CN107655403A - A kind of high-rate laser spot detection system - Google Patents
A kind of high-rate laser spot detection system Download PDFInfo
- Publication number
- CN107655403A CN107655403A CN201610583298.2A CN201610583298A CN107655403A CN 107655403 A CN107655403 A CN 107655403A CN 201610583298 A CN201610583298 A CN 201610583298A CN 107655403 A CN107655403 A CN 107655403A
- Authority
- CN
- China
- Prior art keywords
- image
- fpga
- detection system
- laser
- spot detection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/002—Measuring arrangements characterised by the use of optical techniques for measuring two or more coordinates
- G01B11/005—Measuring arrangements characterised by the use of optical techniques for measuring two or more coordinates coordinate measuring machines
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
A kind of high-rate laser spot detection system.The present invention is in order to realize capture, tracking and aiming of the laser communication system to beacon beam, high-rate laser spot detection system based on FPGA has been invented using embedded image treatment technology, has been that the laser facula target in 300pixel × 300pixel, frame frequency 1000Hz sequence images has carried out extraction operation to resolution ratio.Image preprocessing is carried out using 1 dimension filter operator construction azimuth filtering device, while is easy to hardware to realize, through Threshold segmentation, morphology opening operation and connected domain analysis after filtering, is finally extracted the position of form center information of laser facula.As a result show, invention system can be handled the high frame frequency image comprising laser facula target in real time.Target centre of form extraction frequency can meet requirement of the laser communication system in terms of dynamic link foundation.
Description
Art
The present invention relates to a kind of detecting system, more particularly to a kind of high-rate laser spot detection system.
Background technology
Because laser communication system has bandwidth, high speed, anti-intercepting and capturing ability strong, strong antijamming capability and small-sized etc. prominent
Advantage, make its with being especially suitable for deep space, interspace, star, the link communication such as absolutely empty, vacant lot, therefore obtained rapid hair in recent years
Exhibition.The foundation of laser communication link with keep be unableing to do without capture to beacon beam, tracking and aiming (acquisition,
Tracking and pointing, ATP).Hot spot target position information extraction based on Digital Image Processing is one in ATP
Item key technology, its extraction rate directly affect the break-make of link with precision, and carrying out correlative study has great significance.
Research to the laser communication spot detection technology based on Digital Image Processing can be divided mainly into theory of algorithm and reality
Border is using two aspects.In terms of algorithm, ZHANG et al. proposes the automatic threshold Processing Algorithm based on noise characteristic;LIU etc.
People proposes the centralized positioning algorithm based on center of circle fitting;SUN et al. proposes the hot spot Denoising Algorithm based on small echo; SHAO
Propose the single-point gray scale value-based algorithm of amendment.The above-mentioned research for algorithm is to carry out testing for treatment effect by simulated environment
Card, its hardware is portable and algorithm real-time still needs to be considered.In application aspect, LU et al. propose based on particle filter
Method is 640piexl × 480piexl applied to resolution ratio, and frame frequency is in 24Hz sequence image;CHENG is filtered using intermediate value
Ripple realizes resolution ratio 1024piexl × 1040piexl, the hot spot extraction of frame frequency 25Hz sequence images;The base that XU et al. is proposed
Resolution ratio 320piexl × 256piexl, frame frequency 83Hz image can be located in real time in upper three neighborhoods continuity point counting method
Reason.As what laser communication was studied deepens continuously, the frequency requirement also more and more higher to extracting hot spot target, 100Hz and with
Under extraction frequency can not meet the foundation and holding of dynamic link.Some external scholars pass through with field-programmable gate array
Row (fieldprogrammablegate array, FPGA) are that the system of core devices is had been realized in high frame frequency image
Real-time processing.The content of the invention
The invention aims to realize capture, tracking and aiming of the laser communication system to beacon beam, a kind of height is devised
Fast detecting a laser beam system.
The technical solution adopted for the present invention to solve the technical problems is:
High-rate laser spot detection system is made up of FPGA and peripheral circuit.The main working process of system is image receiver module
The picture signal that collects of camera is received, and digital image information is transmitted into cache module, image processing module is from caching
Extract successively data and be filtered with the operation such as spot center extraction, image processing module is by the coordinates of laser spot center of extraction
Transmit into signal integration module, image is carried out data with its spot center coordinate information and merged by signal integration module, is passed through
PCI-e communication modules transmit original image information and spot center coordinate into main control computer.
Described FPGA uses the XC5VLX110T types fpga chip of match company of SEL (Xilinx) as its main operational device
Part.This FPGA internal resource enriches, and is shown according to its technical data, has 17280 resource units, is wrapped in each resource unit
4 display look-up tables and 4 triggers are included, caching can be used as by possessing 5328 × 1000 pieces of random access memory;Operational capability side
Face, the FPGA possess 64 DSP48E data signal computing kernels, can carry out the complex calculations such as convolution, division.Utilize
Functional kernel in XC5VLX110T types FPGA and abundant storage resource can realize real-time processing to high frame frequency image with
The functions such as Control on Communication.
The filtering method that described image processing module is combined using 1 dimension operator filtering and multi-faceted wave filter.
The beneficial effects of the invention are as follows:
High-rate laser spot detection system is using field programmable gate array as core devices, and system is by azimuth filtering device to image
Pre-processed, the target centre of form is extracted using connected domain analysis, realizes the inspection to the laser spot center in high frame frequency image
Survey.The system is applied in laser communication system, can provide necessary Real-Time Parameter information with keeping for the foundation of dynamic link.
Embodiment
High-rate laser spot detection system is made up of FPGA and peripheral circuit.The main working process of system is image-receptive
Module receives the picture signal that collects of camera, and digital image information is transmitted into cache module, image processing module from
Caching extracts data and is filtered and sits the spot center of extraction with operation, image processing modules such as spot center extractions successively
Scale value is transmitted into signal integration module, and image is carried out data with its spot center coordinate information and merged by signal integration module,
Original image information and spot center coordinate are transmitted into main control computer by PCI-e communication modules.
The design cycle of image processing module is as follows:To ensure the real-time of system, set when being pre-processed to image
Meter uses airspace filter, and compared with frequency domain filtering, FPGA parallel processing mechanism is more suitable for realizing spatial domain filter algorithms.Digitized map
As information is 1 group of 2 dimension word signal, therefore start with expansion research from 2 dimension wave filters first in design, Laplce-Gauss
(Laplacian of Gaussian, LOG) operator is typical 2 dimension filter operator., algorithm basic thought is as follows:Image
Reach smooth effect with Gauss operator convolution in the form of 2 position digital signals, marginal point is converted into zero crossing by Laplace operator
Point, pass through the detection for intersecting zero point to realize to image border.Because Gaussian function is on axial symmetry, therefore LOG operators belong to
Isotropism operator, when objective trait is irregular, the edge of different directions can not be carried out using classical LOG operator filterings
Effective detection, and be unfavorable for FPGA and carry out algorithm realization.To solve the above problems, proposing 1 dimension LOG operator concepts, and treat
It is 2 dimension word signals to handle image, is only capable of handling image from line direction or column direction using 1 dimension operator filtering, because
This need to design multi-faceted wave filter and 2 dimension word picture signals are filtered.The Gaussian Profile standard deviation in each orientation is represented with σ l,
The directions of rays that multi-faceted filter operator can be formed to original image each point with its eight neighborhood point is filtered to image.Filtering
Effect and sequence length Mj (j=1,2 ..., 8) are relevant with σ l.σ l are bigger, better to the filter effect of noise, but can lose
More marginal informations;σ l are smaller, better to the Detection results of target but weaker to the smoothing capability of noise.Sequence length
Longer, filter effect is better, but the 1 dimension filter operator in each orientation is the e negative exponential functions on variable square, when variable
When absolute value is very big, operator function value is close to 0, therefore sequence length is long nonsensical, can influence arithmetic speed on the contrary.If
The σ l values put on different directions can be filtered with sequence length for the image comprising different target type.Due to actual work
Cheng Zhong, hot spot shape formed thereby are not strict circles, therefore by setting different σ l values to be adapted to respective shapes hot spot
Processing.In terms of Threshold Segmentation Algorithm, to ensure system real time, global threshold is set using control computer dynamic in design
Value;To solve the problems, such as that hot spot is broken, the morphology opening operation of binary image has been carried out after Threshold segmentation;Finally by
FPGA carries out the operation of connected domain analysis and Objective extraction, determines whether include hot spot target in image, if then exporting mesh
Mark the coordinate value of the centre of form.
Claims (3)
1. a kind of high-rate laser spot detection system is made up of FPGA and peripheral circuit;The main working process of system connects for image
Receive module and receive the picture signal that camera collects, and digital image information is transmitted into cache module, image processing module
Extracted successively from caching data and be filtered with the operation such as spot center extraction, image processing module is by the spot center of extraction
Coordinate value is transmitted into signal integration module, and image and its spot center coordinate information are carried out data conjunction by signal integration module
And original image information and spot center coordinate are transmitted into main control computer by PCI-e communication modules.
2. high-rate laser spot detection system according to claim 1, it is characterized in that described FPGA is public using match SEL
The XC5VLX110T types fpga chip for taking charge of (Xilinx) is its main operational device;This FPGA internal resource enriches, according to its skill
Art data shows there are 17280 resource units, include 4 display look-up tables and 4 triggers in each resource unit, possess
5328 × 1000 pieces of random access memory can be used as caching;In terms of operational capability, the FPGA possesses 64 DSP48E data signals fortune
Kernel is calculated, the complex calculations such as convolution, division can be carried out;Deposited using the functional kernel in XC5VLX110T types FPGA and abundant
Storage resource can realize the functions such as real-time processing and Control on Communication to high frame frequency image.
3. high-rate laser spot detection system according to claim 1, it is characterized in that described image processing module uses 1
The filtering method that dimension operator filtering and multi-faceted wave filter are combined.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610583298.2A CN107655403A (en) | 2016-07-24 | 2016-07-24 | A kind of high-rate laser spot detection system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610583298.2A CN107655403A (en) | 2016-07-24 | 2016-07-24 | A kind of high-rate laser spot detection system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107655403A true CN107655403A (en) | 2018-02-02 |
Family
ID=61126357
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610583298.2A Pending CN107655403A (en) | 2016-07-24 | 2016-07-24 | A kind of high-rate laser spot detection system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107655403A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111767875A (en) * | 2020-07-06 | 2020-10-13 | 中兴飞流信息科技有限公司 | Tunnel smoke detection method based on instance segmentation |
CN112629831A (en) * | 2020-12-09 | 2021-04-09 | 北京航天飞腾装备技术有限责任公司 | Laser weapon facula monitoring and tracking system |
CN112781591A (en) * | 2019-11-06 | 2021-05-11 | 深圳市优必选科技股份有限公司 | Robot positioning method and device, computer readable storage medium and robot |
CN116433577A (en) * | 2023-01-18 | 2023-07-14 | 中国航天三江集团有限公司 | Light spot detection and light beam quality estimation method and device based on one-dimensional multi-scale convolution feature extraction |
-
2016
- 2016-07-24 CN CN201610583298.2A patent/CN107655403A/en active Pending
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112781591A (en) * | 2019-11-06 | 2021-05-11 | 深圳市优必选科技股份有限公司 | Robot positioning method and device, computer readable storage medium and robot |
CN111767875A (en) * | 2020-07-06 | 2020-10-13 | 中兴飞流信息科技有限公司 | Tunnel smoke detection method based on instance segmentation |
CN111767875B (en) * | 2020-07-06 | 2024-05-10 | 中兴飞流信息科技有限公司 | Tunnel smoke detection method based on instance segmentation |
CN112629831A (en) * | 2020-12-09 | 2021-04-09 | 北京航天飞腾装备技术有限责任公司 | Laser weapon facula monitoring and tracking system |
CN116433577A (en) * | 2023-01-18 | 2023-07-14 | 中国航天三江集团有限公司 | Light spot detection and light beam quality estimation method and device based on one-dimensional multi-scale convolution feature extraction |
CN116433577B (en) * | 2023-01-18 | 2024-02-06 | 中国航天三江集团有限公司 | Light spot detection and light beam quality estimation method and device based on one-dimensional multi-scale convolution feature extraction |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107655403A (en) | A kind of high-rate laser spot detection system | |
CN105550685B (en) | The large format remote sensing image area-of-interest exacting method of view-based access control model attention mechanism | |
CN110246115A (en) | A kind of detection method of far-field laser light spot image | |
WO2022000862A1 (en) | Method and apparatus for detecting object in fisheye image, and storage medium | |
CN111666944B (en) | Infrared weak and small target detection method and device | |
CN104657945A (en) | Infrared small target detection method for multi-scale spatio-temporal union filtering under complex background | |
CN105518709A (en) | Method, system and computer program product for identifying human face | |
CN110298860A (en) | A kind of high bar silk ball detection number system based on machine vision | |
Dong et al. | Infrared target detection in backlighting maritime environment based on visual attention model | |
CN110400294A (en) | A kind of Infrared Target Detection system and detection method | |
Zhang et al. | Saliency-based centernet for ship detection in sar images | |
Jiang et al. | Automatic space debris extraction channel based on large field of view photoelectric detection system | |
Jiang et al. | An infrared small target detection algorithm based on peak aggregation and Gaussian discrimination | |
Li et al. | [Retracted] Image Processing and Recognition Algorithm Design in Intelligent Imaging Device System | |
CN102194249B (en) | Water current modeling data capturing device with combination of infrared rays and visible light | |
Dong et al. | Extracting sea-sky-line based on improved local complexity | |
CN103164857A (en) | Image-based high-efficiency detection method for short-distance large-deflection circular target | |
Zhang et al. | Hierarchical and parallel pipelined heterogeneous SoC for embedded vision processing | |
Gu et al. | A light-weight object detection framework with FPA module for optical remote sensing imagery | |
Huang et al. | Infrared small target detection with directional difference of Gaussian filter | |
Li et al. | Building recognition based on improved faster R-CNN in high point monitoring image | |
CN109492654B (en) | Detection method and device for indoor endoscopic camera | |
Salahat et al. | Automated real-time video surveillance algorithms for soc implementation: A survey | |
CN105046637B (en) | Optical flow tracking implementation method based on OmapL138 chips | |
Xiong et al. | Research on surface quality monitoring of workpiece based on CornerNet |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180202 |
|
WD01 | Invention patent application deemed withdrawn after publication |