CN106203353B - A kind of detection system and method for undercarriage - Google Patents
A kind of detection system and method for undercarriage Download PDFInfo
- Publication number
- CN106203353B CN106203353B CN201610554236.9A CN201610554236A CN106203353B CN 106203353 B CN106203353 B CN 106203353B CN 201610554236 A CN201610554236 A CN 201610554236A CN 106203353 B CN106203353 B CN 106203353B
- Authority
- CN
- China
- Prior art keywords
- undercarriage
- image
- chain code
- feature
- starting point
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- 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
- 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
Abstract
The present invention provides the detection system and method for a kind of undercarriage;Wherein include in system, image pre-processing module, for being pre-processed to the undercarriage image to be detected;Aircraft Targets detection module realizes the detection of Aircraft Targets in image;Characteristic extracting module, for extracting the lower boundary feature of Aircraft Targets;Undercarriage suspicious region detection module realizes the detection in suspicious undercarriage region in Aircraft Targets;Undercarriage detection module realizes the detection of undercarriage.The present invention can be quickly detected whether undercarriage puts down, and be easy to Project Realization;Undercarriage detects accuracy height, has filled up the blank currently for this aspect research;Position and the quantity information of undercarriage can be provided, it is more stable.
Description
Technical field
The invention mainly relates to security protection technical field of data processing, and in particular to a kind of detection system of undercarriage
System and method.
Background technique
Before aircraft landing, to ensure that undercarriage has been put down, other than the indicating equipment on machine, also by control tower
If an observer is monitored with telescope, for this method due to being influenced by human factor and weather conditions, reliability is poor.In recent years
Come, because of airborne equipment failure and observer neglect one's duty flying grade accident caused by wrong diagnosis when have notification.If can be on airport
One undercarriage detection system is installed, detects whether undercarriage puts down automatically by the method for machine vision, it can not only
Aircraft accident caused by reducing because of observer's fault, and alleviate the intensity of manual labor.
Currently, the research whether domestic method detection undercarriage for using machine vision puts down is considerably less, almost locate
In blank stage." the intelligent undercarriage control monitoring system " delivered for domestic scholars 1998 in " Chinese journal of scientific instrument "
In, using aircraft and imaging sensor distance, aircraft length-width ratio, the feature vector of aircraft elevation angle composition aircraft, and using manually
Neural network judges whether undercarriage puts down.Judging characteristic used by this method is influenced very by aircraft flight attitude
Greatly, false alarm rate is high.In addition, this method can only determine that aircraft either with or without drop, can not detect the position of undercarriage
The quantity of undercarriage can not be provided.
Summary of the invention
Technical problem to be solved by the invention is to provide a kind of detection system of undercarriage and methods, meet airport
It is used to judge whether undercarriage puts down when aircraft landing, detection accuracy is high, can provide undercarriage position and
Quantity information.
The technical scheme to solve the above technical problems is that a kind of detection system of undercarriage, including figure
As preprocessing module, Aircraft Targets detection module, characteristic extracting module, undercarriage suspicious region detection module and undercarriage detection
Module,
Described image preprocessing module carries out color for acquiring the initial pictures of undercarriage, and to initial pictures
Processing and noise reduction process, the image after being optimized;
The Aircraft Targets detection module, for successively carrying out background inhibition processing to the image after optimization, at binaryzation
Reason and connected region detection, and the connected region by detecting extracts the Aircraft Targets regional scope in image;
The characteristic extracting module, for extracting the lower boundary in the Aircraft Targets regional scope, then extract it is described under
The chain code feature on boundary;
The undercarriage suspicious region detection module, for the starting point and chain code direction according to setting under Aircraft Targets
The chain code feature on boundary is encoded, and selectes the coding on specific direction according to chain code direction to detect whether that there are undercarriages
Suspicious region;
The undercarriage detection module, the judgement for extracting undercarriage suspicious region according to histograms of oriented gradients are special
Sign, and according to the authenticity for determining that feature determines the undercarriage suspicious region.
The beneficial effects of the present invention are: (1) carries out color, noise reduction optimization processing to initial pictures and examines to connected region
It surveys, can be quickly detected whether undercarriage puts down, be easy to Project Realization;(2) Aircraft Targets regional scope is following
Boundary's feature extraction and according to chain code angle detecting undercarriage suspicious region, so that undercarriage detection accuracy is high, has filled up and has worked as
The preceding blank for this aspect research;(3) undercarriage suspicious region detection module can provide the position sum number of undercarriage
Information is measured, it is more stable.
Based on the above technical solution, the present invention can also be improved as follows.
Further, described image preprocessing module includes:
Gray scale processing unit, for initial pictures to be converted to gray level image;
Filter processing unit, for gray level image to be carried out median filter process, the image after being optimized.
Beneficial effect using above-mentioned further scheme is: initial pictures being carried out gray proces and filtering processing, to subtract
The interference of few noise in image, the image that can be optimized are convenient for post-processing.
Further, the Aircraft Targets detection module includes:
Binary conversion treatment unit carries out background inhibition processing for the image after optimizing, and using adaptive threshold
Background is inhibited treated image to carry out binary conversion treatment by mode;
Connected area disposal$ unit, the image for that will pass through binary conversion treatment carries out connected domain detection, and extracts each company
The area of the area and mass center information in logical domain, the connected domain of the extraction is greater than given value and the smallest connected domain of mass center row coordinate
As Aircraft Targets regional scope.
Beneficial effect using above-mentioned further scheme is: the recall rate of Aircraft Targets is high, and runing time is short.
Further, the undercarriage suspicious region detection module includes:
Setup unit, for setting the starting point of chain code feature, and setting point on the right of the starting point is coding starting point,
And using starting point as initial point, digital coding is successively counterclockwise carried out around starting point, and using starting point as terminal, successively
It is encoded to 0 to 7 direction (totally 8 directions), to complete the setting in chain code direction;
Coding unit, for being carried out according to the starting point and chain code direction of setting to the chain code feature of Aircraft Targets lower boundary
Coding;
Suspicious region detection unit is specific direction for selecting 2 directions and 6 directions according to chain code direction, and according to 2 sides
Detect whether that there are undercarriage suspicious regions to maximum column, minimum column, maximum row, the minimum row on 6 direction positions.
Beneficial effect using above-mentioned further scheme is: realizing the detection of suspicious landing gear position in Aircraft Targets.
Further, the undercarriage detection module includes:
Determine feature extraction unit, the histograms of oriented gradients feature for extracting the undercarriage suspicious region, which is used as, to be sentenced
Determine feature;
Undercarriage detection unit, for determining in the judgement feature input limits learning machine classifier by extraction, and
The authenticity of institute's undercarriage suspicious region is determined according to the result that extreme learning machine classifier exports.
Using the beneficial effect of above-mentioned further scheme is: realize suspicious undercarriage whether sentencing for true undercarriage
Fixed, undercarriage determines that accuracy is high, and runing time is short.
Another technical solution that the present invention solves above-mentioned technical problem is as follows: a kind of detection method of undercarriage,
It is characterized in that, includes the following steps:
Step S1: the initial pictures of undercarriage are acquired, and colors countenance and noise reduction process are carried out to initial pictures, are obtained
Image after to optimization;
Step S2: binary conversion treatment and connected region detection are successively carried out to the image after optimization, and by detecting
Connected region extracts the Aircraft Targets regional scope in image;
Step S3: extracting the lower boundary in the Aircraft Targets regional scope, then extracts the chain code feature of the lower boundary;
Step S4: encoding the chain code feature of Aircraft Targets lower boundary according to the starting point of setting and chain code direction,
And the coding on specific direction is selected to detect whether that there are undercarriage suspicious regions according to chain code direction;
Step S5: the judgement feature of undercarriage suspicious region is extracted according to histograms of oriented gradients, and according to judgement feature
Determine the authenticity of the undercarriage suspicious region.
Further, the specific steps of step S1 are realized are as follows:
Step S101: initial pictures are converted into gray level image;
Step S102: gray level image is subjected to median filter process, the image after being optimized.
Further, the specific steps of step S2 are realized are as follows:
Step S201: the image after optimization is subjected to background inhibition processing, and by background by the way of adaptive threshold
Treated image is inhibited to carry out binary conversion treatment;
Step S202: the image Jing Guo binary conversion treatment is subjected to connected domain detection, and extracts the area of each connected domain
With mass center information, the area of the connected domain of the extraction is greater than given value and the smallest connected domain of mass center row coordinate as aircraft mesh
Mark regional scope.
Further, the specific steps of step S4 are realized are as follows:
Step S401: the starting point of setting chain code feature, and setting point on the right of the starting point is coding starting point, and with
Starting point is initial point, counterclockwise successively carries out digital coding around starting point, and using starting point as terminal, successively obtain 0
To 7 direction encodings, to complete the setting in chain code direction;
Step S402: the chain code feature of Aircraft Targets lower boundary is compiled according to the starting point of setting and chain code direction
Code;
Step S403: selecting 2 directions and 6 directions according to chain code direction is specific direction, and according to 2 directions and 6 direction positions
The maximum column set, minimum column, maximum row, minimum row detect whether that there are undercarriage suspicious regions.
Further, the specific steps of step S5 are realized are as follows:
Step S501: the histograms of oriented gradients feature of the undercarriage suspicious region is extracted as judgement feature;
Step S502: will determine in the judgement feature input limits learning machine classifier of extraction, and according to the limit
The result of habit machine classifier output determines the authenticity of institute's undercarriage suspicious region.
Detailed description of the invention
Fig. 1 is the module frame chart of present system;
Fig. 2 is the schematic diagram in present invention setting chain code direction;
Fig. 3 is the schematic diagram of the following boundary coding of the present invention;
Fig. 4 is the image after optimizing in the embodiment of the present invention;
Fig. 5 is the schematic diagram that 2 directions and 6 directions are detected in the embodiment of the present invention;
Fig. 6 is the schematic diagram of undercarriage suspicious region in the embodiment of the present invention;
Fig. 7 is the method flow diagram of the method for the present invention;
Fig. 8 is the method flow diagram that the present invention realizes step S2.
Specific embodiment
The principle and features of the present invention will be described below with reference to the accompanying drawings, and the given examples are served only to explain the present invention, and
It is non-to be used to limit the scope of the invention.
The invention discloses a kind of undercarriage detection system and method, can be used in determining to fly when the aircraft landing of airport
Whether machine undercarriage puts down.
As shown in Figure 1, a kind of detection system of undercarriage, including image pre-processing module, Aircraft Targets detect mould
Block, characteristic extracting module, undercarriage suspicious region detection module and undercarriage detection module,
Described image preprocessing module carries out color for acquiring the initial pictures of undercarriage, and to initial pictures
Processing and noise reduction process, the image after being optimized;
The Aircraft Targets detection module, for successively carrying out background inhibition processing to the image after optimization, at binaryzation
Reason and connected region detection, and the connected region by detecting extracts the Aircraft Targets regional scope in image;
The characteristic extracting module, for extracting the lower boundary in the Aircraft Targets regional scope, then extract it is described under
The chain code feature on boundary;Extract direction character of the chain code feature of lower boundary as Aircraft Targets lower boundary;
The undercarriage suspicious region detection module, for the starting point and chain code direction according to setting under Aircraft Targets
The chain code feature on boundary is encoded, and selectes the coding on specific direction according to chain code direction to detect whether that there are undercarriages
Suspicious region;
The undercarriage detection module, the judgement for extracting undercarriage suspicious region according to histograms of oriented gradients are special
Sign, and according to the authenticity for determining that feature determines the undercarriage suspicious region.
Preferably, described image preprocessing module includes:
Gray scale processing unit, for initial pictures to be converted to gray level image;
Filter processing unit, for gray level image to be carried out median filter process, the image after being optimized.
Initial pictures are carried out gray proces and filtering processing can be optimized with reducing the interference of noise in image
Image, be convenient for post-processing.
The Aircraft Targets detection module includes:
Binary conversion treatment unit carries out background inhibition processing for the image after optimizing, and using adaptive threshold
Background is inhibited treated image to carry out binary conversion treatment by mode;
Connected area disposal$ unit, the image for that will pass through binary conversion treatment carries out connected domain detection, and extracts each company
The area of the area and mass center information in logical domain, the connected domain of the extraction is greater than given value and the smallest connected domain of mass center row coordinate
As Aircraft Targets regional scope.The recall rate of Aircraft Targets is high, and runing time is short.
Preferably, the undercarriage suspicious region detection module includes:
Setup unit, for setting the starting point of chain code feature, and setting point on the right of the starting point is coding starting point,
And using starting point as initial point, digital coding is successively counterclockwise carried out around starting point, and using starting point as terminal, successively
It is encoded to 0 to 7 direction (totally 8 directions), to complete the setting in chain code direction;
Coding unit, for being carried out according to the starting point and chain code direction of setting to the chain code feature of Aircraft Targets lower boundary
Coding;
Suspicious region detection unit is specific direction for selecting 2 directions and 6 directions according to chain code direction, and according to 2 sides
Detect whether that there are undercarriage suspicious regions to maximum column, minimum column, maximum row, the minimum row on 6 direction positions.
Specifically, the chain code on boundary relies on starting point, when extracting the chain code of Aircraft Targets lower boundary, to arrange in lower boundary
The smallest pixel of coordinate counterclockwise carries out chain code coding, as Figure 2-3, root to lower boundary as coding starting point
According to coding direction, the figure of Fig. 3 can obtain 022760117071 number.To find the region where undercarriage, only need
The position in direction 2 and direction 6 in Aircraft Targets lower boundary chain code feature is found, undercarriage direction is then at a distance of the two nearest
Direction surrounds.The position in direction 1,2 and the position in direction 6,7 have been found in design, obtain the position letter of these specific directions
After breath, both direction is matched according to the rule of arest neighbors each other, the region that the both direction position being matched to surrounds is i.e.
For suspicious undercarriage region, these positions are extracted from the gray level image after optimization, as Figure 4-Figure 6.In order to improve suspicious
The verification and measurement ratio in frame region is fallen, counts the position in direction 1,2 and the position in direction 6,7 in Aircraft Targets lower boundary direction character respectively
It sets.For the maximum column l of each 6,7 direction positions6-7, find and be less than l at a distance of nearest and column with it6-7Direction 1,2 position
It sets, according to the maximum column of the two positions, minimum column, as soon as maximum row, minimum row extract piece region of image, this region
It is a suspicious undercarriage region.There can be multiple suspicious undercarriage regions in one Aircraft Targets, it can also be without suspicious
Frame region is fallen, to realize the detection of suspicious landing gear position in Aircraft Targets.
The undercarriage detection module includes:
Determine feature extraction unit, the histograms of oriented gradients feature for extracting the undercarriage suspicious region, which is used as, to be sentenced
Determine feature;
Undercarriage detection unit, for determining in the judgement feature input limits learning machine classifier by extraction, and
The authenticity of institute's undercarriage suspicious region is determined according to the result that extreme learning machine classifier exports.
Realize undercarriage suspicious region whether be true undercarriage region judgement, undercarriage regional determination accuracy
Height, runing time are short.
As shown in fig. 7, a kind of detection method of undercarriage, includes the following steps:
Step S1: the initial pictures of undercarriage are acquired, and colors countenance and noise reduction process are carried out to initial pictures, are obtained
Image after to optimization;
Step S2: binary conversion treatment and connected region detection are successively carried out to the image after optimization, and by detecting
Connected region extracts the Aircraft Targets regional scope in image;
Step S3: extracting the lower boundary in the Aircraft Targets regional scope, then extracts the chain code feature of the lower boundary;
Step S4: encoding the chain code feature of Aircraft Targets lower boundary according to the starting point of setting and chain code direction,
And the coding on specific direction is selected to detect whether that there are undercarriage suspicious regions according to chain code direction;
Step S5: encoding the chain code feature of Aircraft Targets lower boundary according to the starting point of setting and chain code direction,
And the coding on specific direction is selected to detect whether that there are undercarriage suspicious regions according to chain code direction;
Step S6: the judgement feature of undercarriage suspicious region is extracted according to histograms of oriented gradients, and according to judgement feature
Determine the authenticity of the undercarriage suspicious region.
Preferably, the specific steps of step S1 are realized are as follows:
Step S101: initial pictures are converted into gray level image;
Step S102: gray level image is subjected to median filter process, the image after being optimized.
As shown in figure 8, realizing the specific steps of step S2 are as follows:
Step S201: the image after optimization is subjected to background inhibition processing, and by background by the way of adaptive threshold
Treated image is inhibited to carry out binary conversion treatment;
Step S202: the image Jing Guo binary conversion treatment is subjected to connected domain detection, and extracts the area of each connected domain
With mass center information, the area of the connected domain of the extraction is greater than given value and the smallest connected domain of mass center row coordinate as aircraft mesh
Mark regional scope.
Realize the specific steps of step S4 are as follows:
Step S401: the starting point of setting chain code feature, and setting point on the right of the starting point is coding starting point, and with
Starting point is initial point, counterclockwise successively carries out digital coding around starting point, and using starting point as terminal, successively obtain 0
To 7 direction encodings, to complete the setting in chain code direction;
Step S402: the chain code feature of Aircraft Targets lower boundary is compiled according to the starting point of setting and chain code direction
Code;
Step S403: selecting 2 directions and 6 directions according to chain code direction is specific direction, and according to 2 directions and 6 direction positions
The maximum column set, minimum column, maximum row, minimum row detect whether that there are undercarriage suspicious regions.
Preferably, the specific steps of step S5 are realized are as follows:
Step S501: the histograms of oriented gradients feature of the undercarriage suspicious region is extracted as judgement feature;
Step S502: will determine in the judgement feature input limits learning machine classifier of extraction, and according to the limit
The result of habit machine classifier output determines the authenticity of institute's undercarriage suspicious region.
The present invention can be quickly detected whether undercarriage puts down, and be easy to Project Realization;Undercarriage detection is correct
Rate is high, has filled up the blank currently for this aspect research;Position and the quantity information of undercarriage can be provided, more surely
It is fixed.
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all in spirit of the invention and
Within principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.
Claims (8)
1. a kind of detection system of undercarriage, which is characterized in that detect mould including image pre-processing module, Aircraft Targets
Block, characteristic extracting module, undercarriage suspicious region detection module and undercarriage detection module,
Described image preprocessing module carries out colors countenance for acquiring the initial pictures of undercarriage, and to initial pictures
And noise reduction process, the image after being optimized;
The Aircraft Targets detection module, for the image after optimization is successively carried out background inhibition processing, binary conversion treatment and
Connected region detection, and the connected region by detecting extracts the Aircraft Targets regional scope in image;
The characteristic extracting module for extracting the lower boundary in the Aircraft Targets regional scope, then extracts the lower boundary
Chain code feature;
The undercarriage suspicious region detection module, for the starting point and chain code direction according to setting to Aircraft Targets lower boundary
Chain code feature encoded, and the coding on specific direction is selected according to chain code direction and detects whether that it is suspicious that there are undercarriages
Region;The undercarriage suspicious region detection module includes:
Setup unit, for setting the starting point of chain code feature, and setting point on the right of the starting point is coding starting point, and with
Starting point is initial point, counterclockwise successively carries out digital coding around starting point, and using starting point as terminal, successively obtain 0
To 7 direction encodings, to complete the setting in chain code direction;
Coding unit, for being compiled according to the starting point and chain code direction of setting to the chain code feature of Aircraft Targets lower boundary
Code;
Suspicious region detection unit is specific direction for selecting 2 directions and 6 directions according to chain code direction, and according to 2 directions and
Maximum column on 6 directions position, minimum column, maximum row, minimum row detect whether that there are undercarriage suspicious regions;
The undercarriage detection module, for extracting the judgement feature of undercarriage suspicious region according to histograms of oriented gradients, and
According to the authenticity for determining that feature determines the undercarriage suspicious region.
2. a kind of detection system of undercarriage according to claim 1, which is characterized in that described image pre-processes mould
Block includes:
Gray scale processing unit, for initial pictures to be converted to gray level image;
Filter processing unit, for gray level image to be carried out median filter process, the image after being optimized.
3. a kind of detection system of undercarriage according to claim 1, which is characterized in that the Aircraft Targets detection
Module includes:
Binary conversion treatment unit carries out background inhibition processing for the image after optimizing, and by the way of adaptive threshold
Treated image is inhibited to carry out binary conversion treatment background;
Connected area disposal$ unit, the image for that will pass through binary conversion treatment carries out connected domain detection, and extracts each connected domain
Area and mass center information, the area of the connected domain of the extraction be greater than given value and the smallest connected domain conduct of mass center row coordinate
Aircraft Targets regional scope.
4. a kind of detection system of undercarriage according to claim 1, which is characterized in that the undercarriage detects mould
Block includes:
Determine feature extraction unit, the histograms of oriented gradients feature for extracting the undercarriage suspicious region is special as determining
Sign;
Undercarriage detection unit, for determining in the judgement feature input limits learning machine classifier by extraction, and according to
The result of extreme learning machine classifier output determines the authenticity of institute's undercarriage suspicious region.
5. a kind of detection method of undercarriage, which comprises the steps of:
Step S1: the initial pictures of undercarriage are acquired, and colors countenance and noise reduction process are carried out to initial pictures, are obtained excellent
Image after change;
Step S2: background inhibition processing, binary conversion treatment and connected region detection are successively carried out to the image after optimization, and is passed through
The connected region detected extracts the Aircraft Targets regional scope in image;
Step S3: extracting the lower boundary in the Aircraft Targets regional scope, then extracts the chain code feature of the lower boundary;
Step S4: the chain code feature of Aircraft Targets lower boundary is encoded according to the starting point of setting and chain code direction, and root
The coding on specific direction is selected according to chain code direction to detect whether that there are undercarriage suspicious regions;Realize the specific step of step S4
Suddenly are as follows:
Step S401: the starting point of setting chain code feature, and setting point on the right of the starting point is coding starting point, and to set out
Point is initial point, counterclockwise successively carries out digital coding around starting point, and using starting point as terminal, successively obtain 0 to 7 side
To coding, to complete the setting in chain code direction;
Step S402: the chain code feature of Aircraft Targets lower boundary is encoded according to the starting point of setting and chain code direction;
Step S403: selecting 2 directions and 6 directions according to chain code direction is specific direction, and according on 2 directions and 6 direction positions
Maximum column, minimum column, maximum row, minimum row detect whether that there are undercarriage suspicious regions;
Step S5: the judgement feature of undercarriage suspicious region is extracted according to histograms of oriented gradients, and is determined according to judgement feature
The authenticity of the undercarriage suspicious region.
6. a kind of detection method of undercarriage according to claim 5, which is characterized in that realize that step S1's is specific
Step are as follows:
Step S101: initial pictures are converted into gray level image;
Step S102: gray level image is subjected to median filter process, the image after being optimized.
7. a kind of detection method of undercarriage according to claim 5, which is characterized in that realize that step S2's is specific
Step are as follows:
Step S201: the image after optimization is subjected to background inhibition processing, and is inhibited background by the way of adaptive threshold
Treated, and image carries out binary conversion treatment;
Step S202: the image Jing Guo binary conversion treatment is subjected to connected domain detection, and extracts the area and matter of each connected domain
Heart information, the area of the connected domain of the extraction are greater than given value and the smallest connected domain of mass center row coordinate as Aircraft Targets area
Domain range.
8. a kind of detection method of undercarriage according to claim 5, which is characterized in that realize that step S5's is specific
Step are as follows:
Step S501: the histograms of oriented gradients feature of the undercarriage suspicious region is extracted as judgement feature;
Step S502: will determine in the judgement feature input limits learning machine classifier of extraction, and according to extreme learning machine
The result of classifier output determines the authenticity of institute's undercarriage suspicious region.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610554236.9A CN106203353B (en) | 2016-07-14 | 2016-07-14 | A kind of detection system and method for undercarriage |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610554236.9A CN106203353B (en) | 2016-07-14 | 2016-07-14 | A kind of detection system and method for undercarriage |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106203353A CN106203353A (en) | 2016-12-07 |
CN106203353B true CN106203353B (en) | 2019-11-15 |
Family
ID=57475282
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610554236.9A Active CN106203353B (en) | 2016-07-14 | 2016-07-14 | A kind of detection system and method for undercarriage |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106203353B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106951823B (en) * | 2017-02-06 | 2019-11-15 | 南京莱斯电子设备有限公司 | A kind of infrared image undercarriage automatic identifying method |
CN107633505A (en) * | 2017-08-24 | 2018-01-26 | 南京理工大学 | A kind of undercarriage detection method based on target gray distribution character |
CN114577254B (en) * | 2022-05-07 | 2022-09-09 | 成都凯天电子股份有限公司 | High-reliability detection method and system based on undercarriage inductive proximity sensor |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101551823A (en) * | 2009-04-20 | 2009-10-07 | 浙江师范大学 | Comprehensive multi-feature image retrieval method |
CN103413136A (en) * | 2013-08-01 | 2013-11-27 | 西北工业大学 | Method for detecting high-voltage power lines in infrared image |
CN103839268A (en) * | 2014-03-18 | 2014-06-04 | 北京交通大学 | Method for detecting fissure on surface of subway tunnel |
-
2016
- 2016-07-14 CN CN201610554236.9A patent/CN106203353B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101551823A (en) * | 2009-04-20 | 2009-10-07 | 浙江师范大学 | Comprehensive multi-feature image retrieval method |
CN103413136A (en) * | 2013-08-01 | 2013-11-27 | 西北工业大学 | Method for detecting high-voltage power lines in infrared image |
CN103839268A (en) * | 2014-03-18 | 2014-06-04 | 北京交通大学 | Method for detecting fissure on surface of subway tunnel |
Non-Patent Citations (1)
Title |
---|
基于数学形态学的管道缺陷特征提取方法研究;殷宇殿;《中国优秀硕士学位论文全文数据库信息科技辑(月刊)》;20150515;第2015年卷(第5期);第19-70页 * |
Also Published As
Publication number | Publication date |
---|---|
CN106203353A (en) | 2016-12-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109522793B (en) | Method for detecting and identifying abnormal behaviors of multiple persons based on machine vision | |
CN104483326B (en) | High-voltage line defects of insulator detection method and system based on depth belief network | |
CN102855622B (en) | A kind of infrared remote sensing image sea ship detection method based on significance analysis | |
CN103442209B (en) | Video monitoring method of electric transmission line | |
CN106203353B (en) | A kind of detection system and method for undercarriage | |
CN110070530A (en) | A kind of powerline ice-covering detection method based on deep neural network | |
CN103235938B (en) | The method and system of car plate detection and indentification | |
CN106950197B (en) | The Remotely sensed acquisition methods, devices and systems of sewage draining exit polluted-water | |
CN105184271A (en) | Automatic vehicle detection method based on deep learning | |
CN104200668B (en) | A kind of motorcycle based on graphical analysis is not helmeted violation event detection method | |
CN108537154A (en) | Transmission line of electricity Bird's Nest recognition methods based on HOG features and machine learning | |
CN104463097B (en) | High-voltage line image detecting method based on local auto-adaptive Threshold Segmentation Algorithm | |
CN107392885A (en) | A kind of method for detecting infrared puniness target of view-based access control model contrast mechanism | |
CN110120155A (en) | A kind of chemical industry plant area vehicle overload overload intelligent monitoring and alarming system | |
TW201120814A (en) | Method for determining if an input image is a foggy image, method for determining a foggy level of an input image and cleaning method for foggy images | |
CN109670515A (en) | A kind of detection method and system changed for building in unmanned plane image | |
CN110008932A (en) | A kind of vehicle violation crimping detection method based on computer vision | |
CN103632158A (en) | Forest fire prevention monitor method and forest fire prevention monitor system | |
CN108280952A (en) | A kind of passenger's trailing monitoring method based on foreground object segmentation | |
CN104463121A (en) | Crowd density information obtaining method | |
CN108960133A (en) | Method, electronic equipment, system and the storage medium of ridership monitoring | |
CN106447699A (en) | High-speed rail overhead contact line equipment object detection and tracking method based on Kalman filtering | |
CN113807220A (en) | Traffic event detection method and device, electronic equipment and readable storage medium | |
CN107368786A (en) | A kind of passenger based on machine vision crosses handrail detection algorithm | |
CN104573650B (en) | A kind of electric wire detection sorting technique based on filter response |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |