WO2015101062A1 - 一种多波段共光路图谱联合遥感测量系统及方法 - Google Patents

一种多波段共光路图谱联合遥感测量系统及方法 Download PDF

Info

Publication number
WO2015101062A1
WO2015101062A1 PCT/CN2014/085762 CN2014085762W WO2015101062A1 WO 2015101062 A1 WO2015101062 A1 WO 2015101062A1 CN 2014085762 W CN2014085762 W CN 2014085762W WO 2015101062 A1 WO2015101062 A1 WO 2015101062A1
Authority
WO
WIPO (PCT)
Prior art keywords
interest
infrared
mirror
module
feature
Prior art date
Application number
PCT/CN2014/085762
Other languages
English (en)
French (fr)
Inventor
张天序
戴小兵
刘祥燕
季剑飞
何旭东
高鹏程
Original Assignee
华中科技大学
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 华中科技大学 filed Critical 华中科技大学
Priority to US15/108,175 priority Critical patent/US9869793B2/en
Publication of WO2015101062A1 publication Critical patent/WO2015101062A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V8/00—Prospecting or detecting by optical means
    • G01V8/10—Detecting, e.g. by using light barriers
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
    • G01J3/00—Spectrometry; Spectrophotometry; Monochromators; Measuring colours
    • G01J3/02—Details
    • G01J3/0205—Optical elements not provided otherwise, e.g. optical manifolds, diffusers, windows
    • G01J3/0208—Optical elements not provided otherwise, e.g. optical manifolds, diffusers, windows using focussing or collimating elements, e.g. lenses or mirrors; performing aberration correction
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
    • G01J3/00—Spectrometry; Spectrophotometry; Monochromators; Measuring colours
    • G01J3/28—Investigating the spectrum
    • G01J3/45—Interferometric spectrometry
    • G01J3/453—Interferometric spectrometry by correlation of the amplitudes
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
    • G01J3/00—Spectrometry; Spectrophotometry; Monochromators; Measuring colours
    • G01J3/28—Investigating the spectrum
    • G01J3/2823—Imaging spectrometer
    • G01J2003/2826—Multispectral imaging, e.g. filter imaging

Definitions

  • the invention belongs to the field of remote sensing measurement technology, and more particularly relates to a multi-band common light path map combined remote sensing measurement system and method.
  • the target infrared information refers to the difference between the target acquired by the infrared spectrum sensor and the background radiation, reflection and scattering characteristics, including the radiation, reflection of the short, medium and long wave infrared fine spectral (line) bands. And scattering properties.
  • the multiple fine-spectrum information can more fully and accurately reflect the characteristics inherent in the target and the background itself.
  • Conventional detection methods generally use single-segment imaging detection methods to acquire target/background energy over a wide spectral range. At this time, the target is often submerged in or disturbed by complex background clutter, and the signal is very weak, signal-to-noise ratio, The signal to noise ratio is very low.
  • the map joint device combines the target infrared spectrum with the infrared imaging information, and utilizes the unique spectral features of the target on the infrared spectral line to greatly improve the detectability of the target.
  • This technology is widely used in the field of remote sensing. It provides a data foundation for studying the spectral characteristics of various target backgrounds, and then classifying, monitoring and detecting targets.
  • imaging spectrometers are typically mounted on an aircraft that scans the mirror for rotation so that the received instantaneous field of view is moved perpendicular to the direction of flight for scanning. Due to the forward motion of the aircraft, the imaging spectrometer completes the two-dimensional scanning, and the ground scene is scanned point by point and measured point by point to obtain multi-spectral remote sensing image information.
  • the AVIPIS system completed by the JPL laboratory in the United States and the GERIS system of the US GER company and the PHI push-scan imaging spectrometer system studied by the Shanghai Institute of Technical Physics of the Chinese Academy of Sciences.
  • Such systems can provide rich two-dimensional spatial information and third-dimensional spectral data, that is, each point of two-dimensional spatial imaging can extract spectra information.
  • the device extracts the image cube data of the target scene, the information processing amount is very large, the spatial resolution is low, and the measurement is suitable for the stationary target, which is difficult for the moving target to be coped with, and is expensive, which is difficult for the general user to accept.
  • the spectrum of stationary features and sky background does not need to be acquired in real time, but it is necessary to use the spectral characteristics to automatically detect real-time detection of moving targets (time-varying objects) or local regions in the scene, such as the air.
  • moving targets time-varying objects
  • local regions in the scene such as the air. Aircraft, ships at sea, vehicles in motion, fires, explosions, etc.
  • the “Huazhong University of Science and Technology applied for and authorized “a method and device for acquiring time-varying object spectral information” (ZL200910272679.9) and “a multi-band moving target spectral feature detection and identification method and The device (ZL201110430969.9) adopts two lens combinations to realize the combination of the maps, and the device is bulky; the planar infrared window is adopted, and the field of view is small; the measurement method is for a single moving target, and there is no intelligent automatic control strategy to guide the remote sensing measurement. process.
  • the currently used spectral imaging devices have the following disadvantages: (1) not suitable for spectral measurement of local regions in the scene; (2) unable to achieve multiple moving targets Automatic tracking of the spectrum; (3) online processing and identification of the target spectrum; (4) large amount of data, slow speed and high price.
  • the present invention provides a multi-band common optical path map combined remote sensing measurement system and method, the purpose of which is to achieve a joint measurement of a multi-band common optical axis of a moving target or a local region, Therefore, the technical problems of detecting and identifying weak targets and moving targets are solved.
  • the invention provides a multi-band common light path map combined remote sensing measurement system, which is wrapped by a protective shell, including an infrared window, a two-dimensional rotating mirror, a plane mirror, a reflective multi-band infrared lens, and a Fourier interference spectrum module.
  • the infrared window is coupled to the housing for transmitting infrared light in the target scene;
  • the two-dimensional rotating mirror is fixedly mounted inside the casing, the center of the two-dimensional rotating mirror coincides with the central axis of the field of view of the infrared window, and the mirror surface of the two-dimensional rotating mirror and the field of view of the infrared window
  • the central axis is placed at 45° for tracking moving targets in the target scene, realizing local regions and multi-target sampling;
  • the center of the planar mirror is at the same level as the center of the two-dimensional rotating mirror, the planar mirror
  • the mirror surface is parallel to the mirror surface of the two-dimensional mirror; the central axis of the reflective multi-band infrared lens coincides with the center of the plane mirror for multi-band infrared light focusing, and achieves imaging and spectral measurement.
  • An optical path input end of the Fourier interference spectrum module is connected to the reflective multi-band infrared lens, and an electrical signal output end of the Fourier interference spectrum module is connected to the map joint processing module for Obtaining infrared spectrum data;
  • the map joint processing module is connected to the reflective multi-band infrared lens for realizing joint processing of the map;
  • the power module, the cooling module, and the display module The blocks are respectively connected to the map joint processing module.
  • the two-dimensional rotating mirror comprises a plane mirror, a two-dimensional turntable and a servo motor; the plane mirror is placed on the two-dimensional turntable and fixed by a card slot on the turntable; two of the servo motors
  • the drive shafts are mechanically coupled to the two-dimensional turret pitch axis and the rotary shaft, respectively.
  • the reflective multi-band infrared lens comprises a mirror body, a beam splitter, an infrared detector and an infrared fiber coupler; a mirror center of the beam splitter coincides with a central axis of the mirror body, and the mirror surface of the beam splitter The central axis is at 45°; the infrared detector is mounted at the tail, the center of the infrared detector is in line with the central axis; and the infrared fiber coupler is disposed at the trailing edge.
  • the map joint processing module includes: a communication level conversion unit, an image processing unit, a system control unit, a spectrum processing unit, and a data fusion master control unit; the communication level conversion unit and the reflective type are connected through an RS422 interface
  • the infrared detector connection in the band infrared lens is used to implement level conversion between modules; one end of the image processing unit is connected to the infrared detector through an image interface, and the other end of the image processing unit is
  • the data fusion control unit is connected to perform image enhancement, feature extraction, and segmentation processing on the acquired infrared image to implement system tracking; the input end of the system control unit is connected to the communication level conversion unit, and the system a control end of the control unit is connected to the data fusion master control unit, and the system controls
  • An output end of the unit is connected to the two-dimensional rotating mirror; for performing motion control on the two-dimensional rotating mirror through a serial port; an input end of the spectral processing unit and an electrical signal output end of the Fourier interference spectrum module Connecting
  • the invention also provides a measurement method based on the above multi-band common optical path map combined with a remote sensing measurement system, comprising the following steps:
  • the region of interest refers to a number of closely connected images Pixels, extracting the region of interest specifically includes:
  • the feature description matrix of the region of interest comprises a location feature, a shape feature, a gray feature, and a spectral feature; and the location feature includes a collection point coordinate
  • the boundary chain (c 2 , c 2 , ..., c n ) and the acquisition distance L
  • the shape feature includes area of interest area A and perimeter
  • the area A M, the circumference P is equal to the length of the region of interest RGN V ;
  • the grayscale feature includes an average grayscale a gray gradient matrix and an average gradient; wherein f(x t , y t ) represents a gray value of the image at the pixel point (x t , y t );
  • the spectral features include a spectral curve change trend, peak and trough positions, a number of peaks greater than the mean, a number of peaks less than the mean, a number of troughs greater than the mean, a number of troughs less than the mean, a peak width, and a peak height.
  • the above technical solution conceived by the present invention can increase the system measurement field of view and reduce the measurement dead angle due to the spherical structure of the infrared window; the two-dimensional rotating mirror adopts a servo motor, which can realize Flexible field-of-view switching and fast target tracking; reflective multi-band infrared mirrors enable common optical paths for imaging and spectral measurements; combined image processing with optimal automatic control strategies enables system-to-target scenes to be time-space-spectral Sparse sampling of dimensions, simplifying the amount of data and maximizing the measurement value.
  • FIG. 1 is a schematic structural diagram of a module of a multi-band common light path map combined with a remote sensing measurement system according to an embodiment of the present invention
  • FIG. 2 is a schematic diagram of an optical path structure in a multi-band common optical path map combined with a remote sensing measurement system according to an embodiment of the present invention
  • FIG. 3 is a schematic structural diagram of a two-dimensional mirror in a multi-band common light path map combined with a remote sensing measurement system according to an embodiment of the present invention
  • FIG. 4 is a flowchart of implementing a multi-band common optical path map combined with remote sensing measurement method according to an embodiment of the present invention
  • FIG. 5 is a schematic diagram of full field of view scanning of a multi-band common light path map combined with remote sensing measurement system according to an embodiment of the present invention
  • Figure 6 is an aircraft of the present invention for capturing a flight in a sparse sky background
  • Figure 7 is a view of the present invention for capturing a maneuvering fleet under the constraints of a bridge space
  • Figure 8 is a real-time tracking of airborne aircraft measurement data in the field experiment of the present invention.
  • Figure 8-1 is a long-wave image of a civil aviation aircraft in the air; clockwise, the measurement distance is 1 km, 2.5 km, 5 km, and 8 km, respectively;
  • Figure 8-2 Infrared spectral data of civil aviation aircraft in the air; clockwise, measuring distances are 1km, 2.5km, 5km, 8km;
  • 1-infrared window 2-two-dimensional mirror, 21-plane mirror, 22-two-dimensional turntable, 23-servo motor, 3-plane mirror, 4-reflective multi-band infrared lens, 41-kaset Green structure mirror body, 411-parabolic mirror, 412-hyperbolic mirror, 42-beam splitter, 43-infrared detector, 44-infrared fiber coupler, 5-Fourier interference spectroscopy module, 51-beam splitter , 52-fixed mirror, 53-moving mirror, 54-unit detector, 6-map information fusion module, 61-communication level conversion unit, 62-image processing unit, 63-system control unit, 64-spectral processing unit, 65-Data fusion master control unit, 7-power module, 8-cooling module, 9-display module.
  • 4-reflective multi-band infrared lens 41-kaset Green structure mirror body, 411-parabolic mirror, 412-hyperbolic mirror, 42-beam splitter, 43-infrared detector, 44-in
  • the invention relates to a multi-band common light path map combined with remote sensing measurement system and a combined method for infrared spectrum acquisition based on the system, in particular to an infrared image and spectrum measurement system and method for moving targets.
  • the invention is a smart on-site remote sensing device with portability.
  • the invention provides a multi-band common optical path map combined with remote sensing measurement system, see FIG. 1 and FIG. 2, mainly including infrared window 1, two-dimensional rotating mirror 2, plane mirror 3, reflective multi-band infrared lens 4, and Fuli
  • the system is mainly divided into a processor entity and an optical measuring entity in physical structure.
  • the processor entity is composed of the map joint processing module 6, and the other components constitute an optical measurement entity.
  • the former mainly realizes all the information processing work of the system, and the latter implements the optical measurement work of the whole system.
  • the processing entity uses the Ethernet Ethernet interface to connect and control the Fourier interference spectrum module 5, and obtains infrared spectral data; uses the serial bus to realize the motion control of the two-dimensional rotating mirror 2; adopts the RS422 interface and the image interface and the reflective multi-band The connection of the infrared lens 4.
  • the RS422 interface realizes level conversion, controls lens focusing and non-uniformity correction, and the image interface has analog PAL system and digital Camera LINK interface to obtain infrared image information.
  • the optical measuring body is wrapped in a specially designed protective casing and the housing is made of aluminium.
  • the infrared window 1 and the housing are connected by rivets;
  • the two-dimensional rotating mirror 2 is fixedly mounted inside the casing, the center coincides with the central axis of the field of view of the infrared window 1, the mirror surface is placed at 45° with the central axis;
  • the plane mirror 3 center and two The center of the mirror 2 is at the same level, the mirror is placed parallel to the mirror of the 2D mirror 2, and is fixed inside the casing;
  • the reflective multi-band infrared lens 4 is designed without a heating, the central axis of the lens and the plane mirror 3 center coincides, the mirror body is reinforced and fixed inside the casing;
  • the Fourier interference spectroscopy module 5 is connected to the reflective multi-band infrared lens 4 through the infrared fiber;
  • the power module 7 is placed in the inner space of the casing to realize the system power supply and provide a universal power interface.
  • the refrigeration module 8 is placed inside the casing by a grid structure to realize temperature control of the casing, and a desiccant filling port is provided at the same time, so that it is convenient to add colorless silica gel at any time;
  • the display module 9 adopts a touch-type liquid crystal display and is embedded in the shell. Body side.
  • the infrared window 1 is used to transmit infrared light in the target scene to filter other stray light, and the infrared window 1 also protects the inside and outside of the system from being protected. In practical applications, the infrared window 1 is connected to the housing of the multi-band common light path map combined with the remote sensing measurement system through a fixed structure.
  • the infrared window 1 material can be a multi-spectral CVD ZnS material.
  • the multi-spectral ZnS material under CVD has high transmittance in several bands of near-infrared, short-wave infrared, medium-wave infrared and long-wave infrared.
  • the infrared window 1 can adopt a spherical structure when the window size is constant, which can increase the visual field of view of the system window and reduce the system detection dead angle.
  • the two-dimensional rotating mirror 2 provided by the embodiment of the present invention includes a plane mirror 21, a two-dimensional turntable 22, and a servo motor 23.
  • the plane mirror 21 is placed on the two-dimensional turntable 22 and fixed by a card slot on the turntable; the two drive shafts of the servo motor 23 are mechanically coupled to the pitch axis and the rotary shaft of the two-dimensional turntable 22, respectively.
  • the plane mirror 21 can adopt K9 glass, and the gold-plated reflective layer has high reflectivity for short, medium and long-wave infrared light; the two-dimensional turntable 22 can drive the plane mirror 21 to rotate and pitch under the driving of the servo motor 23. Two dimensions of motion, thereby increasing the flexibility of system detection and achieving localized sparse sampling.
  • the reflective multi-band infrared lens 4 includes: a Cassegrain mirror 41, a beam splitter 42, an infrared detector 43, and an infrared fiber coupler 44; wherein the Cassegrain mirror 41 is a main structure of a reflective multi-band infrared lens.
  • the central axis coincides with the central axis of the lens;
  • the center of the mirror of the beam splitter 42 coincides with the central axis of the Cassegrain mirror 41, and the mirror surface is placed at 45° with the central axis;
  • the infrared detector 43 is mounted at the tail of the lens, and the center is formed with the central axis of the lens.
  • Straight line; the infrared fiber coupler can be mounted at the edge of the rear of the lens.
  • Cassegrain mirror 41 consisting of a parabolic mirror 411 and a hyperbolic mirror 412 Composition and correction of aberrations by several lenses.
  • the reflective multi-band infrared lens 4 After the light enters the reflective multi-band infrared lens 4, it first enters the Cassegrain mirror 41 to focus, and then splits through the beam splitter 42. The transmitted portion is concentrated on the infrared detector 43 for imaging, and the reflected portion is concentrated on the infrared fiber coupler 44, and The fiber is transmitted to the Fourier interference spectroscopy module 5.
  • the reflective multi-band infrared lens 4 can realize long-wave infrared imaging and short, medium and long-wave infrared spectrum common light paths, and the two fields of view coincide.
  • the lens system has less energy loss and light overall weight.
  • a non-heating design is carried out.
  • the beam splitter 42 is coated with a double-layer antireflection film, so that it has high reflectivity for short and medium wave infrared light, and has a transflective function for long-wave infrared light. This design has the advantages of simple structure and no moving parts.
  • the Fourier interference spectroscopy module 5 is configured to perform interference sampling on the incident light and obtain the infrared spectrum by Fourier transform; in the embodiment of the invention, the spectral detection unit EM27 of Bruker Optics or the process control spectroscopy can be used.
  • System IRCube OEM both using Michelson interferometer system, spectral resolution 2cm -1 , 4cm -1 , 8cm -1 , 16cm -1 , 32cm -1 optional , measuring spectral range including short , medium and long wave , using Sterling or liquid nitrogen cooled MCT detectors.
  • the map joint processing module 6 includes a communication level conversion unit 61, an image processing unit 62, a system control unit 63, a spectrum processing unit 64, and a data fusion master unit 65.
  • the communication level conversion unit 61 is interconnected with the outside world through the RS422 interface, and mainly realizes the level conversion function between the modules;
  • the image processing unit 62 mainly performs image enhancement, feature extraction, segmentation, etc. on the collected infrared images to implement system tracking.
  • the system control unit 63 performs motion control on the high-speed servo drive motor 23 through the serial port;
  • the spectrum processing unit 64 mainly performs acquisition control, data pre-processing, and radiation calibration on the infrared spectrum collected by the Fourier interference spectrum module 5 through the Ethernet interface.
  • the processing results of the three modules are finally fed back to the data fusion master control unit 65 for data fusion processing, and finally the optimal system control strategy is selected for data acquisition.
  • the data fusion master control unit 65 also interacts with the power module 7, the cooling module 8, and the display module 9, thereby implementing functions such as configuration parameterization and data visualization.
  • the working principle of the invention is that the light of the target scene is incident from the infrared window 1, is specularly reflected by the plane mirror 21 of the two-dimensional mirror 2, and is again reflected by the plane mirror 3 into the reflective multi-band infrared lens 4, in the lens.
  • the Cassegrain mirror 41 is focused and split by the beam splitter 42.
  • the map information fusion module 6 receives, processes, and stores the multi-dimensional information of the probe object through each sub-unit, and controls the rotation of the two-dimensional mirror 2 to change the optical axis of the system, thereby achieving target tracking and recognition.
  • the display module 9 uses the data acquired by the map information fusion module to display the map information acquired by the current system in the form of a graphical interface, and implements a simple human-computer interaction function.
  • the invention proposes a measurement method based on the above multi-band common light path map combined with remote sensing measurement system, see FIG. 4, comprising the following steps:
  • the multi-band common optical path map combined with remote sensing measurement method proposed by the present invention refers to a plurality of pixel points closely connected in the image in the step (2), and the extraction method is determined as follows:
  • image segmentation processing may be performed by using a threshold segmentation method
  • the multi-band common optical path map combined with remote sensing measurement method proposed by the present invention includes a position feature, a shape feature, a gray feature and a spectral feature in the feature region of the region of interest in the step (3).
  • the characterization matrix contains:
  • Grayscale features including average grayscale a gray gradient matrix and an average gradient; wherein f(x t , y t ) represents a gray value of the image at the pixel point (x t , y t );
  • Spectral characteristics include spectral curve change trend, peak (maximum value) and trough (minimum value) position, number of peaks larger than mean, number of peaks smaller than mean, number of troughs larger than mean, number of troughs smaller than mean, Peak width and peak height; using Gaussian scale space theory, the inflection point and curvature of the spectral curve at each Gaussian spatial scale can be calculated, and the variation trend of the spectral curve is extracted by the inflection point and curvature.
  • the multi-band common optical path map combined with remote sensing measurement method proposed by the present invention uses the feature weight matrix in the step (4) to describe the contribution degree of the feature in the matching association and the information entropy.
  • the design principles of the feature weight matrix are as follows:
  • the matching association can be realized by using the correlation distance matching calculation method according to the feature description matrix and the feature weight matrix.
  • the position feature adopts minimum distance matching
  • the shape and gray feature adopt SAD (Sum of Absolute Difference) method
  • the spectral feature adopts SAM (spectral angle mapper) method.
  • the feature information entropy in the step (5) of the multi-band common light path map combined with the remote sensing measurement method proposed by the present invention is mainly used to describe the total amount of feature information included in the region of interest, and is a criterion for the system to determine the value of the region of interest collection.
  • the multi-band common optical path map combined with remote sensing measurement system and method of the invention not only effectively realizes joint acquisition of image and spectral information, but also realizes intelligent target capture under the maximum information entropy automatic control strategy through the above steps, and finally achieves the goal. Multidimensional information measurement.
  • the invention utilizes long-wave infrared imaging to automatically detect, capture and track the region of interest, and automatically aligns the optical axis of the system to track the region of interest to obtain spatial shape, motion and spectral multi-dimensional information, which satisfactorily solves the spectral measurement of the local region. .
  • the aircraft target is more prominent, and the aircraft engine nozzle is the most obvious position due to the highest temperature, and the spectral information is also the most obvious, and the target information entropy is the largest.
  • the automatic entropy optimization automatic control strategy of the present invention performs spatial sparse sampling of the aircraft in the sky-capture, and then performs real-time spectral measurement on the captured aircraft. It can be clearly seen from the figure that the optical axis center is aligned with the engine nozzle position. .
  • the invention performs automatic target detection and tracking by acquiring infrared images and using continuously measured infrared spectra. Under the automatic control strategy of feature information entropy optimization, the measurement ordering of multiple targets is realized, thereby realizing multi-objective spectral measurement.
  • the bridge landmark constraint area determines the spatial extent of the detection of the motor vehicle fleet.
  • Figure 8-1 shows the infrared results of continuous tracking measurements of moving targets over successive measurement cycles
  • Figure 8-2 shows the results of infrared spectroscopy measurements for the corresponding cases. From the experimental data, it can be seen that the spectral characteristic intensity of the aircraft tail flame becomes weaker with the increase of the detection distance, while the main spectral trend characteristics remain basically unchanged, providing a data basis for the subsequent identification of targets from the background. It can be seen that as the distance of the aircraft is getting farther and farther, the spectral intensity of the aircraft is gradually weakened, but the main peaks, troughs and curves of the spectrum are basically consistent, which provides a solid data foundation for subsequent target recognition.
  • the above illustrations illustrate the superiority and pre-existence of the joint measurement system and method under the automatic control strategy of information entropy optimization.

Landscapes

  • Physics & Mathematics (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Life Sciences & Earth Sciences (AREA)
  • Geophysics (AREA)
  • Spectrometry And Color Measurement (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)

Abstract

一种多波段共光路图谱联合遥感测量系统及方法，该系统包括红外窗口（1）、二维转镜（2）、平面反射镜（3）、反射式多波段红外镜头（4）、傅里叶干涉光谱模块（5）、图谱联合处理模块（6）、电源模块（7）、制冷模块（8）和显示模块（9）；入射光从红外窗口（1）进入，经二维转镜（2）反射后再由平面反射镜（3）反射至反射式多波段红外镜头（4）并由分光镜（42）分光；透射的光线经过会聚透镜聚焦，在红外探测器（43）上成像；反射的光线聚焦于红外光纤耦合器（44），由红外光纤进入傅里叶干涉光谱模块（5）形成干涉图，并经傅里叶变换得到光谱数据；图谱联合处理模块（6）有效联合宽谱成像与非成像光谱数据，控制二维转镜（2）指向目标，实现智能化的遥感测量，具有对场景局部区域光谱测量以及多目标跟踪光谱测量的能力，速度快，数据量适宜，成本低。

Description

一种多波段共光路图谱联合遥感测量系统及方法 【技术领域】
本发明属于遥感测量技术领域，更具体地，涉及一种多波段共光路图谱联合遥感测量系统及方法。
【背景技术】
按照电磁波谱的结构和特性，目标红外信息是指红外谱段传感器获取的目标与背景辐射、反射和散射特性的差异，它包括短、中、长波红外各精细光谱(线)波段的辐射、反射和散射特性。与单谱段的目标背景信息相比，多个精细谱段信息更能全面而准确地反映目标与背景本身固有的特性。常规探测手段一般使用单谱段成像探测方法，在较宽的谱段上采集目标/背景能量，此时目标往往淹没于复杂的背景杂波之中或被干扰，信号非常微弱，信噪比、信杂比很低。
图谱联合设备把目标红外光谱和红外成像信息结合起来，利用目标在红外光谱谱线上的独特的光谱特征，可以大大提高目标的可探测性。该技术广泛应用于遥感领域，为研究各种目标背景的光谱特性，进而对场景进行分类、监视与目标探测识别提供数据基础。
国内外都非常重视研发此类图谱联合设备。目前，常用的图谱联合设备为成像光谱仪。此类成像光谱仪一般安装在飞行器上，其扫描转镜旋转，使接收的瞬时视场作垂直于飞行方向的运动，从而实现扫描。由于飞行器的前向运动，成像光谱仪即完成二维扫描，地物景象被逐点扫过，并逐点分波段测量，从而获得多光谱的遥感图像信息。例如美国JPL实验室完成的AVIPIS系统和美国GER公司GERIS系统以及国内中科院上海技术物理研究所研究的PHI推扫式成像光谱仪系统。此类系统能够提供丰富的二维空间信息及第三维的光谱数据，即二维空间成像的每一点都可以提取光谱 信息。但是这种设备提取目标场景的图像立方体数据，信息处理量非常大，空间分辨率较低，适用于静止目标的测量，对于运动目标难以应付，且价格昂贵，一般用户难以接受。
在许多实际应用中，静止的地物和天空背景的光谱并不需要实时获取，而需要利用光谱特性对场景中运动目标(时变对象)或局部区域进行自动实时地检测识别探测识别，如空中的飞机、海上的船舶、行驶中的车辆、火灾、爆炸等。
在相关专利中，国内的华中科技大学申请并授权的“一种图谱一体化的时变对象光谱信息获取方法与装置”(ZL200910272679.9)和“一种多波段动目标光谱特征探测识别方法和装置”(ZL201110430969.9)采用两个镜头组合实现图谱联合的方式，设备体积大；采用平面红外窗口，视场小；测量方法针对单个运动目标，没有一种智能化的自动控制策略指导遥感测量过程。
对于上述多个运动目标和时变对象的自动检测与光谱识别，目前常用的光谱成像设备存在以下缺点：(1)不适于场景中局部区域的光谱测量；(2)不能实现多个运动目标的自动跟踪测谱；(3)不能进行目标光谱的在线处理与识别；(4)数据量大、速度慢且价格昂贵。
【发明内容】
针对现有技术的以上缺陷或改进需求，本发明提供了一种多波段共光路图谱联合遥感测量系统及方法，其目的在于实现对动目标或者局部区域的多波段共光轴的图谱联合测量，由此解决弱小目标、动目标的探测与识别的技术问题。
本发明提供了一种多波段共光路图谱联合遥感测量系统，由防护性的壳体包裹，包括红外窗口、二维转镜、平面反射镜、反射式多波段红外镜头、傅里叶干涉光谱模块、图谱联合处理模块、电源模块、制冷模块和显示模块；所述红外窗口与壳体连接，用于透过目标场景中的红外光线；所 述二维转镜固定安装在所述壳体内部，所述二维转镜的中心与所述红外窗口的视场中心轴重合，所述二维转镜的镜面与所述红外窗口的视场中心轴成45°放置，用于跟踪目标场景中动目标，实现局部区域以及多目标采样；所述平面反射镜的中心与所述二维转镜的中心在同一水平高度，所述平面反射镜的镜面与所述二维转镜的镜面平行；所述反射式多波段红外镜头的中心轴与所述平面反射镜的中心重合，用于多波段的红外光线聚焦，并实现成像与光谱测量共光路；所述傅里叶干涉光谱模块的光信号输入端与所述反射式多波段红外镜头连接，所述傅里叶干涉光谱模块的电信号输出端与所述图谱联合处理模块连接，用于获取红外光谱数据；所述图谱联合处理模块与所述反射式多波段红外镜头连接，用于实现图谱联合处理；所述电源模块、制冷模块和显示模块分别与所述图谱联合处理模块连接。
其中，所述二维转镜包括平面反射镜、二维转台和伺服电机；所述平面反射镜放置在所述二维转台上，并通过转台上的卡槽固定；所述伺服电机的两个驱动轴分别与所述二维转台俯仰轴以及旋转轴机械连接。
其中，所述反射式多波段红外镜头包括镜体、分光镜、红外探测器和红外光纤耦合器；所述分光镜的镜面中心与所述镜体的中心轴重合，所述分光镜的镜面与中心轴成45°；所述红外探测器安装在尾部，所述红外探测器中心与中心轴成一条直线；所述红外光纤耦合器设置在尾部边缘。
其中，所述图谱联合处理模块包括：通讯电平转换单元、图像处理单元、系统控制单元、光谱处理单元和数据融合总控单元；所述通讯电平转换单元通过RS422接口与所述反射式多波段红外镜头中的所述红外探测器连接；用于实现模块间的电平转换；所述图像处理单元的一端通过图像接口与所述红外探测器连接，所述图像处理单元的另一端与所述数据融合总控单元连接，用于对采集的红外图像进行图像增强、特征提取、分割处理，实现系统跟踪；所述系统控制单元的输入端与所述通讯电平转换单元连接，所述系统控制单元的控制端与所述数据融合总控单元连接，所述系统控制 单元的输出端与所述二维转镜连接；用于通过串口对所述二维转镜进行运动控制；所述光谱处理单元的输入端与所述傅里叶干涉光谱模块的电信号输出端连接，所述光谱处理单元的输出端与所述连接；用于对所述傅里叶干涉光谱模块采集的红外光谱进行采集控制、数据预处理、辐射定标、光谱特征提取处理；所述数据融合总控单元用于将所述图像处理单元、所述系统控制单元和所述光谱处理单元的输出进行数据融合处理，并根据处理结果选择最优的系统控制策略进行数据采集。
本发明还提供了一种基于上述的多波段共光路图谱联合遥感测量系统的测量方法，包括下述步骤：
(1)通过红外探测器实现目标场景的全视场扫描采集获得图像序列；
(2)在所述图像序列中提取至少一个感兴趣区；
(3)计算所述图像序列中所有感兴趣区的特征描述矩阵；
(4)判断当前时刻是否处于数据获取初始阶段，若是，则对所有感兴趣区的特征权重矩阵赋予初值；若否，则根据所有感兴趣区的特征描述矩阵将图像序列中所有感兴趣区与上一采集周期中获取的所有感兴趣区进行逐次匹配关联，提取关联成功的各感兴趣区的光谱特征，并将光谱特征添加到对应感兴趣区的特征描述矩阵中，同时将其特征权重矩阵复制到对应的感兴趣区的特征权重矩阵；
(5)计算所有感兴趣区的特征信息熵，并按照特征信息熵的值从大到小将与之对应的感兴趣区进行排序，并赋予所有感兴趣区采集顺序的编号；
(6)根据编号依次对各个感兴趣区中的采集点进行排序形成采集点序列；
(7)依次将光轴点移动到所述采集点序列中的采集点；同时采用非成像的傅里叶干涉光谱模块获取感兴趣区的光谱数据；
(8)重复步骤(1)～(7)进行下一个采集周期，直到系统采集结束。
其中，在所述步骤(2)中，所述感兴趣区是指图像中紧密相连的若干 像素点，提取感兴趣区具体包括：
(21)利用非线性视觉原理进行图像增强处理；
(22)对增强处理后的图像进行图像分割处理；
(23)对分割处理后的图像进行膨胀腐蚀处理并获得所述感兴趣区。
其中，在步骤(3)中，所述感兴趣区的特征描述矩阵包括位置特征、形状特征、灰度特征和光谱特征；所述位置特征包括采集点坐标
Figure PCTCN2014085762-appb-000001
边界链(c2，c2，......，cn)和采集距离L＝|x0-x0|+|y0-y0|；感兴趣区RGNV包含M个像素点(x1，y1)，(x2，y2)，......，(xM，yM)，n为感兴趣区边界像素点的个数，ct＝（xt，yt)为感兴趣区边界像素点；(x0，y0)为当前光轴点，其位置随着二维转镜的移动而不断变化；
所述形状特征包括感兴趣区面积A和周长
Figure PCTCN2014085762-appb-000002
面积A＝M，周长P等于感兴趣区RGNV的长度；
所述灰度特征包括平均灰度
Figure PCTCN2014085762-appb-000003
灰度梯度矩阵以及平均梯度；其中f(xt，yt)表示图像在像素点(xt，yt)处的灰度值；
所述光谱特征包括光谱曲线变化趋势、波峰和波谷位置、大于均值的波峰数、小于均值的波峰数、大于均值的波谷数、小于均值的波谷数、峰宽、峰高。
总体而言，通过本发明所构思的以上技术方案与现有技术相比，由于红外窗口采用球面结构，能够增大系统测量视场，减小测量死角；二维转镜采用伺服电机，能够实现灵活的视场切换和快速的目标跟踪；反射式多波段红外镜能实现成像和光谱测量的共光路；图谱联合处理采用最优自动控制策略，能够实现系统对目标场景在时间-空间-光谱多维度稀疏采样，简化数据量，最大程度实现测量价值。
【附图说明】
图1为本发明实施例提供的多波段共光路图谱联合遥感测量系统的模块结构示意图；
图2为本发明实施例提供的多波段共光路图谱联合遥感测量系统中的光路结构示意图；
图3为本发明实施例提供的多波段共光路图谱联合遥感测量系统中二维转镜的结构示意图；
图4为本发明实施例提供的多波段共光路图谱联合遥感测量方法的实现流程图；
图5为本发明实施例提供的多波段共光路图谱联合遥感测量系统的全视场扫描示意图；
图6为本发明在稀疏天空背景下捕获飞行的飞机；
图7为本发明在桥梁空间约束下捕获机动的车队；
图8本发明外场实验中实时跟踪空中民航飞机测量数据结果；其中图8-1为空中的民航飞机的长波图像；顺时针，测量距离依次为1km、2.5km、5km、8km；图8-2为空中的民航飞机的红外光谱数据；顺时针，测量距离依次为1km、2.5km、5km、8km；
其中，1-红外窗口，2-二维转镜，21-平面反射镜，22-二维转台，23-伺服电机，3-平面反射镜，4-反射式多波段红外镜头，41-卡塞格林结构镜体，411-抛物面反射镜，412-双曲面反射镜，42-分光镜，43-红外探测器，44-红外光纤耦合器，5-傅里叶干涉光谱模块，51-分束器，52-定镜，53-动镜，54-单元探测器，6-图谱信息融合模块，61-通讯电平转换单元，62-图像处理单元，63-系统控制单元，64-光谱处理单元，65-数据融合总控单元，7-电源模块，8-制冷模块，9-显示模块。
【具体实施方式】
为了使本发明的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本发明进行进一步详细说明。应当理解，此处所描述的具体 实施例仅用以解释本发明，并不用于限定本发明。此外，下面所描述的本发明各个实施方式中所涉及到的技术特征只要彼此之间未构成冲突就可以相互组合。
本发明涉及一种多波段共光路图谱联合遥感测量系统及基于该系统实现的红外图谱联合采集方法，尤其涉及对动目标的红外图像和光谱测量系统及方法。本发明是灵巧型的现场遥感设备，具有便携性。
本发明提供了一种多波段共光路图谱联合遥感测量系统，参见图1和图2，主要包括红外窗口1，二维转镜2，平面反射镜3，反射式多波段红外镜头4，傅里叶干涉光谱模块5，图谱联合处理模块6，电源模块7，制冷模块8和显示模块9；该系统在物理结构上主要分为处理机实体和光学测量实体。其中，处理机实体由图谱联合处理模块6组成，其他部分则组成光学测量实体。前者主要实现系统所有的信息处理工作，后者则是实现整个系统光学测量工作。
处理机实体利用以太网Ethernet接口对傅里叶干涉光谱模块5进行连接控制，并获取红外光谱数据；采用串口总线实现对二维转镜2运动控制；采用RS422接口和图像接口与反射式多波段红外镜头4的连接。RS422接口实现电平转换，控制镜头调焦和非均匀性校正，图像接口有模拟PAL制式和数字Camera LINK接口，实现获取红外图像信息。
光学测量实体由专门设计的防护性外壳包裹，壳体采用铝合金材质。红外窗口1与壳体采用铆钉连接；二维转镜2固定安装在壳体内部，中心与红外窗口1的视场中心轴重合，镜面与中心轴成45°放置；平面反射镜3中心与二维转镜2中心在同一水平高度，镜面与二维转镜2镜面平行放置，并固定在壳体内部；反射式多波段红外镜头4采用无热化封闭式设计，镜头中心轴与平面反射镜3中心重合，镜体加强固定在壳体内部；傅里叶干涉光谱模块5通过红外光纤连接反射式多波段红外镜头4；壳体内部空间内放置电源模块7实现系统供电并提供通用电源接口方便外接电源的使用； 制冷模块8采用栅格结构放置在壳体内部，实现壳体的温度控制，同时设置干燥剂填充口，方便随时补充添加无色硅胶等；显示模块9采用触控式液晶显示屏，镶嵌在壳体侧面。
红外窗口1用于透过目标场景中的红外光线，过滤其他杂散光，同时红外窗口1也实现系统内部与外面隔绝，起到保护作用。在实际应用中，红外窗口1通过固定结构与多波段共光路图谱联合遥感测量系统的壳体相连接。
红外窗口1材料可以采用多光谱CVD ZnS材料。CVD工艺下的多光谱ZnS材料在近红外、短波红外、中波红外及长波红外几个波段都具有很高的透过率。
作为本发明的一个实施例，红外窗口1在窗口大小尺寸一定的情况下，可以采用球面结构，这样能增大系统窗口的可视视场，减小系统探测死角。
如图3所示，本发明实施例提供的二维转镜2包括平面反射镜21、二维转台22和伺服电机23。平面反射镜21放置在二维转台22上，并通过转台上的卡槽固定；伺服电机23的两个驱动轴分别与二维转台22俯仰轴以及旋转轴机械连接。平面反射镜21可以采用K9玻璃，镀金反射层后对短、中、长波红外光都有较高的反射率；二维转台22在伺服电机23的驱动下可以带动平面反射镜21实现旋转和俯仰两个维度的运动，从而增大系统探测的灵活性，实现局域稀疏采样。
反射式多波段红外镜头4包括：卡塞格林镜体41、分光镜42、红外探测器43和红外光纤耦合器44；其中，卡塞格林镜体41是反射式多波段红外镜头的主体结构，其中心轴与镜头中心轴重合；分光镜42镜面中心与卡塞格林镜体41中心轴重合，镜面与中心轴成45°放置；红外探测器43安装在镜头尾部，中心与镜头中心轴成一条直线；红外光纤耦合器在安装在镜头尾部边缘，位置合适即可。
卡塞格林镜体41，由一个抛物面反射镜411和一个双曲面反射镜412 组成，并由若干透镜校正像差。
光线进入反射式多波段红外镜头4后，首先进入卡塞格林镜体41聚焦，然后经分光镜42分光，透射部分会聚于红外探测器43成像，反射部分会聚于红外光纤耦合器44，并经光纤传输至傅里叶干涉光谱模块5。
反射式多波段红外镜头4可实现长波红外成像和短、中、长波红外测谱共光路，两个视场中心重合。该镜头系统能量损失少，整体重量轻，为降低镜头本身辐射对探测造成的影响，进行了无热化设计。其中，分光镜42镀双层增透膜，使其对短、中波红外光反射率高，对长波红外光则具有半透半反功能,这种设计具有结构简单、无需运动部件的优点。
傅里叶干涉光谱模块5用于将入射光进行干涉采样，并通过傅里叶变换获取红外光谱；本发明实施例中可以采用德国布鲁克(Bruker Optics)公司的光谱探测单元EM27或者过程控制光谱测量系统IRCube OEM，两者均采用迈克尔逊干涉仪体制，光谱分辨率2cm-1、4cm-1、8cm-1、16cm-1、32cm-1可选，测量光谱范围包括短、中、长波,采用斯特林或液氮制冷的MCT探测器。
图谱联合处理模块6包括通讯电平转换单元61、图像处理单元62、系统控制单元63、光谱处理单元64和数据融合总控单元65。其中，通讯电平转换单元61通过RS422接口与外界互联，主要实现模块间的电平转换功能；图像处理单元62主要对采集的红外图像进行图像增强、特征提取、分割等处理等，实现系统跟踪；系统控制单元63通过串口对高速伺服驱动电机23进行运动控制；光谱处理单元64则主要通过Ethernet接口，对傅里叶干涉光谱模块5采集的红外光谱进行采集控制、数据预处理、辐射定标、光谱特征提取处理等。这三个模块的处理结果最终均反馈给数据融合总控单元65，进行数据融合处理，最终选择最优的系统控制策略进行数据采集。同时，数据融合总控单元65也与电源模块7、制冷模块8、显示模块9进行交互，实现配置参数化、数据可视化等功能。
本发明的工作原理为：目标场景的光线从红外窗口1入射，经二维转镜2的平面反射镜21镜面反射，由平面反射镜3再次反射进入反射式多波段红外镜头4，在镜头内的卡塞格林镜体41聚焦后经分光镜42分光，入射光中50％的长波红外光透过分光镜42到达红外探测器成像43；入射光中的短、中波红外光和其余50％长波红外光经过分光镜反射后，经由红外光纤耦合器44利用光纤传输至傅里叶干涉光谱模块5形成干涉图，并最终获取红外光谱。图谱信息融合模块6通过各个分单元接收、处理、存储探测对象的图谱多维信息，并控制二维转镜2的旋转改变系统光轴指向，进而实现目标的跟踪、识别。显示模块9利用图谱信息融合模块获取的数据，以图形化界面的形式展示当前系统获取的图谱信息，并实现简单的人机交互功能。
本发明提出了一种基于上述多波段共光路图谱联合遥感测量系统的测量方法，参见图4，包括以下步骤：
(1)通过红外探测器实现目标场景的全视场扫描(如图5)采集；
(2)在全视场扫描图像序列内提取至少一个感兴趣区；
(3)计算当图像序列中所有感兴趣区的特征描述矩阵；
(4)判断当前时刻是否处于数据获取初始阶段，若是，则对所有感兴趣区的特征权重矩阵赋予初值；若否，则根据所有感兴趣区的特征描述矩阵将图像序列所有感兴趣区与上一采集周期中获取的所有感兴趣区进行逐次匹配关联，提取关联成功的各感兴趣区的光谱特征，并将光谱特征添加到对应感兴趣区的特征描述矩阵中，同时将其特征权重矩阵复制到对应的感兴趣区的特征权重矩阵；
(5)计算所有感兴趣区的特征信息熵，并按照特征信息熵的值从大到小将与之对应的感兴趣区进行排序，并赋予所有感兴趣区采集顺序的编号；
(6)根据编号依次对各个感兴趣区中的采集点进行排序形成采集点序列；
(7)依次将光轴点移动到所述采集点序列中的采集点；同时采用非成像的傅里叶干涉光谱模块获取感兴趣区的光谱数据；
(8)重复步骤(1)～(7)开始下一个采集周期，直到系统采集结束。
本发明提出的多波段共光路图谱联合遥感测量方法所述步骤(2)中感兴趣区是指图像中紧密相连的若干像素点，其提取方法按照如下方式确定：
(21)利用非线性视觉原理进行图像增强处理；其中图像增强处理的目的是突出潜在目标以及其所在的区域；
(22)对增强处理后的图像进行图像分割处理；可以采用阈值分割法对其进行图像分割处理；
(23)对分割处理后的图像进行膨胀腐蚀处理，得到感兴趣区；其中膨胀腐蚀处理是合并单个的残余像素点，消除干扰因素。
本发明提出的多波段共光路图谱联合遥感测量方法所述步骤(3)中的感兴趣区的特征描述矩阵包括位置特征、形状特征、灰度特征和光谱特征。
设感兴趣区RGNV包含M个像素点(x1，y1)，(x2，y2)，......，(xM，yM)，则特征描述矩阵包含：
(31)位置特征包括采集点坐标
Figure PCTCN2014085762-appb-000004
边界链(c1，c2，......，cn)，其中n为感兴趣区边界像素点的个数，ct＝(xt，yt)为感兴趣区边界像素点；采集距离L＝|x0-x0|+|y0-y0|,其中(x0，y0)为当前光轴点，其位置随着二维转镜的移动而不断变化；
(32)形状特征包括感兴趣区面积A＝M；周长
Figure PCTCN2014085762-appb-000005
即感兴趣区RGNV的长度，计算周长的过程中，两个4-邻域像素点之间的距离定义为1，两个8-邻域像素点之间的距离定义为1.414；形状数
Figure PCTCN2014085762-appb-000006
(33)灰度特征包括平均灰度
Figure PCTCN2014085762-appb-000007
灰度梯度矩阵以及平均梯 度；其中f(xt，yt)表示图像在像素点(xt，yt)处的灰度值；
(34)光谱特征包括光谱曲线变化趋势、波峰(极大值)和波谷(极小值)位置、大于均值的波峰数、小于均值的波峰数、大于均值的波谷数、小于均值的波谷数、峰宽、峰高；利用高斯尺度空间理论，可以计算出光谱曲线在各个高斯空间尺度下的拐点和曲率，通过拐点和曲率提取光谱曲线变化趋势。
本发明提出的多波段共光路图谱联合遥感测量方法所述步骤(4)中的特征权重矩阵用于描述特征在匹配关联以及信息熵中的贡献程度。特征权重矩阵的设计原则如下：
(41)处在光轴中心指向范围的边界附近，包含正向边界外运动目标的感兴趣区,权重等级Level＝5；
(42)处于光轴指向范围的内部，包含正向边界移动目标的感兴趣区，权重等级Level＝4；
(43)包含向光轴指向范围内侧移动目标的感兴趣区，权重等级Level＝3；
(44)包含静止目标的感兴趣区，权重等级Level＝2；
(45)不包含目标的感兴趣区，权重等级Level＝1；
(46)同一权重等级的感兴趣区，依据其特征矩阵中的各个分特征贡献权重分量。
根据特征描述矩阵和特征权重矩阵利用相关距离匹配计算方法即可实现匹配关联。其中，位置特征采用最小距离匹配；形状与灰度特征采用SAD(Sum of Absolute Difference)方法；光谱特征采用SAM(spectral angle mapper)方法。
本发明提出的多波段共光路图谱联合遥感测量方法所述步骤(5)中的特征信息熵主要用于描述该感兴趣区所包含特征信息的总量，是系统确定兴趣区采集价值的标准。
假设当前采集周期内有N个感兴趣区RGNV依次为1，2，3，……，n，其特征描述矩阵为T，特征权重矩阵为W，特征信息熵(S)定义为：
Figure PCTCN2014085762-appb-000008
本发明的多波段共光路图谱联合遥感测量系统及方法，不仅有效实现了图像和光谱信息的联合采集，而且通过上述步骤实现了最大信息熵自动控制策略下的智能化目标捕获，并最终实现目标的多维信息测量。
某些目标具有空间稀疏性，他们仅占场景中的有限空间区域。本发明利用长波红外成像自动检测、捕获、跟踪感兴趣区，将系统光轴自动对准跟踪感兴趣区后获取空间形态、运动和光谱多维度信息，很好的解决了局域区域的光谱测量。
如图6，天空稀疏背景下，飞机目标比较突出，而飞机发动机喷口由于温度最高，相对飞机整个感兴趣区也是最为明显的位置，光谱信息也最为强烈明显，目标信息熵最大。本发明信息熵优化的自动控制策略下对天空中的飞机进行空间的稀疏采样——捕获，而后对捕获的飞机进行实时光谱测量，从图中可以清楚看到光轴中心对准了发动机喷口位置。
复杂背景下的多目标，在有限的受约束的空间中机动。本发明通过获取红外图像并利用连续测量的红外光谱进行自动目标检测、跟踪。在特征信息熵优化的自动控制策略下，实现对多目标的测量排序，从而实现多目标的光谱测量。
如图7，行驶在桥梁上的车队为例，在空间范围内，桥梁地标约束区域决定了对机动车队检测的空间范围。
图8-1则是在连续多个测量周期内对运动目标进行持续跟踪测量的红外结果，而图8-2是上述对应情况下的红外光谱测量结果。从实验数据结果可以看出飞机尾焰光谱特征强度随着探测距离的加大而变弱，而主要光谱趋势特征却依然保持基本不变，为后续从背景中识别目标提供数据基础。 从中可以看出，随着飞机距离愈来愈远，飞机光谱辐射强度逐渐减弱，但是光谱主要的波峰、波谷以及曲线趋势基本保持一致，为后续进行目标识别提供了坚实的数据基础。上述图例均说明了在信息熵优化的自动控制策略下图谱联合测量系统及方法的优越性和先进行。
本领域的技术人员容易理解，以上所述仅为本发明的较佳实施例而已，并不用以限制本发明，凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等，均应包含在本发明的保护范围之内。

Claims (7)

  1. 一种多波段共光路图谱联合遥感测量系统，由防护性的壳体包裹，其特征在于，包括红外窗口(1)、二维转镜(2)、平面反射镜(3)、反射式多波段红外镜头(4)、傅里叶干涉光谱模块(5)、图谱联合处理模块(6)、电源模块(7)、制冷模块(8)和显示模块(9)；
    所述红外窗口(1)与壳体连接，用于透过目标场景中的红外光线；
    所述二维转镜(2)固定安装在所述壳体内部，所述二维转镜(2)的中心与所述红外窗口(1)的视场中心轴重合，所述二维转镜(2)的镜面与所述红外窗口(1)的视场中心轴成45°放置，用于跟踪目标场景中动目标，实现局部区域以及多目标采样；
    所述平面反射镜(3)的中心与所述二维转镜(2)的中心在同一水平高度，所述平面反射镜(3)的镜面与所述二维转镜(2)的镜面平行；
    所述反射式多波段红外镜头(4)的中心轴与所述平面反射镜(3)的中心重合，用于多波段的红外光线聚焦，并实现成像与光谱测量共光路；
    所述傅里叶干涉光谱模块(5)的光信号输入端与所述反射式多波段红外镜头(4)连接，所述傅里叶干涉光谱模块(5)的电信号输出端与所述图谱联合处理模块(6)连接，用于获取红外光谱数据；
    所述图谱联合处理模块(6)与所述反射式多波段红外镜头(4)连接，用于实现图谱联合处理；
    所述电源模块(7)、制冷模块(8)和显示模块(9)分别与所述图谱联合处理模块(6)连接。
  2. 如权利要求1所述的遥感测量系统，其特征在于，所述二维转镜(2)包括平面反射镜(21)、二维转台(22)和伺服电机(23)；
    所述平面反射镜(21)放置在所述二维转台(22)上，并通过转台上 的卡槽固定；所述伺服电机(23)的两个驱动轴分别与所述二维转台(22)俯仰轴以及旋转轴机械连接。
  3. 如权利要求1或2所述的遥感测量系统，其特征在于，所述反射式多波段红外镜头(4)包括镜体(41)、分光镜(42)、红外探测器(43)和红外光纤耦合器(44)；
    所述分光镜(42)的镜面中心与所述镜体(41)的中心轴重合，所述分光镜(42)的镜面与中心轴成45°；所述红外探测器(43)安装在尾部，所述红外探测器(43)中心与中心轴成一条直线；所述红外光纤耦合器(44)设置在尾部边缘。
  4. 如权利要求3所述的遥感测量系统，其特征在于，所述图谱联合处理模块(6)包括：通讯电平转换单元(61)、图像处理单元(62)、系统控制单元(63)、光谱处理单元(64)和数据融合总控单元(65)；
    所述通讯电平转换单元(61)通过RS422接口与所述反射式多波段红外镜头(4)中的所述红外探测器(43)连接；用于实现模块间的电平转换；
    所述图像处理单元(62)的一端通过图像接口与所述红外探测器(43)连接，所述图像处理单元(62)的另一端与所述数据融合总控单元(65)连接，用于对采集的红外图像进行图像增强、特征提取、分割处理，实现系统跟踪；
    所述系统控制单元(63)的输入端与所述通讯电平转换单元(61)连接，所述系统控制单元(63)的控制端与所述数据融合总控单元(65)连接，所述系统控制单元(63)的输出端与所述二维转镜(2)连接；用于通过串口对所述二维转镜(2)进行运动控制；
    所述光谱处理单元(64)的输入端与所述傅里叶干涉光谱模块(5)的电信号输出端连接，所述光谱处理单元(64)的输出端与所述(65)连接；用于对所述傅里叶干涉光谱模块(5)采集的红外光谱进行采集控制、数据预处理、辐射定标、光谱特征提取处理；
    所述数据融合总控单元(65)用于将所述图像处理单元(62)、所述系统控制单元(63)和所述光谱处理单元(64)的输出进行数据融合处理，并根据处理结果选择最优的系统控制策略进行数据采集。
  5. 一种基于权利要求1-4任一项所述的多波段共光路图谱联合遥感测量系统的测量方法，其特征在于，包括下述步骤：
    (1)通过红外探测器实现目标场景的全视场扫描采集获得图像序列；
    (2)在所述图像序列中提取至少一个感兴趣区；
    (3)计算所述图像序列中所有感兴趣区的特征描述矩阵；
    (4)判断当前时刻是否处于数据获取初始阶段，若是，则对所有感兴趣区的特征权重矩阵赋予初值；若否，则根据所有感兴趣区的特征描述矩阵将图像序列中所有感兴趣区与上一采集周期中获取的所有感兴趣区进行逐次匹配关联，提取关联成功的各感兴趣区的光谱特征，并将光谱特征添加到对应感兴趣区的特征描述矩阵中，同时将其特征权重矩阵复制到对应的感兴趣区的特征权重矩阵；
    (5)计算所有感兴趣区的特征信息熵，并按照特征信息熵的值从大到小将与之对应的感兴趣区进行排序，并赋予所有感兴趣区采集顺序的编号；
    (6)根据编号依次对各个感兴趣区中的采集点进行排序形成采集点序列；
    (7)依次将光轴点移动到所述采集点序列中的采集点；同时采用非成像的傅里叶干涉光谱模块获取感兴趣区的光谱数据；
    (8)重复步骤(1)～(7)进行下一个采集周期，直到系统采集结束。
  6. 如权利要求5所述的测量方法，其特征在于，在所述步骤(2)中，所述感兴趣区是指图像中紧密相连的若干像素点，提取感兴趣区具体包括：
    (21)利用非线性视觉原理进行图像增强处理；
    (22)对增强处理后的图像进行图像分割处理；
    (23)对分割处理后的图像进行膨胀腐蚀处理并获得所述感兴趣区。
  7. 如权利要求5所述的测量方法，其特征在于，在步骤(3)中，所述感兴趣区的特征描述矩阵包括位置特征、形状特征、灰度特征和光谱特征；所述位置特征包括采集点坐标
    Figure PCTCN2014085762-appb-100001
    边界链(c1，c2，......，cn)和采集距离L＝|x0-x0|+|y0-y0|；感兴趣区RGNV包含M个像素点(x1，y1)，(x2，y2)，......，(xM，yM)，n为感兴趣区边界像素点的个数，ct＝(xt，yt)为感兴趣区边界像素点；(x0，y0)为当前光轴点，其位置随着二维转镜的移动而不断变化；
    所述形状特征包括感兴趣区面积A和周长
    Figure PCTCN2014085762-appb-100002
    面积A＝M，周长P等于感兴趣区RGNV的长度；
    所述灰度特征包括平均灰度
    Figure PCTCN2014085762-appb-100003
    灰度梯度矩阵以及平均梯度；其中f(xt，yt)表示图像在像素点(xt，yt)处的灰度值；
    所述光谱特征包括光谱曲线变化趋势、波峰和波谷位置、大于均值的波峰数、小于均值的波峰数、大于均值的波谷数、小于均值的波谷数、峰宽、峰高。
PCT/CN2014/085762 2013-12-30 2014-09-02 一种多波段共光路图谱联合遥感测量系统及方法 WO2015101062A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US15/108,175 US9869793B2 (en) 2013-12-30 2014-09-02 Multiband common-optical-path image-spectrum associated remote sensing measurement system and method

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201310750061.5A CN103776540B (zh) 2013-12-30 2013-12-30 一种多波段共光路图谱联合遥感测量系统及方法
CN201310750061.5 2013-12-30

Publications (1)

Publication Number Publication Date
WO2015101062A1 true WO2015101062A1 (zh) 2015-07-09

Family

ID=50569057

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2014/085762 WO2015101062A1 (zh) 2013-12-30 2014-09-02 一种多波段共光路图谱联合遥感测量系统及方法

Country Status (3)

Country Link
US (1) US9869793B2 (zh)
CN (1) CN103776540B (zh)
WO (1) WO2015101062A1 (zh)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107703555A (zh) * 2016-06-20 2018-02-16 北京英特威视科技有限公司 一种危险源探测方法及系统
CN109934788A (zh) * 2019-03-22 2019-06-25 鲁东大学 一种基于标准遥感图像的遥感图像缺失数据修复方法
CN111323839A (zh) * 2020-04-07 2020-06-23 中国人民解放军国防科技大学 一种光电探测机构用嵌入式控制系统
CN112304904A (zh) * 2019-07-15 2021-02-02 松山湖材料实验室 基于滤波阵列的硅片反射率检测方法
CN114374779A (zh) * 2021-12-16 2022-04-19 中国科学院上海高等研究院 一种全光场成像相机及其成像方法及全光场成像装置

Families Citing this family (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103776540B (zh) * 2013-12-30 2015-07-08 华中科技大学 一种多波段共光路图谱联合遥感测量系统及方法
CN104501956B (zh) * 2014-12-30 2016-07-13 华中科技大学 一种超宽波段图谱关联探测装置与探测方法
CN104501959B (zh) 2014-12-30 2016-08-17 华中科技大学 一种红外图谱关联智能探测方法及装置
CN104535186B (zh) 2014-12-30 2016-08-17 华中科技大学 一种动平台红外图谱关联探测系统及方法
CN104502918B (zh) * 2014-12-30 2016-09-07 华中科技大学 一种低轨卫星星载图谱关联探测方法与载荷
CN104516110A (zh) 2014-12-30 2015-04-15 华中科技大学 一种共孔径宽波段红外光学系统
CN104748849A (zh) * 2015-03-05 2015-07-01 青岛市光电工程技术研究院 一种现场远距离光谱探测装置和方法
CN105676305B (zh) * 2015-12-31 2017-05-31 南京华图信息技术有限公司 一种共口径多视场图谱协同探测系统与方法
CN106527502A (zh) * 2016-12-09 2017-03-22 中国电子科技集团公司第三十四研究所 一种无线光天线自动跟踪系统
WO2019180571A1 (en) * 2018-03-18 2019-09-26 Unispectral Ltd. Generating narrow-band spectral images from broad-band spectral images
US10778912B2 (en) * 2018-03-31 2020-09-15 Open Water Internet Inc. System and device for optical transformation
CN108896188B (zh) * 2018-06-06 2020-07-14 北京空间飞行器总体设计部 一种共口径高轨遥感卫星红外通道大气校正方法
CN108875649A (zh) * 2018-06-22 2018-11-23 北京佳格天地科技有限公司 一种地物分类方法、系统、设备和存储介质
CN108956101B (zh) * 2018-08-20 2024-05-07 中国科学院上海技术物理研究所 一种相机视轴变化的测量装置和测量方法
CN109284706B (zh) * 2018-09-12 2023-12-01 国际商业机器（中国)投资有限公司 基于多源卫星遥感数据的热点网格工业聚集区域识别方法
CN109612512B (zh) * 2018-10-23 2021-11-23 南京航空航天大学 一种空基光电系统多模态一体化测试平台及测试方法
CN109990902B (zh) * 2018-12-29 2020-02-14 华中科技大学 一种图谱关联的分谱段估计目标亮温的方法
CN110118965B (zh) * 2019-04-01 2023-06-23 贵州航天电子科技有限公司 一种无线电对空动目标探测特性外场试验装置
CN111208961B (zh) * 2019-12-30 2024-07-23 西安闻泰电子科技有限公司 将镜面影像映射到电子屏幕的装置及方法
CN112017156B (zh) * 2020-07-17 2023-02-14 中国科学院西安光学精密机械研究所 基于多光谱视频的空间点目标旋转周期估计方法
CN112945381B (zh) * 2021-02-02 2022-08-16 上海机电工程研究所 一种基于视域选通的图谱分时探测系统及方法
CN113504181B (zh) * 2021-08-04 2024-06-28 梁宵 基于傅里叶红外光谱技术的气云监测装置及监测方法
CN113945952B (zh) * 2021-09-30 2022-08-19 中国空间技术研究院 空间分布式合成孔径光学探测方法
CN114563084A (zh) * 2022-02-07 2022-05-31 中电科思仪科技股份有限公司 一种实时傅立叶红外光谱辐射测量系统及测量方法
CN115105784B (zh) * 2022-06-20 2023-04-07 连云港新江环保材料有限公司 一种垃圾焚烧飞灰中二噁英的脱氯解毒处理方法
CN115294439B (zh) * 2022-08-02 2023-04-07 北京卫星信息工程研究所 空中弱小运动目标检测方法、系统、设备及存储介质
CN116049641B (zh) * 2023-04-03 2023-06-30 中国科学院光电技术研究所 一种基于红外光谱的点目标特征提取方法
CN117115679B (zh) * 2023-10-25 2024-06-25 北京佳格天地科技有限公司 一种时空融合遥感影像对筛选方法
CN117169138B (zh) * 2023-11-02 2024-01-09 天津市农业发展服务中心 一种基于遥感数据的区域墒情监测方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6781127B1 (en) * 2000-06-08 2004-08-24 Equinox Corporation Common aperture fused reflective/thermal emitted sensor and system
CN1604139A (zh) * 2004-10-28 2005-04-06 上海交通大学 图像融合评价系统的构建方法
CN101702021A (zh) * 2009-11-06 2010-05-05 华中科技大学 一种图谱一体化的时变对象光谱信息获取方法与装置
CN102564589A (zh) * 2011-12-20 2012-07-11 华中科技大学 一种多波段动目标光谱特征探测识别方法和装置
CN102651124A (zh) * 2012-04-07 2012-08-29 西安电子科技大学 基于冗余字典稀疏表示和评价指标的图像融合方法
CN103776540A (zh) * 2013-12-30 2014-05-07 华中科技大学 一种多波段共光路图谱联合遥感测量系统及方法

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8759773B2 (en) * 2012-04-18 2014-06-24 Raytheon Company Infrared spectrometer with enhanced readout speed

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6781127B1 (en) * 2000-06-08 2004-08-24 Equinox Corporation Common aperture fused reflective/thermal emitted sensor and system
CN1604139A (zh) * 2004-10-28 2005-04-06 上海交通大学 图像融合评价系统的构建方法
CN101702021A (zh) * 2009-11-06 2010-05-05 华中科技大学 一种图谱一体化的时变对象光谱信息获取方法与装置
CN102564589A (zh) * 2011-12-20 2012-07-11 华中科技大学 一种多波段动目标光谱特征探测识别方法和装置
CN102651124A (zh) * 2012-04-07 2012-08-29 西安电子科技大学 基于冗余字典稀疏表示和评价指标的图像融合方法
CN103776540A (zh) * 2013-12-30 2014-05-07 华中科技大学 一种多波段共光路图谱联合遥感测量系统及方法

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107703555A (zh) * 2016-06-20 2018-02-16 北京英特威视科技有限公司 一种危险源探测方法及系统
CN109934788A (zh) * 2019-03-22 2019-06-25 鲁东大学 一种基于标准遥感图像的遥感图像缺失数据修复方法
CN112304904A (zh) * 2019-07-15 2021-02-02 松山湖材料实验室 基于滤波阵列的硅片反射率检测方法
CN112304904B (zh) * 2019-07-15 2023-11-03 松山湖材料实验室 基于滤波阵列的硅片反射率检测方法
CN111323839A (zh) * 2020-04-07 2020-06-23 中国人民解放军国防科技大学 一种光电探测机构用嵌入式控制系统
CN111323839B (zh) * 2020-04-07 2023-04-07 中国人民解放军国防科技大学 一种光电探测机构用嵌入式控制系统
CN114374779A (zh) * 2021-12-16 2022-04-19 中国科学院上海高等研究院 一种全光场成像相机及其成像方法及全光场成像装置

Also Published As

Publication number Publication date
US9869793B2 (en) 2018-01-16
US20160327682A1 (en) 2016-11-10
CN103776540B (zh) 2015-07-08
CN103776540A (zh) 2014-05-07

Similar Documents

Publication Publication Date Title
WO2015101062A1 (zh) 一种多波段共光路图谱联合遥感测量系统及方法
US9759835B2 (en) Infrared image-spectrum associated intelligent detection method and apparatus
US9921102B2 (en) Moving platform borne infrared image-spectrum associated detection system and method
WO2016106957A1 (zh) 一种超宽波段图谱关联探测装置与探测方法
CN101738619B (zh) 双波段红外光学系统
CN106772417B (zh) 一种动目标多维度多尺度红外光谱特征测量方法及系统
WO2017041335A1 (zh) 一种全光学波段图谱协同探测动目标的装置和方法
WO2016106954A1 (zh) 一种低轨卫星星载图谱关联探测方法与载荷
WO2013091286A1 (zh) 一种多波段动目标光谱特征探测识别方法和装置
CN109490223B (zh) 一种基于可编程高光谱成像的目标探测识别系统及方法
CN204439211U (zh) 一种超宽波段图谱关联探测装置
Deng et al. A compact mid-wave infrared imager system with real-time target detection and tracking
CN108844628A (zh) 一种多光谱成像探测系统
CN108305290B (zh) 一种动目标的精确测谱方法
CN102519596B (zh) 地球静止轨道高分辨率干涉光谱成像系统
CN116740459A (zh) 一种联合空间特征和光谱特征的点云分类方法
CN214334007U (zh) 一种基于扫描振镜的凝视型高光谱相机
CN109781259B (zh) 一种图谱关联的精准测量空中运动小目标红外光谱的方法
CN115793271B (zh) 基于光学结构的偏振/红外视觉导航与环境感知终端
Xue et al. Polarization imaging detection technology research
Tang et al. Remote sensing small object detection algorithm based on contextual information enhancement
Brady et al. Reflective Signatures of Unresolved Objects
Dong et al. Characteristic analysis of infrared target and design of target detection system
Wu et al. An Infrared Target Images Recognition and Processing Method Based on the Fuzzy Comprehensive Evaluation
CN113223143A (zh) 基于飞行时间三维成像测量传感器的电弧测量方法及系统

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 14876834

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 15108175

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 14876834

Country of ref document: EP

Kind code of ref document: A1