2017 Vol. 38, No. 3
Display Method:
2017, 38(3): 341-347.
doi: 10.5768/JAO201738.0301001
Abstract:
It is difficult for traditional gear drive to meet requirements of high precision drive of eletro-optic system because of problems of backlash and gap. Based on analysis of technical characteristics of cable and steel belt drive, 1:1 angle measuring mechanism of periscope based on cable drive and 2:1 drive mechanism for electro-optic system based on steel belt drive are designed. Based on optical measurement method, transmission error of cable and steel belt drive mechanism is measured by rotary transformer and theodolite respectively. Results show that transmission accuracy of cable drive can reach minute level, corner error is less than 3.5′ in range of -110°~+135°; while transmission accuracy of steel belt drive can reach second level, angle error is less than 11.9″ in range of -3°~+3°. In conclusion, flexible drive has high transmission accuracy and can satisfy performance requirements of electro-optical system.
It is difficult for traditional gear drive to meet requirements of high precision drive of eletro-optic system because of problems of backlash and gap. Based on analysis of technical characteristics of cable and steel belt drive, 1:1 angle measuring mechanism of periscope based on cable drive and 2:1 drive mechanism for electro-optic system based on steel belt drive are designed. Based on optical measurement method, transmission error of cable and steel belt drive mechanism is measured by rotary transformer and theodolite respectively. Results show that transmission accuracy of cable drive can reach minute level, corner error is less than 3.5′ in range of -110°~+135°; while transmission accuracy of steel belt drive can reach second level, angle error is less than 11.9″ in range of -3°~+3°. In conclusion, flexible drive has high transmission accuracy and can satisfy performance requirements of electro-optical system.
2017, 38(3): 348-351.
doi: 10.5768/JAO201738.0301002
Abstract:
Dual-band common aperture camera can take into account advantages of both near infrared and visible light. It not only ensures rich color information access, but also can take full advantage of better through fog and high sensitivity low light detection capability of near infrared detection. Dual-band common aperture camera with working band of 450 nm ~ 650 nm and 750 nm ~ 900 nm is designed, with field angle of 35°× 20°, F number 1.5, and globally complex double-Gaussian structure used. Results show that MTF is more than> 0.4 at 50 lp/mm, system diffuse spot is less than 10 μm, and distortion is less than 3%.
Dual-band common aperture camera can take into account advantages of both near infrared and visible light. It not only ensures rich color information access, but also can take full advantage of better through fog and high sensitivity low light detection capability of near infrared detection. Dual-band common aperture camera with working band of 450 nm ~ 650 nm and 750 nm ~ 900 nm is designed, with field angle of 35°× 20°, F number 1.5, and globally complex double-Gaussian structure used. Results show that MTF is more than> 0.4 at 50 lp/mm, system diffuse spot is less than 10 μm, and distortion is less than 3%.
2017, 38(3): 352-357.
doi: 10.5768/JAO201738.0301003
Abstract:
Infrared optical system is the core component of infrared imaging devices. Athermalization is widely adopted in mid-infrared optical design in order to improve performance and stability of infrared imaging optical systems, and meet performance requirements of civil and military applications. In this paper, passive athermalization design of a non-thermal mid-infrared imaging optical system is realized through combining different infrared materials. Contribution of narcissus is calculated by NITD method, curvature and optical power of surface with large contribution to narcissus are optimized. Results show that MTF of optical system is greater than 0.5 at 30 lp/mm in temperature range of -40 ℃ ~ 60 ℃, defocus distance is less than depth of light, RMS diameter is less than pixel size. Maximum NITD value of narcissus is reduced by 40%, which is smaller than minimum resolution temperature difference of the system. Application of this method can be very good to achieve athermalization design of infrared imaging optics system, and suppression effect of narcissus is obvious.
Infrared optical system is the core component of infrared imaging devices. Athermalization is widely adopted in mid-infrared optical design in order to improve performance and stability of infrared imaging optical systems, and meet performance requirements of civil and military applications. In this paper, passive athermalization design of a non-thermal mid-infrared imaging optical system is realized through combining different infrared materials. Contribution of narcissus is calculated by NITD method, curvature and optical power of surface with large contribution to narcissus are optimized. Results show that MTF of optical system is greater than 0.5 at 30 lp/mm in temperature range of -40 ℃ ~ 60 ℃, defocus distance is less than depth of light, RMS diameter is less than pixel size. Maximum NITD value of narcissus is reduced by 40%, which is smaller than minimum resolution temperature difference of the system. Application of this method can be very good to achieve athermalization design of infrared imaging optics system, and suppression effect of narcissus is obvious.
2017, 38(3): 358-364.
doi: 10.5768/JAO201738.0301004
Abstract:
Visible light positioning (VLP) based on mobile phone not only has high positioning accuracy, but also has advantages of visible light communication and mobile internet. Due to high computational complexity of image processing, most existing system can not support real-time navigation because of its low positioning speed. A fast indoor VLP system with high accuracy is proposed in this paper. Positioning latency is reduced by designing elaborate flicker-free line coding scheme and lightweight image processing algorithm. In addition, this system has advantage of supporting flicker mitigation and dimming, which are important for illumination. Android-based system prototype has been developed for field tests on an off-the-shelf smartphone. Experimental results show that it can achieve high accuracy of 7.5 cm, and positioning time is reduced to 22.7 ms for single-luminaire and to 35.7 ms for dual-luminaries respectively. In addition, it supports real-time indoor navigation for users moving at a speed of up to 18 km/h.
Visible light positioning (VLP) based on mobile phone not only has high positioning accuracy, but also has advantages of visible light communication and mobile internet. Due to high computational complexity of image processing, most existing system can not support real-time navigation because of its low positioning speed. A fast indoor VLP system with high accuracy is proposed in this paper. Positioning latency is reduced by designing elaborate flicker-free line coding scheme and lightweight image processing algorithm. In addition, this system has advantage of supporting flicker mitigation and dimming, which are important for illumination. Android-based system prototype has been developed for field tests on an off-the-shelf smartphone. Experimental results show that it can achieve high accuracy of 7.5 cm, and positioning time is reduced to 22.7 ms for single-luminaire and to 35.7 ms for dual-luminaries respectively. In addition, it supports real-time indoor navigation for users moving at a speed of up to 18 km/h.
2017, 38(3): 365-371.
doi: 10.5768/JAO201738.0301005
Abstract:
A multipurpose calibration device has been successfully designed to measure variable angle luminosity of light source and body surface and chroma parameter, which adopts two-dimensional automatic rotation control combined with servo motor, sensor and measurement software. It solves problems that measurement devices to measure the luminosity and chromaticity parameter are all independent device. Results show that this device can measure such parameters as total luminous flux, A/B field average luminous intensity with CIE, isocandela/illumination curve and small size display photometric chromaticity and variable angle chromaticity of body surface. Average luminous intensity, luminous flux, Y value and chromaticity coordinates and the standard deviation of the device are-1.8%, 1.5%, 0.9 and 0.005 respectively. So this device has met design requirements of high accuracy, convenient function switch, high level of automation and simple operation.
A multipurpose calibration device has been successfully designed to measure variable angle luminosity of light source and body surface and chroma parameter, which adopts two-dimensional automatic rotation control combined with servo motor, sensor and measurement software. It solves problems that measurement devices to measure the luminosity and chromaticity parameter are all independent device. Results show that this device can measure such parameters as total luminous flux, A/B field average luminous intensity with CIE, isocandela/illumination curve and small size display photometric chromaticity and variable angle chromaticity of body surface. Average luminous intensity, luminous flux, Y value and chromaticity coordinates and the standard deviation of the device are-1.8%, 1.5%, 0.9 and 0.005 respectively. So this device has met design requirements of high accuracy, convenient function switch, high level of automation and simple operation.
2017, 38(3): 372-380.
doi: 10.5768/JAO201738.0301006
Abstract:
Aviation gravimeter is a precision instrument for measuring gravity of the earth. In order to make the gravimeter isolate all kinds of disturbances of flight process in working state, and always keep an upright and stable state, inertial stable platform of aviation gravimeter with high accuracy is designed. At the same time, in order to shorten design cycle of stable platform and ensure performance of designed platform to meet technical requirements, mechatronic joint modeling and simulation are carried out on high accuracy inertial stable platform of aviation gravimeter. Mechatronic joint model based on Adams and SimuLink is built and time domain and frequency domain characteristics are simulated and analyzed. Simulation results and experimental results are compared. Results show that simulation results of time domain and frequency domain approximately agree with experimental results under conditions of open loop, elevation axis amplitude-frequency matching degree is greater than 83% and phase-frequency matching degree is greater than 79.2%; roll axis amplitude-frequency matching degree is greater than 92.9% and phase-frequency matching degree is greater than 86.1%. Time domain matching of two axes is similar, and correctness and validity of simulation model are verified. This model can predict servo performance of stable platform and lay foundation for actual test and debugging of platform.
Aviation gravimeter is a precision instrument for measuring gravity of the earth. In order to make the gravimeter isolate all kinds of disturbances of flight process in working state, and always keep an upright and stable state, inertial stable platform of aviation gravimeter with high accuracy is designed. At the same time, in order to shorten design cycle of stable platform and ensure performance of designed platform to meet technical requirements, mechatronic joint modeling and simulation are carried out on high accuracy inertial stable platform of aviation gravimeter. Mechatronic joint model based on Adams and SimuLink is built and time domain and frequency domain characteristics are simulated and analyzed. Simulation results and experimental results are compared. Results show that simulation results of time domain and frequency domain approximately agree with experimental results under conditions of open loop, elevation axis amplitude-frequency matching degree is greater than 83% and phase-frequency matching degree is greater than 79.2%; roll axis amplitude-frequency matching degree is greater than 92.9% and phase-frequency matching degree is greater than 86.1%. Time domain matching of two axes is similar, and correctness and validity of simulation model are verified. This model can predict servo performance of stable platform and lay foundation for actual test and debugging of platform.
2017, 38(3): 381-384.
doi: 10.5768/JAO201738.0302001
Abstract:
In order to realize diagnosis of inertial confinement fusion (ICF), high temperature plasma X-ray energy spectrum information and two-dimensional spatial resolution information of detonation target are obtained. Spherical crystal analyzer is fabricated by using Bragg diffraction property of crystal, with bent radius 125 mm. In order to verify spatial resolution of spherical crystal, a backlight imaging platform is designed for backlight imaging experiment. Quartz spherical crystal is diffractive core element, and IP plate of receiving device gets two-dimensional spatial resolution information of Cu target. Experimental results show that spatial resolution of imaging system based on spherically bent Quartz crystal is about 100 μm.
In order to realize diagnosis of inertial confinement fusion (ICF), high temperature plasma X-ray energy spectrum information and two-dimensional spatial resolution information of detonation target are obtained. Spherical crystal analyzer is fabricated by using Bragg diffraction property of crystal, with bent radius 125 mm. In order to verify spatial resolution of spherical crystal, a backlight imaging platform is designed for backlight imaging experiment. Quartz spherical crystal is diffractive core element, and IP plate of receiving device gets two-dimensional spatial resolution information of Cu target. Experimental results show that spatial resolution of imaging system based on spherically bent Quartz crystal is about 100 μm.
2017, 38(3): 385-391.
doi: 10.5768/JAO201738.0302002
Abstract:
As necessary part of guiding process carried by large space telescope, fine guidance star catalog is key basis to determine absolute pointing of LOS of spacecraft telescope. Selection of guide stars, catalog based storage capacity, division and regional search speeds are vital to realize functions and performances of fine guidance sensor. Under conditions of FOV about 0.08deg2 and high precision requirements, UCAC4 is selected as basic star catalog, and mixed partition visual magnitude filtering algorithm is proposed to select guide stars. In order to realize fast searching of catalog, guide star table access mode is analyzed in details; 10 mv~16 mv fine guidance star catalog, containing about 4 million stars, is constructed to analyze star catalog uniformity and search speed. Simulation results show that global uniformity of 16 mv fine guidance star catalog based on UCAC4 reaches 0.44, and star search speed reach 0.05 s. Stars number in FOV is moderate and efficiently identified probability reaches 75.51%. This star catalog can be used as basis of fine guidance star map identification to support large-aperture space telescope observations.
As necessary part of guiding process carried by large space telescope, fine guidance star catalog is key basis to determine absolute pointing of LOS of spacecraft telescope. Selection of guide stars, catalog based storage capacity, division and regional search speeds are vital to realize functions and performances of fine guidance sensor. Under conditions of FOV about 0.08deg2 and high precision requirements, UCAC4 is selected as basic star catalog, and mixed partition visual magnitude filtering algorithm is proposed to select guide stars. In order to realize fast searching of catalog, guide star table access mode is analyzed in details; 10 mv~16 mv fine guidance star catalog, containing about 4 million stars, is constructed to analyze star catalog uniformity and search speed. Simulation results show that global uniformity of 16 mv fine guidance star catalog based on UCAC4 reaches 0.44, and star search speed reach 0.05 s. Stars number in FOV is moderate and efficiently identified probability reaches 75.51%. This star catalog can be used as basis of fine guidance star map identification to support large-aperture space telescope observations.
2017, 38(3): 392-399.
doi: 10.5768/JAO201738.0302003
Abstract:
This paper introduces a kind of embossed imaging technology based on phase-shifted structured light illumination. The principle and imaging characteristics of imaging technology are studied experimentally, and color embossed imaging is realized by phase-shifted illumination method. In our experiments, multi-steps phase-shifting sinusoidal patterns are projected obliquely onto the object to be photographed, and digital camera sequentially captures illuminated object, and then reconstructs desired image by calculated method using multiple captured images. Theoretical analysis and experimental results show that phase shift structured light illumination imaging technique can effectively eliminate influence of environment uneven illumination. The obtained reflectance distribution image is only related to structural light illumination, which highlights shadows formed by oblique projection illumination. Change of light and shade reflects normal variation of object surface, and thus apparent embossing effect is formed visually.
This paper introduces a kind of embossed imaging technology based on phase-shifted structured light illumination. The principle and imaging characteristics of imaging technology are studied experimentally, and color embossed imaging is realized by phase-shifted illumination method. In our experiments, multi-steps phase-shifting sinusoidal patterns are projected obliquely onto the object to be photographed, and digital camera sequentially captures illuminated object, and then reconstructs desired image by calculated method using multiple captured images. Theoretical analysis and experimental results show that phase shift structured light illumination imaging technique can effectively eliminate influence of environment uneven illumination. The obtained reflectance distribution image is only related to structural light illumination, which highlights shadows formed by oblique projection illumination. Change of light and shade reflects normal variation of object surface, and thus apparent embossing effect is formed visually.
2017, 38(3): 400-405.
doi: 10.5768/JAO201738.0302004
Abstract:
Calibration of multi-camera is essential in stereo visual measurement. When calibration plane board can not be visualized simultaneously by multiple cameras, relative position of cameras can not be achieved. In order to solve this problem, a method utilizing 2-axis turntable for multi-camera calibration without public field of view (FOV) is proposed in this paper. Before calibration, multi-camera system is installed and restrained on turntable properly, and relative positions are determined between coordinate of targets and turntable using rotation of turntable.Then turntable is rotated to ensure calibration board enters each camera's FOV, positons of each camera in recent target coordinate system are determined, and angles of turntable's axes are recorded simultaneously. Finally, relationship of relative positions between each camera is solved according to the relationship between coordinate of target turntable.The experimental results indicate that this solution is feasible, and calculating error is within 0.5%. Accordingly, this solution is proved to have certain accuracy in determining positon relationship of multi-camera with non-public FOV and wide angles between optical axes.
Calibration of multi-camera is essential in stereo visual measurement. When calibration plane board can not be visualized simultaneously by multiple cameras, relative position of cameras can not be achieved. In order to solve this problem, a method utilizing 2-axis turntable for multi-camera calibration without public field of view (FOV) is proposed in this paper. Before calibration, multi-camera system is installed and restrained on turntable properly, and relative positions are determined between coordinate of targets and turntable using rotation of turntable.Then turntable is rotated to ensure calibration board enters each camera's FOV, positons of each camera in recent target coordinate system are determined, and angles of turntable's axes are recorded simultaneously. Finally, relationship of relative positions between each camera is solved according to the relationship between coordinate of target turntable.The experimental results indicate that this solution is feasible, and calculating error is within 0.5%. Accordingly, this solution is proved to have certain accuracy in determining positon relationship of multi-camera with non-public FOV and wide angles between optical axes.
2017, 38(3): 406-414.
doi: 10.5768/JAO201738.0302005
Abstract:
In order to realize selective encryption of infrared images, an infrared image selection algorithm based on multi-feature difference detection and joint control mapping is proposed. Initial infrared image is enhanced by introducing segmented Sine transform to highlight true infrared target. Then target decision factor is designed by considering gray difference between target and background, and infrared background suppression mechanism is constructed by segmenting structure elements of Top-Hat transform to filter clutter and noise. Multi-feature difference detection model is established by using gray level, contrast and similarity degree to extract region of interest of real target and suspicious object. Joint control chaotic mapping is designed by Logistic map, Tent map and chebysbev map to permutation region of interest. Finally, gravity model is introduced to spread pixels in region of interest for complete infrared target selection encryption. Experimental results show that compared with existing local encryption mechanism, entropy of cipher text information is 7.982 6, which can be used for local selective encryption of infrared images.
In order to realize selective encryption of infrared images, an infrared image selection algorithm based on multi-feature difference detection and joint control mapping is proposed. Initial infrared image is enhanced by introducing segmented Sine transform to highlight true infrared target. Then target decision factor is designed by considering gray difference between target and background, and infrared background suppression mechanism is constructed by segmenting structure elements of Top-Hat transform to filter clutter and noise. Multi-feature difference detection model is established by using gray level, contrast and similarity degree to extract region of interest of real target and suspicious object. Joint control chaotic mapping is designed by Logistic map, Tent map and chebysbev map to permutation region of interest. Finally, gravity model is introduced to spread pixels in region of interest for complete infrared target selection encryption. Experimental results show that compared with existing local encryption mechanism, entropy of cipher text information is 7.982 6, which can be used for local selective encryption of infrared images.
2017, 38(3): 415-420.
doi: 10.5768/JAO201738.0302006
Abstract:
Polarization imaging has unique advantages in target detection, recognition and processing compared with traditional photoelectric detection technology. To overcome degraged images taken in haze weather, an image defogging method based on polarization information is proposed. By obtaining target polarized image at three angles, Stokes vector of scene object is solved. From relationship between Stokes vector and Mueller matrix, variation law of polarized image intensity with polarization angle is analyzed, orthogonal polarization images under maximum and minimum light intensity are obtained automatically and accurately. Polarization degree of atmosphere and its infinity atmospheric light intensity value are estimated using polarization filtering and bright channel priori method, and fog-free images are reconstructed. Experimental results show that clear images can be reconstructed in haze weather by using orthogonal polarized images obtained, average gradient and edge intensity of reconstructed image are improved by about 3 times and grey standard deviation is improved by about 88%.
Polarization imaging has unique advantages in target detection, recognition and processing compared with traditional photoelectric detection technology. To overcome degraged images taken in haze weather, an image defogging method based on polarization information is proposed. By obtaining target polarized image at three angles, Stokes vector of scene object is solved. From relationship between Stokes vector and Mueller matrix, variation law of polarized image intensity with polarization angle is analyzed, orthogonal polarization images under maximum and minimum light intensity are obtained automatically and accurately. Polarization degree of atmosphere and its infinity atmospheric light intensity value are estimated using polarization filtering and bright channel priori method, and fog-free images are reconstructed. Experimental results show that clear images can be reconstructed in haze weather by using orthogonal polarized images obtained, average gradient and edge intensity of reconstructed image are improved by about 3 times and grey standard deviation is improved by about 88%.
2017, 38(3): 421-428.
doi: 10.5768/JAO201738.0302007
Abstract:
In order to assist unmanned vehicles in understanding scene of night vision images, detecting and identifying surrounding environment more quickly and accurately at night, a semantic segmentation method of unmmanned vehicle night vision images based on convolution-deconvolution neural network is proposed, which uses deep learning to segment scenery semant of night vision images. Convolution-deconvolution neural network is constructed by adding deconvolution network to traditional convolutional neural network, without selecting feature manually. By learning and training pixels-to-pixels, image semantic segmentation model can be obtained. The model can be used to predict scene semantic category of each pixel in night vision image, realizing environment perception of unmanned vehicles at night, which is import for automatic driving at night. Experimental results show that this method has good accuracy and real-time performance, and average IU reaches 68.47.
In order to assist unmanned vehicles in understanding scene of night vision images, detecting and identifying surrounding environment more quickly and accurately at night, a semantic segmentation method of unmmanned vehicle night vision images based on convolution-deconvolution neural network is proposed, which uses deep learning to segment scenery semant of night vision images. Convolution-deconvolution neural network is constructed by adding deconvolution network to traditional convolutional neural network, without selecting feature manually. By learning and training pixels-to-pixels, image semantic segmentation model can be obtained. The model can be used to predict scene semantic category of each pixel in night vision image, realizing environment perception of unmanned vehicles at night, which is import for automatic driving at night. Experimental results show that this method has good accuracy and real-time performance, and average IU reaches 68.47.
2017, 38(3): 429-433.
doi: 10.5768/JAO201738.0302008
Abstract:
Visual locating technology is one of main development direction of mobile robot locating. Aiming at problem of poor real-time performance of traditional visual positioning technology, this paper proposes a fast visual locating technology based on adaptive down sampling. Down sampling rate is calculated firstly based on previous size characteristic and minimum resolution size. Secondly down sampling and image segmentation are made on next image. Object locating zone in original source image are determined according to coordinates of object and down sampling rate. Then image segmentation and feature extraction of source image of the area are made. Size of feature extraction is taken as an image input, and required features extraction of visual location are taken as robot locating solution. Thus visual positioning time is effectively reduced with ensured visual positioning accuracy. Experiments show that it takes 20.23 s for traditional technology to process 100 figures while it takes 1.78 s for technology given out by this paper, which is only 8.8% of traditional technology. This technology can reduce image processing time effectively and improve real-time performance of mobile robot locating.
Visual locating technology is one of main development direction of mobile robot locating. Aiming at problem of poor real-time performance of traditional visual positioning technology, this paper proposes a fast visual locating technology based on adaptive down sampling. Down sampling rate is calculated firstly based on previous size characteristic and minimum resolution size. Secondly down sampling and image segmentation are made on next image. Object locating zone in original source image are determined according to coordinates of object and down sampling rate. Then image segmentation and feature extraction of source image of the area are made. Size of feature extraction is taken as an image input, and required features extraction of visual location are taken as robot locating solution. Thus visual positioning time is effectively reduced with ensured visual positioning accuracy. Experiments show that it takes 20.23 s for traditional technology to process 100 figures while it takes 1.78 s for technology given out by this paper, which is only 8.8% of traditional technology. This technology can reduce image processing time effectively and improve real-time performance of mobile robot locating.
2017, 38(3): 434-437.
doi: 10.5768/JAO201738.0303001
Abstract:
Aiming at the problems of shift of lubricant films on head slider, and dynamic change visualization of lubricant films on head slider, this paper presents caculation model of lubricant thickness distribution on a head slider surface using phase shift method, which is base on improved vertical-objective-based ellipsometric microscope (VEM). Measurement of lubricant film thickness is achived with this method. Taking nonpolar perfluoropolyether (PFPE) lubricant Z03 as a sample, ellipsometric microscope is calibrated; while taking polar PFPE lubricant Zdol4000 as a sample, dewetting on head slider surface is observed. Experimental results show measuring accuracy and resolution are 0.37 nm and 0.36 μm respectively. This can be reference for other visualization of nm lubricant films.
Aiming at the problems of shift of lubricant films on head slider, and dynamic change visualization of lubricant films on head slider, this paper presents caculation model of lubricant thickness distribution on a head slider surface using phase shift method, which is base on improved vertical-objective-based ellipsometric microscope (VEM). Measurement of lubricant film thickness is achived with this method. Taking nonpolar perfluoropolyether (PFPE) lubricant Z03 as a sample, ellipsometric microscope is calibrated; while taking polar PFPE lubricant Zdol4000 as a sample, dewetting on head slider surface is observed. Experimental results show measuring accuracy and resolution are 0.37 nm and 0.36 μm respectively. This can be reference for other visualization of nm lubricant films.
2017, 38(3): 438-444.
doi: 10.5768/JAO201738.0303002
Abstract:
This paper describes new vision measurement method for shape measurement of large-size object. Combine and intersect two monocular vision measurement systems without public filed of view for measurements of pose estimation and monocular vision intersection respectively. Pose estimation system solves the PnP problem to provide position information for monocular measurement. Monocular measurement system combining with auxiliary information from pose estimation calculates shape information of object. This paper analyses the system principle, gives PnP algorithm's implementation process of pose estimation system, and introduces the principle and implementation process of calibration method for camera without public field of view. Finally, static system verification test of measurement system of indoor large-size object has been carried out, and the feasibilities of this system have been validated. Experimental results show that measuring accuracy of system is 5 mm in the scope of the 3 m×3 m. It is proved that this system can be used in shape measurement of large-size object, and the structure of system is simple and light with great flexibility.
This paper describes new vision measurement method for shape measurement of large-size object. Combine and intersect two monocular vision measurement systems without public filed of view for measurements of pose estimation and monocular vision intersection respectively. Pose estimation system solves the PnP problem to provide position information for monocular measurement. Monocular measurement system combining with auxiliary information from pose estimation calculates shape information of object. This paper analyses the system principle, gives PnP algorithm's implementation process of pose estimation system, and introduces the principle and implementation process of calibration method for camera without public field of view. Finally, static system verification test of measurement system of indoor large-size object has been carried out, and the feasibilities of this system have been validated. Experimental results show that measuring accuracy of system is 5 mm in the scope of the 3 m×3 m. It is proved that this system can be used in shape measurement of large-size object, and the structure of system is simple and light with great flexibility.
2017, 38(3): 445-450.
doi: 10.5768/JAO201738.0303003
Abstract:
The objective of this paper is to explore feasibilities on attitude measurement of flying targets of remote and small field of view in range using weak perspective projection. Taking axis symmetric rotating object as an example, attitude measurement module of monocular vision is derived reversely by using projection image information and weak perspective projection principle. Experiment results verifies feasibilities of range attitude measurement in weak perspective projection, Under mentioned testing environments, error between weak perspective projection and perspective projection is less than 0.05°, and it mainly derives from operation range. Ultimate operating range is derived with algorithm applied. This study above lays a theoretical foundation for attitude measurement of flying targets of remote and small field of view in range.
The objective of this paper is to explore feasibilities on attitude measurement of flying targets of remote and small field of view in range using weak perspective projection. Taking axis symmetric rotating object as an example, attitude measurement module of monocular vision is derived reversely by using projection image information and weak perspective projection principle. Experiment results verifies feasibilities of range attitude measurement in weak perspective projection, Under mentioned testing environments, error between weak perspective projection and perspective projection is less than 0.05°, and it mainly derives from operation range. Ultimate operating range is derived with algorithm applied. This study above lays a theoretical foundation for attitude measurement of flying targets of remote and small field of view in range.
2017, 38(3): 451-455.
doi: 10.5768/JAO201738.0303004
Abstract:
Field of view (FOV) conformity between different channels is an important evaluating indicator of aperture-divided detection system. Accurate estimation of FOV conformity will make positive sense to design and fabrication of aperture-divided detection system. This paper firstly presents the principles of FOV measuring and FOV measuring conformity. Based on matrix scanning measurement, volume and area methods for measuring conformity of FOV between different channels are proposed. Then advantages and disadvantages of three measurement methods in different scenarios were briefly analyzed.Devices for measuring FOV of aperture-divided detection system are set up and degree of FOV conformity of aperture-divided radiometer are measured. Results indicate that: even if parallelism of optical axes among different channels is better than 95%, conformities of volume and area is still possibly less than 90%. Combination of using three methods can have a better understanding of manufacture and assembly of aperture-divided detection system.
Field of view (FOV) conformity between different channels is an important evaluating indicator of aperture-divided detection system. Accurate estimation of FOV conformity will make positive sense to design and fabrication of aperture-divided detection system. This paper firstly presents the principles of FOV measuring and FOV measuring conformity. Based on matrix scanning measurement, volume and area methods for measuring conformity of FOV between different channels are proposed. Then advantages and disadvantages of three measurement methods in different scenarios were briefly analyzed.Devices for measuring FOV of aperture-divided detection system are set up and degree of FOV conformity of aperture-divided radiometer are measured. Results indicate that: even if parallelism of optical axes among different channels is better than 95%, conformities of volume and area is still possibly less than 90%. Combination of using three methods can have a better understanding of manufacture and assembly of aperture-divided detection system.
2017, 38(3): 456-462.
doi: 10.5768/JAO201738.0303005
Abstract:
Infrared modulation detection technique is used to detect performance of scanning integral type electro-optical radar modulator. An optical transmission and laser collimation system are designed in this paper, which matches inherent properties of the modulator. Based on standard servo control theory, axis scanning model of Ⅲ system with stable condition is established for modulation of pulse code signal. Through FOV's switching of intercepting and tracking, and modulation of driving voltage amplitude, simulated infrared-converged laser scanning beam is coincided with 3°×3°, 40′×40′ cross target. Deviation of scanning FOV and zero position precision are metered, calibration mirror vibration position and scanning frequency zero-crossing pulse are collected in order to achieve infrared signal modulation performance detection. Results show that accuracy of infrared modulation detection technique is better than 0.5′(1σ), accuracy of zero error is less than 5′(1σ), and relative error of the system is less than ± 1%.It meets accuracy requirements of electro-optical radar in infrared optical axis assembly process, and can be applied to linear array scanning integral type modulator performance testing.
Infrared modulation detection technique is used to detect performance of scanning integral type electro-optical radar modulator. An optical transmission and laser collimation system are designed in this paper, which matches inherent properties of the modulator. Based on standard servo control theory, axis scanning model of Ⅲ system with stable condition is established for modulation of pulse code signal. Through FOV's switching of intercepting and tracking, and modulation of driving voltage amplitude, simulated infrared-converged laser scanning beam is coincided with 3°×3°, 40′×40′ cross target. Deviation of scanning FOV and zero position precision are metered, calibration mirror vibration position and scanning frequency zero-crossing pulse are collected in order to achieve infrared signal modulation performance detection. Results show that accuracy of infrared modulation detection technique is better than 0.5′(1σ), accuracy of zero error is less than 5′(1σ), and relative error of the system is less than ± 1%.It meets accuracy requirements of electro-optical radar in infrared optical axis assembly process, and can be applied to linear array scanning integral type modulator performance testing.
2017, 38(3): 463-468.
doi: 10.5768/JAO201738.0303006
Abstract:
Validation for radiometric and spectral calibration accuracy of hyperspectral remote sensor is necessary. Taking SVC spectral radiometer as an example, new type spectrum-tunable reference light source (STIS) and muti-band radiance standard transfer detector(MRSTD) are used to design validation method for radiometric and spectral calibration accuracy. In this method, spectral level of spectral shape is monotonically increasing and monotonically decreasing in measured band by new spectral adjustable integrating sphere reference light source. Wavelength of SVC spectral radiance is changed by spectral matching technique. Relative deviation of radiance measured by MRSTD and SVC spectral radiometer are calculated and contrasted. Contrast shows that spectral calibration accuracy is better than ±0.2 nm, and absolute radiometric calibration accuracy is less than 5%.
Validation for radiometric and spectral calibration accuracy of hyperspectral remote sensor is necessary. Taking SVC spectral radiometer as an example, new type spectrum-tunable reference light source (STIS) and muti-band radiance standard transfer detector(MRSTD) are used to design validation method for radiometric and spectral calibration accuracy. In this method, spectral level of spectral shape is monotonically increasing and monotonically decreasing in measured band by new spectral adjustable integrating sphere reference light source. Wavelength of SVC spectral radiance is changed by spectral matching technique. Relative deviation of radiance measured by MRSTD and SVC spectral radiometer are calculated and contrasted. Contrast shows that spectral calibration accuracy is better than ±0.2 nm, and absolute radiometric calibration accuracy is less than 5%.
2017, 38(3): 469-475.
doi: 10.5768/JAO201738.0303007
Abstract:
Aiming at the problem that detection accuracy of plane interferometry is limited by accuracy of reference surface shape, it is proposed to eliminate influence of reference surface error in interference system by using high precision absolute detection method based on parity function. Influences of rotation angle error and rotation eccentricity error on measurement accuracy of absolute detection method are simulated. Experiments and repetitive experiments of absolute detection accuracy are designed and analyzed by commercial Fischer interferometer. Simulation results show that measurement error PV value is 0.000 1λ when rotation angle error reaches 0.13 °; measurement error PV value is 0.005λ when rotation eccentricity error reaches 3 pixels. Experimental results show that absolute detection accuracy of actual sample PV10 value is 0.041 5λ and RMS value is 0.008 7λ, which is smaller than that of conventional interference detection. Point to point processing is done for two independent absolute detection results in the same plane, in this way residual graph is obtained with PV10 value is 0.004 λ and RMS value is 0.000 5 λ. Experimental results show that this method is highly reproducible and effective.
Aiming at the problem that detection accuracy of plane interferometry is limited by accuracy of reference surface shape, it is proposed to eliminate influence of reference surface error in interference system by using high precision absolute detection method based on parity function. Influences of rotation angle error and rotation eccentricity error on measurement accuracy of absolute detection method are simulated. Experiments and repetitive experiments of absolute detection accuracy are designed and analyzed by commercial Fischer interferometer. Simulation results show that measurement error PV value is 0.000 1λ when rotation angle error reaches 0.13 °; measurement error PV value is 0.005λ when rotation eccentricity error reaches 3 pixels. Experimental results show that absolute detection accuracy of actual sample PV10 value is 0.041 5λ and RMS value is 0.008 7λ, which is smaller than that of conventional interference detection. Point to point processing is done for two independent absolute detection results in the same plane, in this way residual graph is obtained with PV10 value is 0.004 λ and RMS value is 0.000 5 λ. Experimental results show that this method is highly reproducible and effective.
2017, 38(3): 476-479,498.
doi: 10.5768/JAO201738.0106001
Abstract:
Ion barrier film of Micro-channel Plate(MCP) is one of unique features of low-light-level(LLL)image intensifier with negative electron affinity. Its main function is to effectively prevent positive ions from bombarding cathode to improve working life of the device. Thickness of ionizing membrane of MCP determines size of electron permeation and ion blocking ability, and electron permeability directly affects contrast and signal to noise ratio(SNR) of the image. According to threshold voltage definition and test principle of ion barrier film of MCP, optimal thickness of ion barrier film is determined through lots of experiments. Results show that threshold voltage of ion barrier film increased linearly with increase of film thickness, linear relationship Y=3.98X+50 is proportional between two sides. Threshold voltage of ion barrier film decreased linearly with the increase of MCP operating voltage, showing an inverse proportion. This study is of great significance to improve performance of MCP in LLL image intensifier with negative electron affinity photonic cathode photoluminescence enhancers.
Ion barrier film of Micro-channel Plate(MCP) is one of unique features of low-light-level(LLL)image intensifier with negative electron affinity. Its main function is to effectively prevent positive ions from bombarding cathode to improve working life of the device. Thickness of ionizing membrane of MCP determines size of electron permeation and ion blocking ability, and electron permeability directly affects contrast and signal to noise ratio(SNR) of the image. According to threshold voltage definition and test principle of ion barrier film of MCP, optimal thickness of ion barrier film is determined through lots of experiments. Results show that threshold voltage of ion barrier film increased linearly with increase of film thickness, linear relationship Y=3.98X+50 is proportional between two sides. Threshold voltage of ion barrier film decreased linearly with the increase of MCP operating voltage, showing an inverse proportion. This study is of great significance to improve performance of MCP in LLL image intensifier with negative electron affinity photonic cathode photoluminescence enhancers.
2017, 38(3): 480-484.
doi: 10.5768/JAO201738.0106002
Abstract:
Micro-channel plate (MCP) with ion barrier film is one of the key components of Low Light Level(LLL)image intensifier with negative electron affinity, and working state of filmed MCP has a serious influence on it, so MCP best working voltage should be determined by testing filmed and un-filmed MCP gain in different voltage of MCP voltage and cathode voltage. ①For LLL image intensifier with negative electron affinity with un-filmed MCP, optimum operating voltage are: if cathode voltage exceeds a certain value Vc1 with MCP gain almost unchanged, this explains that cathode voltage Vc1 at this time is the best operating voltage; if MCP voltage is of a specific value Vm1 (cathod voltage is any value higher than Vc1) with low MCP gain, and MCP gain turns higher enough when > 20 000 MCP voltage becomes (Vm1+100V), thus this value (Vm1+100V) is the optimum MCP voltage; ②For the same material with filmed MCP, operating voltage are: cathode voltage Vc=algebraic sum of un-filmed MCP cathode voltage (Vc1) and threshold voltage, and MCP voltage is Vm > (Vm1+100V), whose specific value should be determined according to linear region of filmed MCP gain value. This study has an important significance to the determination of best working voltage and performance improvement of filmed MCP on NEA photocathode image intensifier.
Micro-channel plate (MCP) with ion barrier film is one of the key components of Low Light Level(LLL)image intensifier with negative electron affinity, and working state of filmed MCP has a serious influence on it, so MCP best working voltage should be determined by testing filmed and un-filmed MCP gain in different voltage of MCP voltage and cathode voltage. ①For LLL image intensifier with negative electron affinity with un-filmed MCP, optimum operating voltage are: if cathode voltage exceeds a certain value Vc1 with MCP gain almost unchanged, this explains that cathode voltage Vc1 at this time is the best operating voltage; if MCP voltage is of a specific value Vm1 (cathod voltage is any value higher than Vc1) with low MCP gain, and MCP gain turns higher enough when > 20 000 MCP voltage becomes (Vm1+100V), thus this value (Vm1+100V) is the optimum MCP voltage; ②For the same material with filmed MCP, operating voltage are: cathode voltage Vc=algebraic sum of un-filmed MCP cathode voltage (Vc1) and threshold voltage, and MCP voltage is Vm > (Vm1+100V), whose specific value should be determined according to linear region of filmed MCP gain value. This study has an important significance to the determination of best working voltage and performance improvement of filmed MCP on NEA photocathode image intensifier.
2017, 38(3): 485-493.
doi: 10.5768/JAO201738.0306003
Abstract:
In order to reduce sensitivity of binocular pupil detection system to the pupil position, a binocular pupil detector based on near-infrared telecentric lens is designed. Calculation process of optical parameters of near-infrared telecentric lens is described. Telecentric lens is designed and optimized in different parts with Zemax.Tolerance is analyzed by MTF, maximum distortion of the whole field and object telecentricity. Finally, standard circle is used to calibrate the instrument to find out fitting correspondence between number of pixels and pupil size. Results show that pupil diameter measurement accuracy reaches 0.05 mm for interpupillary distance of 55 mm to 75 mm, which can meet the requirements.
In order to reduce sensitivity of binocular pupil detection system to the pupil position, a binocular pupil detector based on near-infrared telecentric lens is designed. Calculation process of optical parameters of near-infrared telecentric lens is described. Telecentric lens is designed and optimized in different parts with Zemax.Tolerance is analyzed by MTF, maximum distortion of the whole field and object telecentricity. Finally, standard circle is used to calibrate the instrument to find out fitting correspondence between number of pixels and pupil size. Results show that pupil diameter measurement accuracy reaches 0.05 mm for interpupillary distance of 55 mm to 75 mm, which can meet the requirements.
2017, 38(3): 494-498.
doi: 10.5768/JAO201738.0307001
Abstract:
In this paper, characteristics and differences of ultrasonic signals generated by point light source and line light source are mainly studied under condition of maintaining the same laser pulse energy and the same detection position through established laser ultrasonic testing platform. Amplitude of ultrasonic signal and waveform of ultrasonic signal generated by point light source and line light source are studied at different detection positions. It is found that amplitude of ultrasonic signal is increased from 1.3 V to 1.7 V, which increased by 0.4 V after point light source is replaced by a line source in the same detection position. When detection position is gradually increased, attenuation amplitude of ultrasonic signal amplitude excited by point light source is 75.78%, while attenuation amplitude of ultrasonic signal amplitude excited by line light source is 34.92%. Results show that under the same power density, ultrasonic signal produced by line light source is increased by 40.86% of ultrasonic signal amplitude generated by point light source, resulting in strong surface wave and better signal to noise ratio.
In this paper, characteristics and differences of ultrasonic signals generated by point light source and line light source are mainly studied under condition of maintaining the same laser pulse energy and the same detection position through established laser ultrasonic testing platform. Amplitude of ultrasonic signal and waveform of ultrasonic signal generated by point light source and line light source are studied at different detection positions. It is found that amplitude of ultrasonic signal is increased from 1.3 V to 1.7 V, which increased by 0.4 V after point light source is replaced by a line source in the same detection position. When detection position is gradually increased, attenuation amplitude of ultrasonic signal amplitude excited by point light source is 75.78%, while attenuation amplitude of ultrasonic signal amplitude excited by line light source is 34.92%. Results show that under the same power density, ultrasonic signal produced by line light source is increased by 40.86% of ultrasonic signal amplitude generated by point light source, resulting in strong surface wave and better signal to noise ratio.
2017, 38(3): 499-505.
doi: 10.5768/JAO201738.0307002
Abstract:
In view of current high noise of domestic 561 nm laser, affecting its practicality, 561 nm yellow laser with high stability and low noise is proposed. Nd:YAG crystal is used to obtain 1 123 nm fundamental frequency light, and output of LBO crystal is 561 nm. Threshold pump power of 1 112 nm, 1 116 nm and 1 123 nm wavelength is analyzed theoretically, and a single wavelength oscillation condition of 1 123 nm is proposed to determine requirements of resonant cavity coating. According to theoretical calculation, a reasonable resonant cavity system is designed, and single-wavelength oscillation of 1 123 nm line is realized by suppressing oscillation of 1 112 nm and 1 116 nm lines in the resonator. At a pump power of 5 W, 561 nm laser single wavelength output is achieved with an output power of 107 mW, power instability of 0.7% and noise of 1.2%.
In view of current high noise of domestic 561 nm laser, affecting its practicality, 561 nm yellow laser with high stability and low noise is proposed. Nd:YAG crystal is used to obtain 1 123 nm fundamental frequency light, and output of LBO crystal is 561 nm. Threshold pump power of 1 112 nm, 1 116 nm and 1 123 nm wavelength is analyzed theoretically, and a single wavelength oscillation condition of 1 123 nm is proposed to determine requirements of resonant cavity coating. According to theoretical calculation, a reasonable resonant cavity system is designed, and single-wavelength oscillation of 1 123 nm line is realized by suppressing oscillation of 1 112 nm and 1 116 nm lines in the resonator. At a pump power of 5 W, 561 nm laser single wavelength output is achieved with an output power of 107 mW, power instability of 0.7% and noise of 1.2%.
2017, 38(3): 506-513.
doi: 10.5768/JAO201738.0308001
Abstract:
In order to improve measuring accuracy of nano coordinate measuring machine probe and meet requirements of accurate measurement for complex surfaces or micro structures, a novel optical fiber Bragg grating (FBG) based micro contact probe system is proposed in this paper. The probe has high sensitivity and repeatability. This new FBG probe has a molten spherical top, and as a strain sensor, FBG is built on the spindle. Basic principle of fiber-optic probe is introduced. The finite element software ANSYS 11 is used to simulate strain distribution of probe in axial and lateral loads. Results show that simulation analysis agrees well with theoretical calculation. Sensitivity and resolution of fiber probe are tested by experiment. Experimental results show that performance of probe is measured by piezoelectric transducer with displacement resolution of 1.5 nm under axial load condition, and measurement resolution of fiber probe is 60 nm. Fiber probe has a high sensitivity and resolution, and its performance can meet requirements of actual measurement.
In order to improve measuring accuracy of nano coordinate measuring machine probe and meet requirements of accurate measurement for complex surfaces or micro structures, a novel optical fiber Bragg grating (FBG) based micro contact probe system is proposed in this paper. The probe has high sensitivity and repeatability. This new FBG probe has a molten spherical top, and as a strain sensor, FBG is built on the spindle. Basic principle of fiber-optic probe is introduced. The finite element software ANSYS 11 is used to simulate strain distribution of probe in axial and lateral loads. Results show that simulation analysis agrees well with theoretical calculation. Sensitivity and resolution of fiber probe are tested by experiment. Experimental results show that performance of probe is measured by piezoelectric transducer with displacement resolution of 1.5 nm under axial load condition, and measurement resolution of fiber probe is 60 nm. Fiber probe has a high sensitivity and resolution, and its performance can meet requirements of actual measurement.
2017, 38(3): 514-520.
doi: 10.5768/JAO201738.0308002
Abstract:
Semiconductor laser beam splitting technology can achieve near-diffraction limit of high-power laser output, and has become the current research focus. Performance of diffraction grating directly determines laser output effect of spectral beam. A sub wavelength transmission grating for 940nm wavelength and fused silica material is simulated and designed. Based on strictly coupled wave theory, grating structure is designed, and parameters such as grating duty cycle, ridge height and period are optimized by Rsoft software. Influence of each parameter on diffraction efficiency of grating is also analyzed simultaneously. Designed diffraction grating achieves wavelength division multiplexing function of -1 diffraction order, diffractive efficiency reaches 91.2% for TE mode, and diffraction efficiency is reduced to 1.2% or less by compressing other diffraction orders. Diffraction efficiency of 90% or more is maintained within range of grating incidence angle of 59°±3°, meanwhile high power laser output has high error tolerance, and is easy to adjust to meet requirements of spectral beam combination technology.
Semiconductor laser beam splitting technology can achieve near-diffraction limit of high-power laser output, and has become the current research focus. Performance of diffraction grating directly determines laser output effect of spectral beam. A sub wavelength transmission grating for 940nm wavelength and fused silica material is simulated and designed. Based on strictly coupled wave theory, grating structure is designed, and parameters such as grating duty cycle, ridge height and period are optimized by Rsoft software. Influence of each parameter on diffraction efficiency of grating is also analyzed simultaneously. Designed diffraction grating achieves wavelength division multiplexing function of -1 diffraction order, diffractive efficiency reaches 91.2% for TE mode, and diffraction efficiency is reduced to 1.2% or less by compressing other diffraction orders. Diffraction efficiency of 90% or more is maintained within range of grating incidence angle of 59°±3°, meanwhile high power laser output has high error tolerance, and is easy to adjust to meet requirements of spectral beam combination technology.
New website launched!
陕公网安备 61011302001501号