to the local mean value μ and variances sigma of pixel2It is calculated.N in formula1And n2Indicate the coordinate position of pixel, a
(n1, n2) be the point gray value.MN indicates that reference templates image size is M × N.
B in formula is the true gray value of Image estimation, v2For the global variance of image.Picture is converted into grayscale image,
Increase Gaussian noise.
2. looking for the intensity gradient (intensity gradients) of image based on Canny operator, calculated using first-order difference
Son calculates gradient magnitude component horizontally and vertically, to obtain amplitude M and the direction of gradient of the gradient of image
E calculates partial gradient range value and edge direction to each pixel in image to improve the computational accuracy of gradient.?
The maximum point of range value is image border point on gradient direction.By the extraction to image border, the accommodation etc. of ship is obtained
Information.
3. ship's navigation situation feature is extracted using local binary patterns (LBP), with the strong LBP operator pair of description power
The sailing situation feature of important area extracts.
The basic principle of LBP description is exactly the gray scale size of Correlation Centre pixel Yu its field pixel, between Boolean type
Function is stated as a result, the LBP value of so textural characteristics can be stated are as follows:
Wherein gi (i=1 ..., N) is indicated with gcCentered on circle domain N number of pixel, R be circle domain radius.
The characteristic value that LBP algorithm carries out the image f (x, y) after feature extraction is indicated with histogram:
Finally the statistic histogram each of obtained is attached as a feature vector, that is, whole picture figure
LBP texture feature vector.
4. HOG be not easy it is affected by noise, it by calculate and statistical picture gradient orientation histogram come construction feature.It should
Method can depict the subtle mode in the part such as bright spot, dim spot, edge of ship's navigation situation image and its distribution situation, to light
There is stronger robustness according to equal environmental changes.
Determine that the sliding window of a n × n, sliding window are slided on the entire image, to extract ship's navigation
The HOG feature of situation.Sliding window is uniformly divided into several fritters, calculates the ladder of each pixel on each fritter respectively
Spend amplitude and direction.Horizontal direction gradient, vertical gradient and pixel value at the pixel, the then gradient of the pixel
Are as follows:
Gx(x0)=H (x5)-H(x4)
Gy(x0)=H (x7)-H(x2)
Then the gradient magnitude at pixel 0 and gradient direction are respectively as follows:
Finally all fritters in detection window are carried out with the collection of HOG feature, the histogram of each fritter of concatenated in order
Feature, sliding window completes the extraction of image HOG feature in whole image.
5. ship important area feature is extracted, in ship's navigation based on the AAM positioning feature point algorithm of diverse characteristics fusion
Characteristic point is demarcated on image, is classified to realize to ship's navigation situation.
In conjunction with Canny operator contour feature, local binary patterns feature and histograms of oriented gradients model come at the beginning of AAM
Beginning form parameter does the improvement optimized and to the upgrading of AAM matching template.Modified hydrothermal process using characteristic point partial model and
AAM world model in conjunction with method optimize AAM original shape parameter, and herein under the premise of AAM matching template is risen
Grade, makes it closer to the information of image to be matched.It is final to extract under accurate matching template and the original shape parameter of optimization
The precision of image and database matching can get a promotion.
Step 4: the multiple features fusion model based on support vector machines is established to the ship situation feature and ship identified
Oceangoing ship sailing situation database is matched.Algorithm of support vector machine is for solving to have a small amount of sample but the very high classification of dimension
Problem effect is fine, for example ship's navigation situation identifies problem.After being trained with support vector machines, the ship of interception will be framed
It navigates by water image to handle by gray processing, carries out match check with the feature inside ship's navigation situation database.Fig. 3 is the present invention
Middle ship's navigation situation fixation and recognition figure.
Step 5: ship's navigation situation alarm system is established.If the identification successful match in step 4, illustrates at ship
In security postures, alarm system will will do it alarm if it fails to match, and identification ship is reminded to be careful not to blind navigation process
Bridge zone.
It should be understood that for those of ordinary skills, it can be modified or changed according to the above description,
And all these modifications and variations should all belong to the protection domain of appended claims of the present invention.
Claims (9)
1. a kind of bridge zone ship's navigation Situation analysis method based on image recognition, which comprises the following steps:
Step 1: the underway sailing situation feature of ship history is obtained, according to the underway sailing situation feature of ship history,
Construct ship's navigation situation database；
Step 2: obtaining navigation channel monitor video image, to ship to be detected, navigation channel is monitored using Viola-Jones frame
Vessel position in video image is detected；
Step 3: establishing the AAM positioning feature point model merged based on diverse characteristics, identified by AAM positioning feature point model
The sailing situation feature of ship to be detected in the monitor video image of navigation channel；
Step 4: establish multiple features fusion model based on support vector machines, by multiple features fusion model to identifying
Ship situation feature to be detected is matched with ship's navigation situation database；
Step 5: the sailing situation during ship's navigation to be detected is measured in real time and is alarmed according to matching result.
2. the bridge zone ship's navigation Situation analysis method according to claim 1 based on image recognition, which is characterized in that step
According to the underway sailing situation feature of ship history in rapid one, ship's navigation situation database is constructed method particularly includes:
The ship actual motion that ship's navigation situation refers to that ship is influenced and formed by various interior external force under objective environment becomes
Gesture and state and the subsequent variation for including；By collecting ship sailing situation relevant historical data, according to ship and surrounding ship
The relative motion relation of oceangoing ship, shoal, harbour as the determining ship of comparison of object of reference, relative motion relation includes the transverse direction of ship
Or longitudinal movement information, azimuth-range information；According to vessel position, navigation course and with object of reference or obstacle
The range information of object constructs ship's navigation situation database.
3. the bridge zone ship's navigation Situation analysis method according to claim 1 based on image recognition, which is characterized in that step
The vessel position in the monitor video image of navigation channel is detected using Viola-Jones frame in rapid two, method particularly includes:
Vessel position in the monitor video image of navigation channel is detected, gray processing and pros are carried out to navigation channel monitor video image
Figure equalization pretreatment, carries out image enhancement processing to determining ship image, carries out the feature of gray processing and ship image
The blurring and enhancing at position are handled, and are carried out the operation of illumination weighing apparatusization to image, are removed background, the influence of illumination factor；
The Haar characteristic value of image is extracted with Viola-Jones algorithm, Haar characteristic value uses feature templates, every kind of template packet
Black-and-white two color rectangle is included, the pixel in white rectangle frame and pixel in grey rectangle frame is subtracted and obtains characteristic value, determines ship
The adjacent rectangular area of oceangoing ship specific position quantifies ship's particulars to distinguish ship and non-ship's particulars；
Then vessel position is further accurately positioned using integral image techniques and acceleration classifier.
4. the bridge zone ship's navigation Situation analysis method according to claim 1 based on image recognition, which is characterized in that step
The rapid three kinds sailing situation features by the ship to be detected in AAM positioning feature point model identification navigation channel monitor video image,
Its method particularly includes:
Ship's navigation situation feature is further analyzed and extracts on the basis of vessel position detection, in conjunction with Canny operator, side
The ship's navigation situation identification model that diverse characteristics merge is established to histogram of gradients HOG and local binary patterns LBP, is finally led to
AAM positioning feature point algorithm is crossed to be accurately positioned ship's navigation situation feature.
5. the bridge zone ship's navigation Situation analysis method according to claim 4 based on image recognition, which is characterized in that step
Rapid three method particularly includes:
(1) noise is removed come smoothed image based on Canny operator application gaussian filtering；
(2) intensity gradient that image is looked for based on Canny operator is calculated horizontally and vertically using first difference operator
Gradient magnitude component, the maximum point of range value is image border point on gradient direction；
(3) ship's navigation situation feature is extracted using local binary patterns LBP, with the strong LBP operator of description power to important area
The textural characteristics in domain extract；LBP algorithm carries out the image feature value after feature extraction to be indicated by histogram；
(4) by calculating with statistical picture gradient orientation histogram come construction feature, ship's navigation situation is depicted by method
Bright spot, the subtle mode in part and its distribution situation of dim spot, edge of image；
(5) the AAM positioning feature point algorithm based on diverse characteristics fusion, extracts ship important area feature, in ship's navigation figure
Characteristic point is demarcated as on, is classified to realize to ship's navigation situation.
6. the bridge zone ship's navigation Situation analysis method according to claim 1 based on image recognition, which is characterized in that step
In rapid four by multiple features fusion model to the ship situation feature to be detected that identifies and ship's navigation situation database into
Row matching, method particularly includes:
On the basis of ship's navigation situation identification, ship's navigation situation feature codomain is divided using support vector machines, point
Class identifies the feature extracted, finally matches sailing situation feature with the feature of lane database, examines ship's navigation
Situation accurately identifies the validity and reliability of model.
7. the bridge zone ship's navigation Situation analysis method according to claim 1 based on image recognition, which is characterized in that step
According to matching result in rapid five, the sailing situation during ship's navigation to be detected is measured in real time and is alarmed, it is specific
Method are as follows:
It alarms ship's navigation situation precarious position information, the alarm system if identifying that ship belongs to dangerous operational configuration
Alarm signal will be issued.
8. the bridge zone ship's navigation Situation analysis method according to claim 7 based on image recognition, which is characterized in that step
Alarm signal in rapid five includes: super drinking water, course is dangerous and exceeds safety speed.
9. a kind of bridge zone ship's navigation Study on Trend device based on image recognition, which is characterized in that the device includes:
Ship's navigation situation database is navigated by water for obtaining the underway sailing situation feature of ship history according to ship history
In sailing situation feature, construct ship's navigation situation database；
Image collecting device, for acquiring the navigation channel monitor video image during ship's navigation；
Pattern recognition device, to ship to be detected, utilizes Viola-Jones frame for obtaining navigation channel monitor video image
Vessel position in the monitor video image of navigation channel is detected；Establish the AAM positioning feature point mould merged based on diverse characteristics
Type identifies the sailing situation feature of the ship to be detected in navigation channel monitor video image by AAM positioning feature point model；
Image matching apparatus passes through multiple features fusion model pair for establishing the multiple features fusion model based on support vector machines
The ship situation feature to be detected identified is matched with ship's navigation situation database；
Warning device, for being measured in real time to the sailing situation during ship's navigation to be detected according to matching result and
Alarm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810934480.7A CN109063669B (en) | 2018-08-16 | 2018-08-16 | Bridge area ship navigation situation analysis method and device based on image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810934480.7A CN109063669B (en) | 2018-08-16 | 2018-08-16 | Bridge area ship navigation situation analysis method and device based on image recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109063669A true CN109063669A (en) | 2018-12-21 |
| CN109063669B CN109063669B (en) | 2022-06-03 |
Family
ID=64687112
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810934480.7A Active CN109063669B (en) | 2018-08-16 | 2018-08-16 | Bridge area ship navigation situation analysis method and device based on image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109063669B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109683618A (en) * | 2018-12-28 | 2019-04-26 | 广州中国科学院沈阳自动化研究所分所 | A kind of navigational signals identifying system and its recognition methods |
| CN110619368A (en) * | 2019-09-23 | 2019-12-27 | 北京理工大学 | Planet surface navigation feature imaging matching detection method |
| CN112307026A (en) * | 2020-10-29 | 2021-02-02 | 广东海洋大学 | Method for establishing small ship navigation multi-source information real-time database |
| CN117875949A (en) * | 2024-03-13 | 2024-04-12 | 山东交通学院 | Intelligent bridge apparent disease detection method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN203786891U (en) * | 2014-04-11 | 2014-08-20 | 中国人民解放军61139部队 | Mobile monitoring convergence platform based on GIS (Geographic Information System) |
| CN106816038A (en) * | 2017-03-17 | 2017-06-09 | 武汉理工大学 | A kind of inland waters abnormal behaviour ship automatic identification system and method |
| CN107704865A (en) * | 2017-05-09 | 2018-02-16 | 北京航空航天大学 | Fleet Targets Detection based on the extraction of structure forest edge candidate region |
-
2018
- 2018-08-16 CN CN201810934480.7A patent/CN109063669B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN203786891U (en) * | 2014-04-11 | 2014-08-20 | 中国人民解放军61139部队 | Mobile monitoring convergence platform based on GIS (Geographic Information System) |
| CN106816038A (en) * | 2017-03-17 | 2017-06-09 | 武汉理工大学 | A kind of inland waters abnormal behaviour ship automatic identification system and method |
| CN107704865A (en) * | 2017-05-09 | 2018-02-16 | 北京航空航天大学 | Fleet Targets Detection based on the extraction of structure forest edge candidate region |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109683618A (en) * | 2018-12-28 | 2019-04-26 | 广州中国科学院沈阳自动化研究所分所 | A kind of navigational signals identifying system and its recognition methods |
| CN110619368A (en) * | 2019-09-23 | 2019-12-27 | 北京理工大学 | Planet surface navigation feature imaging matching detection method |
| CN112307026A (en) * | 2020-10-29 | 2021-02-02 | 广东海洋大学 | Method for establishing small ship navigation multi-source information real-time database |
| CN117875949A (en) * | 2024-03-13 | 2024-04-12 | 山东交通学院 | Intelligent bridge apparent disease detection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109063669B (en) | 2022-06-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107463890B (en) | A kind of Foregut fermenters and tracking based on monocular forward sight camera | |
| Alioua et al. | Driver’s fatigue detection based on yawning extraction | |
| CN109063669A (en) | A kind of bridge zone ship's navigation Situation analysis method and device based on image recognition | |
| Siriborvornratanakul | An automatic road distress visual inspection system using an onboard in-car camera | |
| CN101609504B (en) | Method for detecting, distinguishing and locating infrared imagery sea-surface target | |
| CN113592845A (en) | Defect detection method and device for battery coating and storage medium | |
| WO2016015547A1 (en) | Machine vision-based method and system for aircraft docking guidance and aircraft type identification | |
| Tayab Khan et al. | Smart real-time video surveillance platform for drowsiness detection based on eyelid closure | |
| CN103927751A (en) | Water surface optical visual image target area detection method based on gradient information fusion | |
| CN103310194A (en) | Method for detecting head and shoulders of pedestrian in video based on overhead pixel gradient direction | |
| CN112381870B (en) | Binocular vision-based ship identification and navigational speed measurement system and method | |
| CN111553214B (en) | Method and system for detecting smoking behavior of driver | |
| CN109002800A (en) | The real-time identification mechanism of objective and recognition methods based on Multi-sensor Fusion | |
| CN110321855A (en) | A kind of greasy weather detection prior-warning device | |
| CN110348307B (en) | Path edge identification method and system for crane metal structure climbing robot | |
| CN102129556A (en) | Judging method of definition of iris image | |
| Wang et al. | Real-time vehicle classification based on eigenface | |
| Chen | Road vehicle recognition algorithm in safety assistant driving based on artificial intelligence | |
| Guo et al. | Detection of occluded road signs on autonomous driving vehicles | |
| Diao et al. | Image sequence measures for automatic target tracking | |
| Huang et al. | A back propagation based real-time license plate recognition system | |
| Si-ming et al. | Moving shadow detection based on Susan algorithm | |
| Cheng et al. | Vehicle detection and classification using model-based and fuzzy logic approaches | |
| OUYANG et al. | Edge detection in pavement crack image with beamlet transform | |
| Bergevin et al. | Detection and characterization of junctions in a 2D image |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |