CN104636710B - Multi-source forest zone Caiyu object detection system - Google Patents
Multi-source forest zone Caiyu object detection system Download PDFInfo
- Publication number
- CN104636710B CN104636710B CN201310561647.7A CN201310561647A CN104636710B CN 104636710 B CN104636710 B CN 104636710B CN 201310561647 A CN201310561647 A CN 201310561647A CN 104636710 B CN104636710 B CN 104636710B
- Authority
- CN
- China
- Prior art keywords
- caiyu
- target
- data
- forest zone
- 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.)
- Expired - Fee Related
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
- G06V20/188—Vegetation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Image Processing (AREA)
Abstract
The present invention devises a kind of forest zone Caiyu object detection system based on Multi-sensor Fusion.This system is built by scanning laser range finder and thermal infrared imager, is acquired to the thermal-induced imagery, visible images and laser data of forest zone Caiyu target.Image data is handled by the methods of image segmentation, image co-registration, extracts the features such as color, temperature, the region of Caiyu target；Effective laser data point is obtained by filtering, the methods of denoising simultaneously, and laser data carried out with image data to be registrated association, to obtain the width and location information of target.Supporting vector machine model is finally established, by merging the complementary characteristic of different sensors extraction, can effectively forest zone Caiyu target is detected and be identified.This programme makes full use of multi-source information, and the advantages of each sensor is combined, and reduces by the uncertainty caused by single-sensor, can effectively detect forest zone Caiyu target, improves forest operation efficiency, enhances work capacity.
Description
Technical field
The present invention relates to it is a kind of quickly, accurate, system that forest zone Caiyu target is detected.It is that a kind of apply is being combined
The intelligent assistance system that Caiyu operation is carried out on forest Caiyu machine, can help operator to confirm the information of Caiyu target, improve
The working efficiency of forest Caiyu machine simultaneously reduces operation danger.
Background technology
To meet the needs of China's forestry industry fast development, forestry Caiyu Work machine equipment gradually replaces artificial make
Industry, the domestic multi-functional Caiyu machine developed, can complete whole operation process of artificial forest Caiyu, operating efficiency highest can at present
80 times or more of intelligent's work Caiyu production.But due to forest land environment complexity, the influence of barrier causes continuous Caiyu operation
Interruption, and the danger of operator is increased, reduce operating efficiency.Accurately identify Caiyu target, detection operation environmental information,
And carry out safe early warning is timely feedbacked, operating efficiency can be improved, enhances work capacity, while also reducing the operation of operator
It is dangerous.
Invention content
Due to forest land environment complexity, the influence of barrier can cause the interruption of the continuous Caiyu operation of Caiyu machine, increase operation
Danger, and reduce operating efficiency.However the information that single-sensor obtains is not enough, and is not enough to the complicated forest zone of expression
Environment, therefore the present invention uses multi-sensor fusion technology, the complementary characteristic provided by merging each sensor, can quickly,
Effectively, accurately forest zone Caiyu target is detected.
Specifically there is following several respects content：
1. building based on scanning laser range finder（1 in Fig. 2）, thermal infrared imager（2 in Fig. 2）Forest zone Caiyu target information
Acquisition system.The two is fixed on same plane, is placed in cradle head of two degrees of freedom（3 in Fig. 2）On, by tripod（4 in Fig. 2）Branch
Support, and it is respectively connected to host computer（5 in Fig. 2）, the data acquisition of sensor is uniformly controlled by host computer.
2. optimization processings such as pair laser data acquired is filtered, denoisings；Simultaneously to infrared image and visible light figure
As carrying out the processing such as image segmentation, feature extraction.
3. by image fusion technology, the image information of detection target is enriched, the hiding target in single image is obtained, carries
The accuracy of hi-vision data.By laser and image registration, target can be clearly detected, while determining the location information of target
4. by learning to great amount of samples data, support vector machine classifier model is established, the target signature based on extraction
Caiyu target is identified, can quickly and effectively detect forest zone standing tree.
The beneficial effects of the present invention are：
1. building the information acquisition system based on thermal infrared imager and two dimensional laser scanning instrument, the two is fixed on same plane
On, it is demarcated convenient for the fusion of data, obtains really and accurately data.
2. making full use of multi-source information, the advantages of each sensor image, is combined, is obtained to the maximum extent to various
The information of feature describes, and reduces by the uncertainty caused by single-sensor.
3. the foundation of support vector machine classifier model is based on the study to great amount of samples data, reliability is high, can be with
Fast and accurately detect forest zone standing tree.
Description of the drawings
Fig. 1 is overall flow figure of the present invention；
Fig. 2 is data acquisition platform structure of the present invention；
Fig. 3 is inventive algorithm flow chart.
Specific implementation mode
Specific embodiment of the present invention is as follows：
1. according to Fig. 1, general system proposal is established, overall plan document is established, including：
A. detecting system equipment is chosen, major function, technical indicator, principle block diagram and explanatory note；
B. data fusion method principle and explanation；
2. according to fig. 2, building data acquisition platform, gathered data establishes database.
3. according to fig. 3, carrying out corresponding algorithm process to collected data, characteristic is extracted, to support vector machines
It is trained, with trained model come recognition detection forest zone Caiyu target.
Claims (1)
1. the forest zone Caiyu object detection system based on multi-sensor information fusion technology, it is characterised in that：It is by infrared thermal imagery
Instrument and laser range finder are built, and to carry out forest zone data acquisition, collected data include Caiyu target --- forest zone standing tree
Thermal-induced imagery, visible images and two-dimensional laser data；
Include using the method that the forest zone Caiyu object detection system carries out Caiyu target detection：
Lectotype selection simultaneously builds information acquisition system；Build the forest zone Caiyu mesh based on scanning laser range finder, thermal infrared imager
Mark information acquisition system；The two is fixed on same plane, is placed on cradle head of two degrees of freedom, is supported by tripod, and is connected respectively
It is connected to host computer, the data acquisition of sensor is uniformly controlled by host computer；
Laser range finder collects two-dimensional laser data；By target laser data are obtained by filtration；Mesh is extracted from laser data
Target width and location information, and the position of target is positioned by the calibration of laser data and image；
Thermal infrared imager acquires thermal-induced imagery and visible images simultaneously；It is adopted by the infrared acquisition of merging with visible images
Educate the image information of target；For Caiyu target Extracting temperature, color and area from infrared, visible light and blending image respectively
Characteristic of field；
The laser data of Caiyu target is associated with image data and is matched；
By being trained with collected great amount of samples data, support vector machine classifier model is established；
Classify to unknown data, exports the Caiyu objective result detected, specially：Various features based on extraction, fortune
Classified to collected information with support vector machine classifier, Caiyu target is identified, with minimum in visible images
The mode of boundary rectangle is shown.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310561647.7A CN104636710B (en) | 2013-11-13 | 2013-11-13 | Multi-source forest zone Caiyu object detection system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310561647.7A CN104636710B (en) | 2013-11-13 | 2013-11-13 | Multi-source forest zone Caiyu object detection system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104636710A CN104636710A (en) | 2015-05-20 |
| CN104636710B true CN104636710B (en) | 2018-08-10 |
Family
ID=53215442
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310561647.7A Expired - Fee Related CN104636710B (en) | 2013-11-13 | 2013-11-13 | Multi-source forest zone Caiyu object detection system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104636710B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108108720A (en) * | 2018-01-08 | 2018-06-01 | 天津师范大学 | A kind of ground cloud image classification method based on depth multi-modal fusion |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO1999019824A1 (en) * | 1997-10-10 | 1999-04-22 | Case Corporation | Method for monitoring nitrogen status using a multi-sprectral imaging system |
| CN201322681Y (en) * | 2008-11-17 | 2009-10-07 | 浙江红相科技有限公司 | SF6 gas leakage laser imager based on infrared imaging technology |
| CN102520414A (en) * | 2011-11-18 | 2012-06-27 | 西安交通大学 | Laser active and infrared reactive compound detecting device |
-
2013
- 2013-11-13 CN CN201310561647.7A patent/CN104636710B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO1999019824A1 (en) * | 1997-10-10 | 1999-04-22 | Case Corporation | Method for monitoring nitrogen status using a multi-sprectral imaging system |
| CN201322681Y (en) * | 2008-11-17 | 2009-10-07 | 浙江红相科技有限公司 | SF6 gas leakage laser imager based on infrared imaging technology |
| CN102520414A (en) * | 2011-11-18 | 2012-06-27 | 西安交通大学 | Laser active and infrared reactive compound detecting device |
Non-Patent Citations (2)
| Title |
|---|
| 一种单线激光雷达和可见光摄像机的标定方法;刘大学等;《华中科技大学学报》;20081031;第36卷;全文 * |
| 基于红外CCD与激光测距仪融合的行人检测技术研究;余燕;《中国优秀硕士学位论文全文数据库 信息科技辑》;20081115(第11期);第39-79页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104636710A (en) | 2015-05-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liu et al. | Improved kiwifruit detection using pre-trained VGG16 with RGB and NIR information fusion | |
| EP3171292B1 (en) | Driving lane data processing method, device, storage medium and apparatus | |
| CN109472831A (en) | Obstacle recognition range-measurement system and method towards road roller work progress | |
| CN106969706A (en) | Workpiece sensing and three-dimension measuring system and detection method based on binocular stereo vision | |
| CN110910350B (en) | Nut loosening detection method for wind power tower cylinder | |
| CN104378539B (en) | The video structural extraction of semantics video camera and its method of scene adaptive | |
| CN103150730A (en) | Round small target accurate detection method based on image | |
| CN105279756A (en) | Notch circular arc part dimension visual detection method based on self-adapting region division | |
| CN101441769A (en) | Real time vision positioning method of monocular camera | |
| CN109447061A (en) | Reactor oil level indicator recognition methods based on crusing robot | |
| CN106824806A (en) | The detecting system and detection method of the low module plastic gear based on machine vision | |
| JP2020042794A (en) | Self-driving vehicle visual sensing method, self-driving vehicle visual sensing device, control instrument and computer readable storage medium | |
| Yuan et al. | Combining maps and street level images for building height and facade estimation | |
| CN111832760B (en) | Automatic inspection method for well lid based on visual algorithm | |
| CN107610086A (en) | Industrial parallel robot rapid visual detection algorithm based on bionic compound eye structure | |
| KR20160088986A (en) | Lane detection method using disparity based on vanishing point | |
| TWI543117B (en) | Method for recognizing and locating object | |
| CN103903279A (en) | Parallel tracking system and method based on bionic binocular vision onboard platform | |
| CN104636710B (en) | Multi-source forest zone Caiyu object detection system | |
| Jiang et al. | Land segmentation and contour extraction of remote sensing image based on Mask R-CNN | |
| CN114881241A (en) | Deep learning-based lane line detection method and device and automatic driving method | |
| CN113936210A (en) | Anti-collision method for tower crane | |
| JP2020006788A (en) | Construction limit determination device | |
| JP4459162B2 (en) | Velocity measuring device, method and program | |
| Yu et al. | Surface defect detection of hight-speed railway hub based on improved YOLOv3 algorithm |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180810 Termination date: 20201113 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |