CN109581408A - A kind of method and system carrying out target identification using laser complex imaging - Google Patents
A kind of method and system carrying out target identification using laser complex imaging Download PDFInfo
- Publication number
- CN109581408A CN109581408A CN201811503301.0A CN201811503301A CN109581408A CN 109581408 A CN109581408 A CN 109581408A CN 201811503301 A CN201811503301 A CN 201811503301A CN 109581408 A CN109581408 A CN 109581408A
- Authority
- CN
- China
- Prior art keywords
- target
- signal
- point
- point cloud
- echo
- 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
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/88—Lidar systems specially adapted for specific applications
- G01S17/89—Lidar systems specially adapted for specific applications for mapping or imaging
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Electromagnetism (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a kind of method and systems that target identification is carried out using laser complex imaging, which comprises Range Profile, intensity image, azimuth and the pitch angle of target each point are obtained by laser complex imaging；Three dimensional point cloud reconstruct is carried out to the Range Profile, intensity image, azimuth and pitch angle, obtains three dimensional point cloud, three dimensional point cloud includes amplitude, height, shape, frequency spectrum and the scattered information of target；The three dimensional point cloud is filtered, size, shape, height, distance, the physical characteristic parameter of scattering and superficial layer of estimation target are converted by hyperspace, according to Discrimination Functions determined by a large amount of training samples, in conjunction with the property data base of known typical target, identified in classifier.Basis of the present invention using the Range Profile of target, intensity image, azimuth and pitch angle as target identification carries out more accurate identification to target, solves the not high technical problem of existing target identification technology recognition accuracy.
Description
Technical field
The invention belongs to target identification technology field, especially a kind of side that target identification is carried out using laser complex imaging
Method and system.
Background technique
There are mainly three types of the technical solutions of target identification at present, the first is to carry out target identification, the program using radar
It identifies that spatial resolution and the distance resolution (picture quality) of target are poor, is easy by electromagnetic interference；Second is using red
Outer imaging carries out target identification, and the program can only identify the two dimensional image of target and cannot obtain the range information of target, and
The performance of infrared imaging device will receive the influence of target temperature characteristic；The third is somebody's turn to do using the scheme of slice range gating
Scheme stitching algorithm is complicated and is difficult to real-time processing.
Summary of the invention
In order to overcome drawbacks described above, the technical problem to be solved in the present invention is to provide know at least to solve existing target
The not high technical problem of other technology recognition accuracy.
In order to solve the above technical problems, one of embodiment of the present invention carries out target identification using laser complex imaging
Method includes:
Range Profile, intensity image, azimuth and the pitch angle of target each point are obtained by laser complex imaging；
Three dimensional point cloud reconstruct is carried out to the Range Profile, intensity image, azimuth and pitch angle, obtains three-dimensional point cloud number
According to three dimensional point cloud includes amplitude, height, shape, frequency spectrum and the scattered information of target；
The three dimensional point cloud is filtered；
Pass through the size of hyperspace transformation estimation target mathematically, shape, height, distance, scattering and superficial layer
Physical characteristic parameter, last Discrimination Functions according to determined by a large amount of training samples, in conjunction with the characteristic of known typical target
Library is identified in classifier；
Wherein, the three dimensional point cloud is filtered, is specifically included:
Spatially three dimensional point cloud is filtered, a value range in an x, y, z direction is spatially set, with mistake
Filter the point cloud data not in field range；
Single-point filtering, searches the consecutive points cloud data amount check in pre-set radius or adjacent domains, when number is less than default threshold
When value, the point cloud data is removed, to filter out isolated point cloud data；
Statistical filtering calculates the mean value of 32 point cloud data distances of arest neighbors to each point, removes distance in a standard
Point cloud data outside difference removes unsocial discrete point；
Layered shaping is carried out to point cloud data；
Point cloud data is split, point cloud data is divided into different connected region blocks.
Optionally, the Range Profile and intensity image that target each point is obtained by laser complex imaging, specifically include:
The signal that exploring laser light issues, the time that recording laser signal issues；
The laser signal is beamed into multiple beamlets, multiple beamlets are respectively radiated to target each point；
Receive the echo-signal that multiple beamlets are respectively radiated to after target each point, record that every beam echo-signal reaches when
Between and range value；
The Range Profile that target each point is calculated according to the difference of the time of the sending and the time of the arrival, returns every beam
The range value of wave signal is sampled to obtain the intensity image of target each point.
Optionally, the multiple beamlets of reception are respectively radiated to the echo-signal after target each point, specifically include:
The echo-signal after multiple beamlets are respectively radiated to target each point is received respectively by multiple single-element detectors.
Optionally, the laser signal is beamed into after multiple beamlets, further includes:
To the multiple beamlet carry out shaping transformation, make the multiple beamlet respectively with the multiple single-element detector
Match.
Optionally, it is described receive the echo-signal after multiple beamlets are respectively radiated to target each point before, further includes:
The multiple beamlet is respectively radiated to the echo-signal after target each point by receiving optics, makes beamlet
The angle of divergence it is consistent with the spatial resolution of single-element detector.
Optionally, the laser signal is beamed into after multiple beamlets, further includes:
Laser emission optical path is tuned into laser pick-off optical path coaxial.
Optionally, the multiple beamlets of reception are respectively radiated to the echo-signal after target each point, record every Shu Huibo
The time and range value that signal reaches；Target each point is calculated according to the difference of the time of the sending and the time of the arrival
Range Profile is sampled to obtain the intensity image of target each point, be specifically included to the range value of every beam echo-signal:
Multiple single-element detectors receive the echo-signal that multiple beamlets are respectively radiated to after target each point, echo letter respectively
Number by photoelectric conversion be electric signal, electric signal is a multiple signals, the multiple signals packet after amplifying circuit amplifies
It includes:
First via signal, is sent to counting circuit after trigger circuit generates echo-signal, and record echo-signal reaches meter
The time of number circuit, counting circuit calculate the time of the sending and the difference of the time for reaching counting circuit, Jin Erji
Calculate the Range Profile of target each point；
Second road signal, by peak holding circuit, peak holding circuit is used to obtain peak value and the holding of amplifying circuit
For a period of time, make sample circuit there is time enough to sample the range value of echo-signal, obtain the intensity of target each point
Picture.
Optionally, the multiple signals further include:
Third road signal makes unit for removing the grid bias power supply of control unit detector as noise level detection signal
Detector always works in just bias state.
In order to solve the above technical problems, one of embodiment of the present invention carries out target identification using laser complex imaging
System includes:
Module is obtained, for obtaining Range Profile, intensity image, azimuth and the pitching of target each point by laser complex imaging
Angle；
Reconstructed module carries out three dimensional point cloud reconstruct to the Range Profile, intensity image, azimuth and pitch angle, obtains
Three dimensional point cloud, three dimensional point cloud include amplitude, height, shape, frequency spectrum and the scattered information of target；
The filtering processing module, for being filtered to the three dimensional point cloud；
Identification module, by the size of hyperspace mathematically transformation estimation target, shape, height, distance, scattering and
The physical characteristic parameter of superficial layer, last Discrimination Functions according to determined by a large amount of training samples, in conjunction with known typical target
Property data base is identified in classifier；
Module, which is filtered, includes:
An x, y, z side is spatially arranged for spatially filtering to three dimensional point cloud in three dimensional point cloud filtering
To a value range, to filter out the not point cloud data in field range；
Single-point filter module is filtered for single-point, searches the consecutive points cloud data amount check in pre-set radius or adjacent domains,
When number is less than preset threshold, the point cloud data is removed, to filter out isolated point cloud data；
Statistical filtering module is used for statistical filtering, to each point, calculates the mean value of 32 point cloud data distances of arest neighbors,
Point cloud data of the distance outside a standard deviation is removed, unsocial discrete point is removed；
Hierarchical processing module, for carrying out layered shaping to point cloud data；
Divide module and point cloud data is divided into different connected region blocks for being split to point cloud data.
Optionally, the acquisition module includes:
Photodetector, the time issued for the signal that exploring laser light issues, recording laser signal；
Beam expanding lens, for the laser signal to be beamed into multiple beamlets, it is each that multiple beamlets are respectively radiated to target
Point；
Mirror is received, for receiving the echo-signal after multiple beamlets are respectively radiated to target each point；
Recording unit, for recording the time and range value that every beam echo-signal reaches；
Counting circuit, for the difference according to time of time and arrival of the sending calculate target each point away from
From picture；
Sample circuit is sampled to obtain the intensity image of target each point for the range value to every beam echo-signal.
The embodiment of the present invention obtains Range Profile, intensity image, azimuth and the pitching of target each point by laser complex imaging
Angle；Three dimensional point cloud reconstruct is carried out to the Range Profile, intensity image, azimuth and pitch angle, obtains three dimensional point cloud, it is right
The three dimensional point cloud is filtered, and is specifically included: spatially being filtered to three dimensional point cloud, single-point filtering, system
Meter filtering, layered shaping, point cloud data segmentation pass through size, the shape, height of hyperspace transformation estimation target mathematically
Degree, distance, the physical characteristic parameter of scattering and superficial layer, last Discrimination Functions according to determined by a large amount of training samples, in conjunction with
The property data base of known typical target, is identified in classifier, solves existing target identification technology recognition accuracy
Not high technical problem.
The above description is only an overview of the technical scheme of the present invention, in order to better understand the technical means of the present invention,
And it can be implemented in accordance with the contents of the specification, and in order to allow above and other objects of the present invention, feature and advantage can
It is clearer and more comprehensible, the followings are specific embodiments of the present invention.
Detailed description of the invention
By reading the following detailed description of the preferred embodiment, various other advantages and benefits are common for this field
Technical staff will become clear.The drawings are only for the purpose of illustrating a preferred embodiment, and is not considered as to the present invention
Limitation.And throughout the drawings, the same reference numbers will be used to refer to the same parts.In the accompanying drawings:
Fig. 1 is a kind of flow chart for the method that target identification is carried out using laser complex imaging of the embodiment of the present invention；
Fig. 2, which is that the embodiment of the present invention is a kind of, to be carried out in the method for target identification using laser complex imaging to the three-dimensional point
The flow chart that cloud data are filtered；
Fig. 3 is a kind of flow chart for the system that target identification is carried out using laser complex imaging of the embodiment of the present invention；
Fig. 4 is that module is filtered in a kind of system for carrying out target identification using laser complex imaging of the embodiment of the present invention
Flow chart；
Fig. 5 is a kind of flow chart for the system that target identification is carried out using laser complex imaging of the embodiment of the present invention.
Specific embodiment
Exemplary embodiments of the present disclosure are described in more detail below with reference to accompanying drawings.Although showing the disclosure in attached drawing
Exemplary embodiment, it being understood, however, that may be realized in various forms the disclosure without should be by embodiments set forth here
It is limited.On the contrary, these embodiments are provided to facilitate a more thoroughly understanding of the present invention, and can be by the scope of the present disclosure
It is fully disclosed to those skilled in the art.
In subsequent description, it is only using the suffix for indicating such as " module ", " component " or " unit " of element
Be conducive to explanation of the invention, itself there is no a specific meaning.Therefore, " module ", " component " or " unit " can mix
Ground uses.
Using for distinguishing element " first ", the prefixes such as " second " only for being conducive to explanation of the invention,
Itself is without specific meaning.
Embodiment one
As shown in Fig. 1,2 and 5, the embodiment of the present invention provides a kind of side that target identification is carried out using laser complex imaging
Method, comprising:
S101 obtains Range Profile, intensity image, azimuth and the pitch angle of target each point by laser complex imaging；
S102 carries out three dimensional point cloud reconstruct to the Range Profile, intensity image, azimuth and pitch angle, obtains three-dimensional
Point cloud data, three dimensional point cloud include amplitude, height, shape, frequency spectrum and the scattered information of target；
S103 is filtered the three dimensional point cloud, specifically includes:
S104 passes through size, shape, height, distance, scattering and the surface of hyperspace transformation estimation target mathematically
The physical characteristic parameter of layer, last Discrimination Functions according to determined by a large amount of training samples, in conjunction with the feature of known typical target
Database is identified in classifier.
Wherein, the three dimensional point cloud is filtered, is specifically included:
S201 spatially filters three dimensional point cloud, and a value model in an x, y, z direction is spatially arranged
It encloses, to filter out the not point cloud data in field range；
S202, single-point filtering, searches the consecutive points cloud data amount check in pre-set radius or adjacent domains, when number is less than pre-
If when threshold value, removing the point cloud data, to filter out isolated point cloud data；
S203, statistical filtering calculate the mean value of 32 point cloud data distances of arest neighbors to each point, remove distance one
Point cloud data outside a standard deviation removes unsocial discrete point；
S204 carries out layered shaping to point cloud data；
Since laser point cloud data has distance value, only the point cloud data of some specific range value can be handled,
It can reduce processing ambient noise, reduce processing data volume.
S205 is split point cloud data, and point cloud data is divided into different connected region blocks；
The embodiment of the present invention obtains Range Profile, intensity image, azimuth and the pitching of target each point by laser complex imaging
Angle；Three dimensional point cloud reconstruct is carried out to the Range Profile, intensity image, azimuth and pitch angle, obtains three dimensional point cloud, it is right
The three dimensional point cloud is filtered, and is specifically included: spatially being filtered to three dimensional point cloud, single-point filtering, system
Meter filtering, layered shaping, point cloud data segmentation pass through size, the shape, height of hyperspace transformation estimation target mathematically
Degree, distance, the physical characteristic parameter of scattering and superficial layer, last Discrimination Functions according to determined by a large amount of training samples, in conjunction with
The property data base of known typical target, is identified in classifier, solves existing target identification technology recognition accuracy
Not high technical problem.
Optionally, the Range Profile and intensity image that target each point is obtained by laser complex imaging, specifically include:
S301, the signal that exploring laser light issues, the time that recording laser signal issues；
Laser source of the invention can be used up using the impulse ejection source of Gao Zhongying when laser emitting source goes out light
Electric explorer is sampled the laser signal of sending, for the time that recording laser signal issues, after sampled signal is sent to
Continuous processing, marks main wave timing to start.
The laser signal is beamed into multiple beamlets by S302, and multiple beamlets are respectively radiated to target each point；
The laser of laser source transmitting can pass through beam expanding lens, compress laser angular to original 1/5th, by the laser
Signal is beamed into multiple beamlets, which has corrected the basic aberrations such as spherical aberration astigmatism coma well, is meeting aberration
Compression volume, beam expanding lens can use inverted Galilean telescope structure as far as possible while demand.Receive multiple sub-lights
It is spaced between each pixel of multiple single-element detectors of beam, if irradiating target each point using light beam, has
A part of light is reflected on the interval of multiple single-element detectors, will result in the waste of laser energy in this way, influence detection away from
From.
Optionally, the laser signal is beamed into after multiple beamlets, further includes:
To the multiple beamlet carry out shaping transformation, make the multiple beamlet respectively with the multiple single-element detector
Match.
Laser after beam expanding lens remains as Gaussian beam, high due to needing to carry out accurately image to target each point
The intensity of target each point is uneven after this light beam is irradiated in target, influences whether the consistency and precision of imaging.
In this patent using liquid crystal phase controller come to after expanding Gaussian beam carry out shaping transformation, become with
The array beamlet that multiple single-element detectors match.The production of the phase diaphragm of liquid crystal phase controller is imitated according to diffraction optics
Rate > 95%, the criterion for being divided uniformity > 90% are completed, wherein the uniformity and input window of array beamlet peak light intensity
The periodicity of grating is related in mouthful, and when being illuminated with Gauss light wave, the periodicity for choosing grating is equal to 6, beamlet array
There is 90% uniformity.By the beamlet optical efficiency after liquid crystal phase controller, according to the transmittance function by grating
The calculating for carrying out Newton iteration method, obtains phase distribution value when optical efficiency > 95%.The phase of phase diaphragm will be calculated
Bit distribution value is input to computer, opens the control software of liquid crystal phase controller, phase diaphragm is loaded into liquid crystal phase control
Device processed obtains the beamlet for meeting performance indicator.Liquid crystal phase controller is a hardware, which type of light beam it specifically generates
Can be by software control, grating is exactly the process for changing software according to demand in input window.
Optionally, the laser signal is beamed into after multiple beamlets, further includes:
Laser emission optical path is tuned into laser pick-off optical path coaxial.
When carrying out complex imaging to target, need first to be tuned into transmitting optical path and receiving light path coaxially.Due to emitting light
The final output on road be beam splitting after beamlet, therefore can using etc. adjusted than the method for model coaxially.This method comprises: first
The first model according to big little makings one same size of multiple single-element detectors has equally distributed opening on model, opening
Size and single-element detector it is in the same size, be spaced it is also consistent with the interval of single-element detector.Model is placed on unit to visit
It surveys on the position of device, a laser source is placed beside model.Laser emitting source is opened, after optical transmitting system and liquid crystal
After phase controller, laser beamlets, in the focal plane imaging of concave mirror, open simultaneously swashing beside model after concave mirror
Light source, laser source issue array light by the opening on model, and the light issued is also after concave mirror in the coke of concave mirror
Face imaging places the distribution situation that speckle analysis instrument observes two beam array beams in focal plane, and hot spot is beamlet in nothing at focal plane
Remote distribution situation is limited, the pitching azimuth of receiving optics is adjusted, is observed by speckle analysis instrument, when two array sub-lights
Beam completes the coaxial of Laser emission optical path and laser pick-off optical path when being overlapped on laser beam analyzer.Laser source can use
Low-power laser source.
S303 receives the echo-signal that multiple beamlets are respectively radiated to after target each point, records every beam echo-signal and arrives
The time reached and range value；
Optionally, the multiple beamlets of reception are respectively radiated to the echo-signal after target each point, specifically include: passing through
Multiple single-element detectors receive the echo-signal that multiple beamlets are respectively radiated to after target each point respectively.
S304 calculates the Range Profile of target each point according to the difference of the time of the sending and the time of the arrival, right
The range value of every beam echo-signal is sampled to obtain the intensity image of target each point.
The signal that the embodiment of the present invention is issued by exploring laser light, the time that recording laser signal issues；It then will be described
Laser signal is beamed into multiple beamlets, and multiple beamlets are respectively radiated to target each point；Then multiple beamlet difference are received
Echo-signal after being irradiated to target each point records time and range value that every beam echo-signal reaches；Finally according to the hair
The difference of the time of time and the arrival out calculate the Range Profile of target each point, carry out to the range value of every beam echo-signal
Sampling obtains the intensity image of target each point, in this, as the basis of target identification, carries out more accurate identification to target, solves
The not high technical problem of existing target identification technology recognition accuracy.
Laser three-dimensional imaging used by the embodiment of the present invention compared with radar, divide by the space that it has radar incomparable
Resolution and distance resolution (picture quality), also not vulnerable to electromagnetic interference；Compared with visible light and infrared imaging, they all have
Similar picture quality, but visible light and the two dimensional image of infrared only target be without range information, and visible light and red
The performance of outer imaging device is illuminated by the light the influence of condition and target temperature characteristic respectively, and laser imaging is using active laser
Illumination, imaging effect and use condition relationship are little；Compared with sweep type laser imaging, it has imaging rate height, image
Quality do not influenced by optical quality and scanning accuracy, without high-speed moving part, high reliablity, do not need stabilized platform, user
Just the features such as.Laser three-D is modulated with slice gating formula laser three-dimensional imaging, more slit streak tube laser three-dimensional imagings, continuous wave
The imaging modes such as imaging, gain modulation laser three-dimensional imaging are compared, and are had imaging fast, high resolution, are realized that simple, distance is remote
Feature.
Optionally, it is described receive the echo-signal after multiple beamlets are respectively radiated to target each point before, further includes:
The multiple beamlet is respectively radiated to the echo-signal after target each point by receiving optics, makes beamlet
The angle of divergence it is consistent with the spatial resolution of single-element detector.
The echo that beamlet is irradiated to target each point back reflection is entered multiple units to detect after receiving optics
Device, multiple single-element detectors constitute detector array, and detector array can be using array APD (avalanche photodide)
Single-element detector, array APD single-element detector refer to that detector is not unit pixel, are an array, such as 1024*
1024.Receiving optics receives system using non-parallel optical path.Non-parallel optical path receives system and narrow band filter, echo is housed
For laser with certain angle by narrow band filter, such optical path volume is compacter.Echo laser passes through receiving optics
Primary mirror, field stop, collimating eye-piece, 1.55 μm of interferometric filters, the single-element detector for entering array APD after optically focused microscope group,
The angle of divergence of beamlet and the spatial resolution of single-element detector are consistent, if the angle of divergence of beamlet is too big, can waste laser
Energy.
Optionally, the multiple beamlets of reception are respectively radiated to the echo-signal after target each point, record every Shu Huibo
The time and range value that signal reaches；Target each point is calculated according to the difference of the time of the sending and the time of the arrival
Range Profile is sampled to obtain the intensity image of target each point, be specifically included to the range value of every beam echo-signal:
Multiple single-element detectors receive the echo-signal that multiple beamlets are respectively radiated to after target each point, echo letter respectively
Number by photoelectric conversion be electric signal, electric signal is a multiple signals, the multiple signals packet after amplifying circuit amplifies
It includes:
First via signal, is sent to counting circuit after trigger circuit generates echo-signal, and record echo-signal reaches meter
The time of number circuit, counting circuit calculate the time of the sending and the difference of the time for reaching counting circuit, Jin Erji
Calculate the Range Profile of target each point；
Second road signal, by peak holding circuit, peak holding circuit is used to obtain peak value and the holding of amplifying circuit
For a period of time, make sample circuit there is time enough to sample the range value of echo-signal, obtain the intensity of target each point
Picture.
Optionally, the multiple signals further include:
Third road signal makes unit for removing the grid bias power supply of control unit detector as noise level detection signal
Detector always works in just bias state.
Beamlet into array APD single-element detector becomes electric signal after photoelectric conversion, and electric signal is by preposition
Using secondary amplification after amplification, secondary amplification includes trsanscondutance amplifier, variable gain amplifier, driving amplifier.It is secondary to put
The crosstalk between each road echo-signal is considered greatly, by noise and signal in the different characteristics of frequency domain, is introduced frequency domain information, is subtracted
Small crosstalk.
It is the signal of a multichannel by amplified signal, signal is divided into three parts, a part of triggered circuit in road
It generates door signal (echo) and is sent to counting circuit, record echo-signal reaches the time of counting circuit, and counting circuit passes through institute
The difference of the time of sending and the time of the arrival counting circuit are stated to generate Range Profile.Specifically, Range ProfileC is the light velocity, and t is the difference.
For first part, due to the laser emitting source using 1kHz, array APD single-element detector ambient noise,
Dark current noise and thermal noise will generate a large amount of false-alarm, and this patent has a preferable temporal correlation and bias light for signal
False-alarm random distribution this feature generated with noise takes the method for Multiple pulse superposition by true target range information from big
It is extracted in the false-alarm of amount, to obtain the 3-D image of high quality.There is the corresponding Bo Mennei of target, echo occurs general
Rate can be bigger, and the position that noise occurs is random, therefore this method can increase the signal-to-noise ratio of image.When bias light compared with
The false alarm rate of Qiang Shi, array APD single-element detector will be very high.The processing of false-alarm is the master of array APD single-element detector detection
Want difficult point.1000 are uniformly divided into distance range apart from small echo door.When there is target in somewhere, there is echo in the range gate
Probability will increase, for export 20Hz data transfer rate, as long as therefore by after 50 pulse Multiple pulse superpositions, signal can be made to believe
It makes an uproar than being greatly improved with range capability.For the signal after superposition processing, then spatially using the method for filtering, to passing through phase
Range information between adjacent pixel uses Wiener filtering, median filtering, further increases signal-to-noise ratio and range capability.
Second part passes through peak holding circuit, when peak holding circuit obtains the peak value of amplifying circuit and kept for one section
Between, make sample circuit there is time enough to carry out digitized sampling to the range value of echo-signal, obtains the intensity of target each point
Information.The method for also using Multiple pulse superposition simultaneously, the intensity of 50 impulsive measurements is added up, the letter of strength signal is improved
It makes an uproar ratio.
Part III removes control array APD single-element detector grid bias power supply as noise level detection signal, visits unit
It surveys device and always works in just bias state, gain control circuit can make the gain of system smaller in short distance, in long distance
From when gain it is larger.
Step more than, Range Profile and intensity image to target each point carry out three dimensional point cloud reconstruct, target
Three-dimensional data information response is the different distance value of target each point, and intensity image reaction is for difference in a target
The two is combined and carries out target identification by reflectivity.Removal is less than the three dimensional point cloud of predetermined threshold range first and observation regards
Over-the-counter three dimensional point cloud；Then pixel point height and size three-dimensional point cloud identical with each point height of target and size are extracted
Data, and therefrom extract the characteristic information of target each point；Classify later to the three dimensional point cloud；Finally to described three
It ties up point cloud data and carries out spatio-temporal filtering, obtain interested target information, carry out the identification of target.
Hardware/Software Collaborative Design thought is used to the processing of intensity image and Range Profile above, is based on advanced FPGA+DSP structure
Frame, FPGA use the TMS320c6455 of TI company using xilinx virtex6 series, DSP, to the Range Profile of multichannel target and
Intensity image is handled.It can complete acquisition, pretreatment, the function of algorithm process, showing, store and transmit.
Embodiment two
As shown in Figures 3 and 4, the embodiment of the present invention provides a kind of system for carrying out target identification using laser complex imaging,
Include:
Module is obtained, for obtaining Range Profile, intensity image, azimuth and the pitching of target each point by laser complex imaging
Angle；
Reconstructed module carries out three dimensional point cloud reconstruct to the Range Profile, intensity image, azimuth and pitch angle, obtains
Three dimensional point cloud, three dimensional point cloud include amplitude, height, shape, frequency spectrum and the scattered information of target；
The filtering processing module, for being filtered to the three dimensional point cloud；
Identification module, by the size of hyperspace mathematically transformation estimation target, shape, height, distance, scattering and
The physical characteristic parameter of superficial layer, last Discrimination Functions according to determined by a large amount of training samples, in conjunction with known typical target
Property data base is identified in classifier.
Module, which is filtered, includes:
An x, y, z side is spatially arranged for spatially filtering to three dimensional point cloud in three dimensional point cloud filtering
To a value range, to filter out the not point cloud data in field range；
Single-point filter module is filtered for single-point, searches the consecutive points cloud data amount check in pre-set radius or adjacent domains,
When number is less than preset threshold, the point cloud data is removed, to filter out isolated point cloud data；
Statistical filtering module is used for statistical filtering, to each point, calculates the mean value of 32 point cloud data distances of arest neighbors,
Point cloud data of the distance outside a standard deviation is removed, unsocial discrete point is removed；
Hierarchical processing module, for carrying out layered shaping to point cloud data；
Divide module and point cloud data is divided into different connected region blocks for being split to point cloud data；
Optionally, the acquisition module includes:
Photodetector, the time issued for the signal that exploring laser light issues, recording laser signal；
Beam expanding lens, for the laser signal to be beamed into multiple beamlets, it is each that multiple beamlets are respectively radiated to target
Point；
Mirror is received, for receiving the echo-signal after multiple beamlets are respectively radiated to target each point；
Recording unit, for recording the time and range value that every beam echo-signal reaches；
Counting circuit, for the difference according to time of time and arrival of the sending calculate target each point away from
From picture；
Sample circuit is sampled to obtain the intensity image of target each point for the range value to every beam echo-signal.
Optionally, the reception mirror is avalanche photodide, and the avalanche photodide includes multiple unit detections
Device, the multiple single-element detector receive the echo-signal that multiple beamlets are respectively radiated to after target each point respectively.
Optionally, the system also includes:
Phase controller, for the multiple beamlet carry out shaping transformation, make the multiple beamlet respectively with institute
Multiple single-element detectors are stated to match.
Optionally, the system also includes:
Receiving optics, the echo-signal after target each point is respectively radiated to for the multiple beamlet is by receiving
Optical system keeps the angle of divergence of beamlet consistent with the spatial resolution of single-element detector.
Optionally, the system also includes:
It is coaxial for Laser emission optical path to be tuned into laser pick-off optical path with axial adjustment module.
Optionally, the system also includes:
Photoelectric conversion circuit is converted to telecommunications for multiple beamlets to be respectively radiated to the echo-signal after target each point
Number；
Amplification circuit module, for amplifying the electric signal, amplified electric signal is a multiple signals, described more
Road signal includes:
First via signal is sent to logging modle after trigger circuit generates echo-signal；
Second road signal, by peak holding circuit, peak holding circuit is used to obtain peak value and the holding of amplifying circuit
For a period of time, make sample circuit there is time enough to sample the range value of echo-signal, obtain the intensity of target each point
Picture.
Third road signal makes unit for removing the grid bias power supply of control unit detector as noise level detection signal
Detector always works in just bias state.
The function of system described in the embodiment of the present invention is described in embodiment of the method, therefore this system is implemented
Not detailed place in the description of example, may refer to the related description in preceding method embodiment, this will not be repeated here.
It should be noted that, in this document, the terms "include", "comprise" or its any other variant are intended to non-row
His property includes, so that the process, method, article or the device that include a series of elements not only include those elements, and
And further include other elements that are not explicitly listed, or further include for this process, method, article or device institute it is intrinsic
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including being somebody's turn to do
There is also other identical elements in the process, method of element, article or device.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment side
Method can be realized by means of software and necessary general hardware platform, naturally it is also possible to by hardware, but in many cases
The former is more preferably embodiment.Based on this understanding, technical solution of the present invention substantially in other words does the prior art
The part contributed out can be embodied in the form of software products, which is stored in a storage medium
In (such as ROM/RAM, magnetic disk, CD), including some instructions are used so that a terminal (can be mobile phone, computer, service
Device, air conditioner or network equipment etc.) execute method described in each embodiment of the present invention.
The embodiment of the present invention is described with above attached drawing, but the invention is not limited to above-mentioned specific
Embodiment, the above mentioned embodiment is only schematical, rather than restrictive, those skilled in the art
Under the inspiration of the present invention, without breaking away from the scope protected by the purposes and claims of the present invention, it can also make very much
Form, all of these belong to the protection of the present invention.
Claims (10)
1. a kind of method for carrying out target identification using laser complex imaging characterized by comprising
Range Profile, intensity image, azimuth and the pitch angle of target each point are obtained by laser complex imaging；
Three dimensional point cloud reconstruct is carried out to the Range Profile, intensity image, azimuth and pitch angle, obtains three dimensional point cloud,
Three dimensional point cloud includes amplitude, height, shape, frequency spectrum and the scattered information of target；
The three dimensional point cloud is filtered；
By the size of hyperspace mathematically transformation estimation target, shape, height, distance, scattering and superficial layer physics
Characterisitic parameter, last Discrimination Functions according to determined by a large amount of training samples, in conjunction with the property data base of known typical target,
It is identified in classifier；
Wherein, the three dimensional point cloud is filtered, is specifically included:
Spatially three dimensional point cloud is filtered, a value range in an x, y, z direction is spatially set, to filter out
Not in the point cloud data of field range；
Single-point filtering, searches the consecutive points cloud data amount check in pre-set radius or adjacent domains, when number is less than preset threshold,
The point cloud data is removed, to filter out isolated point cloud data；
Statistical filtering calculates the mean value of 32 point cloud data distances of arest neighbors to each point, removes distance outside a standard deviation
Point cloud data, remove unsocial discrete point；
Layered shaping is carried out to point cloud data；
Point cloud data is split, point cloud data is divided into different connected region blocks.
2. the method as described in claim 1, which is characterized in that by laser complex imaging obtain target each point Range Profile and
Intensity image specifically includes:
The signal that exploring laser light issues, the time that recording laser signal issues；
The laser signal is beamed into multiple beamlets, multiple beamlets are respectively radiated to target each point；
Receive the echo-signal that multiple beamlets are respectively radiated to after target each point, record time that every beam echo-signal reaches and
Range value；
The Range Profile that target each point is calculated according to the difference of the time of the sending and the time of the arrival, believes every Shu Huibo
Number range value sampled to obtain the intensity image of target each point.
3. method according to claim 2, which is characterized in that after the multiple beamlets of reception are respectively radiated to target each point
Echo-signal, specifically include:
The echo-signal after multiple beamlets are respectively radiated to target each point is received respectively by multiple single-element detectors.
4. method as claimed in claim 3, which is characterized in that after the laser signal is beamed into multiple beamlets, also
Include:
To the multiple beamlet carry out shaping transformation, make the multiple beamlet respectively with the multiple single-element detector phase
Match.
5. method as claimed in claim 3, which is characterized in that after the multiple beamlets of reception are respectively radiated to target each point
Echo-signal before, further includes:
The multiple beamlet is respectively radiated to the echo-signal after target each point by receiving optics, makes the hair of beamlet
It is consistent with the spatial resolution of single-element detector to dissipate angle.
6. method as claimed in claim 3, which is characterized in that after the laser signal is beamed into multiple beamlets, also
Include:
Laser emission optical path is tuned into laser pick-off optical path coaxial.
7. method according to claim 2, which is characterized in that after the multiple beamlets of reception are respectively radiated to target each point
Echo-signal, record the time and range value that every beam echo-signal reaches；According to the time of the sending and the arrival
The difference of time calculates the Range Profile of target each point, is sampled to obtain the strong of target each point to the range value of every beam echo-signal
Picture is spent, is specifically included:
Multiple single-element detectors receive the echo-signal that multiple beamlets are respectively radiated to after target each point, echo-signal warp respectively
Crossing photoelectric conversion is electric signal, and electric signal is a multiple signals after amplifying circuit amplifies, and the multiple signals include:
First via signal is sent to counting circuit after trigger circuit generates echo-signal, and record echo-signal, which reaches, counts electricity
The time on road, counting circuit calculates the time of the sending and the difference of the time for reaching counting circuit, and then calculates mesh
Mark the Range Profile of each point；
Second road signal, by peak holding circuit, peak holding circuit is used to obtain the peak value of amplifying circuit and is kept for one section
Time makes sample circuit have time enough to sample the range value of echo-signal, obtains the intensity image of target each point.
8. the method for claim 7, which is characterized in that the multiple signals further include:
Third road signal detects unit for removing the grid bias power supply of control unit detector as noise level detection signal
Device always works in just bias state.
9. a kind of system for carrying out target identification using laser complex imaging characterized by comprising
Module is obtained, for obtaining Range Profile, intensity image, azimuth and the pitch angle of target each point by laser complex imaging；
Reconstructed module carries out three dimensional point cloud reconstruct to the Range Profile, intensity image, azimuth and pitch angle, obtains three-dimensional
Point cloud data, three dimensional point cloud include amplitude, height, shape, frequency spectrum and the scattered information of target；
Module is filtered, for being filtered to the three dimensional point cloud；
Identification module passes through size, shape, height, distance, scattering and the surface of hyperspace transformation estimation target mathematically
The physical characteristic parameter of layer, last Discrimination Functions according to determined by a large amount of training samples, in conjunction with the feature of known typical target
Database is identified in classifier；
The filtering processing module includes:
An x, y, z direction is spatially arranged for spatially filtering to three dimensional point cloud in three dimensional point cloud filtering
One value range, to filter out the not point cloud data in field range；
Single-point filter module is filtered for single-point, the consecutive points cloud data amount check in pre-set radius or adjacent domains is searched, when a
When number is less than preset threshold, the point cloud data is removed, to filter out isolated point cloud data；
Statistical filtering module is used for statistical filtering, to each point, calculates the mean value of 32 point cloud data distances of arest neighbors, removal
Point cloud data of the distance outside a standard deviation, removes unsocial discrete point；
Hierarchical processing module, for carrying out layered shaping to point cloud data；
Divide module and point cloud data is divided into different connected region blocks for being split to point cloud data.
10. system as claimed in claim 9, which is characterized in that the acquisition module includes:
Photodetector, the time issued for the signal that exploring laser light issues, recording laser signal；
Beam expanding lens, for the laser signal to be beamed into multiple beamlets, multiple beamlets are respectively radiated to target each point；
Mirror is received, for receiving the echo-signal after multiple beamlets are respectively radiated to target each point；
Recording unit, for recording the time and range value that every beam echo-signal reaches；
Counting circuit, for being calculated at a distance from target each point according to the time of the sending and the difference of the time of the arrival
Picture；
Sample circuit is sampled to obtain the intensity image of target each point for the range value to every beam echo-signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811503301.0A CN109581408B (en) | 2018-12-10 | 2018-12-10 | Method and system for identifying target by using laser composite imaging |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811503301.0A CN109581408B (en) | 2018-12-10 | 2018-12-10 | Method and system for identifying target by using laser composite imaging |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109581408A true CN109581408A (en) | 2019-04-05 |
| CN109581408B CN109581408B (en) | 2023-01-06 |
Family
ID=65927965
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811503301.0A Active CN109581408B (en) | 2018-12-10 | 2018-12-10 | Method and system for identifying target by using laser composite imaging |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109581408B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110187356A (en) * | 2019-06-14 | 2019-08-30 | 中国科学技术大学 | Remote super-resolution single photon image reconstructing method |
| WO2021051281A1 (en) * | 2019-09-17 | 2021-03-25 | 深圳市大疆创新科技有限公司 | Point-cloud noise filtering method, distance measurement device, system, storage medium, and mobile platform |
| WO2021058016A1 (en) * | 2019-09-29 | 2021-04-01 | 睿镞科技(北京)有限责任公司 | Laser radar and method for generating laser point cloud data |
| CN112785615A (en) * | 2020-12-04 | 2021-05-11 | 浙江工业大学 | Engineering surface multi-scale filtering method based on extended two-dimensional empirical wavelet transform |
| CN112924983A (en) * | 2021-03-18 | 2021-06-08 | 哈尔滨工业大学 | Target velocity image detection system and method based on calculation correlation imaging |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002065155A1 (en) * | 2001-02-09 | 2002-08-22 | Commonwealth Scientific And Industrial Research Organisation | Lidar system and method |
| CN104049259A (en) * | 2014-07-01 | 2014-09-17 | 南京大学 | Lidar three-dimensional imaging system based on virtual instrument |
| US20150109603A1 (en) * | 2013-10-21 | 2015-04-23 | Electronics And Telecommunications Research Institute | Multi-wavelength image lidar sensor apparatus and signal processing method thereof |
| CN105044731A (en) * | 2015-08-31 | 2015-11-11 | 中国电子科技集团公司第十一研究所 | Laser three-dimensional imaging system and imaging method |
| CN106250881A (en) * | 2016-08-25 | 2016-12-21 | 深圳大学 | A kind of target identification method based on three dimensional point cloud and system |
| CN106970393A (en) * | 2017-03-14 | 2017-07-21 | 南京航空航天大学 | A kind of face battle array laser radar three-D imaging method based on CDMA |
| CN108107444A (en) * | 2017-12-28 | 2018-06-01 | 国网黑龙江省电力有限公司检修公司 | Substation's method for recognizing impurities based on laser data |
| CN108646260A (en) * | 2018-07-02 | 2018-10-12 | 中国科学院西安光学精密机械研究所 | A kind of gazing type is without lens laser three-dimensional image forming apparatus and imaging method |
-
2018
- 2018-12-10 CN CN201811503301.0A patent/CN109581408B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2002065155A1 (en) * | 2001-02-09 | 2002-08-22 | Commonwealth Scientific And Industrial Research Organisation | Lidar system and method |
| US20150109603A1 (en) * | 2013-10-21 | 2015-04-23 | Electronics And Telecommunications Research Institute | Multi-wavelength image lidar sensor apparatus and signal processing method thereof |
| CN104049259A (en) * | 2014-07-01 | 2014-09-17 | 南京大学 | Lidar three-dimensional imaging system based on virtual instrument |
| CN105044731A (en) * | 2015-08-31 | 2015-11-11 | 中国电子科技集团公司第十一研究所 | Laser three-dimensional imaging system and imaging method |
| CN106250881A (en) * | 2016-08-25 | 2016-12-21 | 深圳大学 | A kind of target identification method based on three dimensional point cloud and system |
| CN106970393A (en) * | 2017-03-14 | 2017-07-21 | 南京航空航天大学 | A kind of face battle array laser radar three-D imaging method based on CDMA |
| CN108107444A (en) * | 2017-12-28 | 2018-06-01 | 国网黑龙江省电力有限公司检修公司 | Substation's method for recognizing impurities based on laser data |
| CN108646260A (en) * | 2018-07-02 | 2018-10-12 | 中国科学院西安光学精密机械研究所 | A kind of gazing type is without lens laser three-dimensional image forming apparatus and imaging method |
Non-Patent Citations (2)
| Title |
|---|
| 李春苗等: "基于特征值的激光成像雷达目标识别算法", 《激光与红外》 * |
| 马超杰等: "成像激光雷达在自动目标识别中的应用", 《航空兵器》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110187356A (en) * | 2019-06-14 | 2019-08-30 | 中国科学技术大学 | Remote super-resolution single photon image reconstructing method |
| WO2021051281A1 (en) * | 2019-09-17 | 2021-03-25 | 深圳市大疆创新科技有限公司 | Point-cloud noise filtering method, distance measurement device, system, storage medium, and mobile platform |
| WO2021058016A1 (en) * | 2019-09-29 | 2021-04-01 | 睿镞科技(北京)有限责任公司 | Laser radar and method for generating laser point cloud data |
| CN112785615A (en) * | 2020-12-04 | 2021-05-11 | 浙江工业大学 | Engineering surface multi-scale filtering method based on extended two-dimensional empirical wavelet transform |
| CN112924983A (en) * | 2021-03-18 | 2021-06-08 | 哈尔滨工业大学 | Target velocity image detection system and method based on calculation correlation imaging |
| CN112924983B (en) * | 2021-03-18 | 2022-06-21 | 哈尔滨工业大学 | Target velocity image detection system and method based on calculation correlation imaging |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109581408B (en) | 2023-01-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109581408A (en) | A kind of method and system carrying out target identification using laser complex imaging | |
| CN103868831B (en) | Cloud particle Spectral structure measuring method and measuring system | |
| Henriksson et al. | Continuously scanning time-correlated single-photon-counting single-pixel 3-D lidar | |
| US6717655B2 (en) | Micropulse lidar system | |
| CN102879835B (en) | A kind of measuring method of laser precipitation weather phenomenon and laser precipitation weather phenomenon instrument | |
| CN103411891A (en) | Terahertz (THz) super-resolution imaging method and system | |
| CN106646426B (en) | A kind of full optical fiber laser radar of multiple illuminators and single receiver telescope array | |
| CN106932097B (en) | A kind of dual-waveband imaging associates the weak signal target detection device and method that full spectrum surveys spectrum | |
| CN101776760A (en) | Laser three-dimensional imaging device based on single-photon detector | |
| WO2009064626A1 (en) | Sensing using polarization diversity and wavelength dependent backscatter | |
| CN110031821A (en) | A kind of vehicle-mounted avoidance laser radar waveform extracting method, laser radar and medium | |
| CN102692394B (en) | Two-dimensional imaging method and device based on thermal lens effect | |
| US7148974B1 (en) | Method for tracking the location of mobile agents using stand-off detection technique | |
| Guo et al. | Development of a single-wavelength airborne bathymetric LiDAR: System design and data processing | |
| Du et al. | Single-photon detection approach for autonomous vehicles sensing | |
| Wang et al. | Multichannel interconnection decomposition for hyperspectral LiDAR waveforms detected from over 500 m | |
| Billard et al. | Sea surface and depth detection in the WRELADS airborne depth sounder | |
| Zhang et al. | The intelligent vehicle target recognition algorithm based on target infrared features combined with lidar | |
| US8805075B2 (en) | Method and apparatus for identifying a vibrometry spectrum in imaging applications | |
| Bobrovnikov et al. | A Multi-Aperture Transceiver System of a Lidar with Narrow Field of View and Minimal Dead Zone | |
| CN111142121B (en) | Brillouin scattering temperature measurement laser radar system based on two-stage virtual image phase array | |
| Razenkov | Optimization of parameters of a turbulent lidar | |
| CN116165635A (en) | Denoising method for photon cloud data of different beams under daytime condition of multistage filtering algorithm | |
| CN205514536U (en) | Frequency domain optics chromatographic device that is concerned with | |
| Li et al. | Achieve accurate recognition of 3D point cloud images by studying the scattering characteristics of typical targets |
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 |