2021 Vol. 42, No. 6
Display Method:
2021, 42(6): 949-955.
doi: 10.5768/JAO202142.0601001
Abstract:
The mirror in the high-energy laser beam combining system needs to make a fast and high-precision pointing adjustment before working. A two-dimensional electrically controlled adjusting mirror was designed. The main body of the system adopted an integrated flexible support design, the drive adopted a stepping motor and a deceleration mechanism to drive the screw, the precision measurement of the mirror deflection angle adopted an eddy current sensor, and the digital signal processor (DSP) was adopted as the main control module. The working principle and design of the system were analyzed, and the system calibration method as well as the control algorithm were deeply studied. In order to meet the requirements of system adjustment speed, the S-shaped acceleration and deceleration algorithm was adopted as the control algorithm of adjusting mirror, and the system calibration method with piecewise linearity was adopted. Finally, the system was tested experimentally. The experimental results show that within the angle range of ±500″, the in-position time of adjusting mirror is within 3 s, and the control error is less than 2″, which can meet the system requirements.
The mirror in the high-energy laser beam combining system needs to make a fast and high-precision pointing adjustment before working. A two-dimensional electrically controlled adjusting mirror was designed. The main body of the system adopted an integrated flexible support design, the drive adopted a stepping motor and a deceleration mechanism to drive the screw, the precision measurement of the mirror deflection angle adopted an eddy current sensor, and the digital signal processor (DSP) was adopted as the main control module. The working principle and design of the system were analyzed, and the system calibration method as well as the control algorithm were deeply studied. In order to meet the requirements of system adjustment speed, the S-shaped acceleration and deceleration algorithm was adopted as the control algorithm of adjusting mirror, and the system calibration method with piecewise linearity was adopted. Finally, the system was tested experimentally. The experimental results show that within the angle range of ±500″, the in-position time of adjusting mirror is within 3 s, and the control error is less than 2″, which can meet the system requirements.
2021, 42(6): 956-962.
doi: 10.5768/JAO202142.0601002
Abstract:
When the split detection light screen is used to test projectile velocity of the indoor target trajectory, the sensitivity of detection light screen directly affects the accuracy of initial velocity measurement. The sensitivity distribution of a large-area triangular detection light screen formed by a line light source coupled with a lens-type receiving device was studied. Considering the off-axis effect of optical lens and the attenuation of line light source with the distance, the illuminance at different positions in the detection light screen was equivalent to the line light source. Assuming that the thickness of the light screen was uniform and the projectile length was always greater than the thickness of the screen, the ratio of the area formed by the projectile blocking the detection light screen to the cross-sectional area of the light screen in the current area was equivalent to the ratio of the projectile diameter to the width of the light screen at the current position. The theoretical simulation and live ammunition test verification were carried out on the 4.8 m×2.4 m triangular detection light screen. The simulation analysis results are consistent with the live ammunition test results. The sensitivity is higher near the lens and lower far away from the lens, and the standard deviation of the analog voltage amplitude in the key detection area was 0.05 V, from which the uniformity meets the test requirements. The research results can provide references for the engineering design of the triangular detection light screen.
When the split detection light screen is used to test projectile velocity of the indoor target trajectory, the sensitivity of detection light screen directly affects the accuracy of initial velocity measurement. The sensitivity distribution of a large-area triangular detection light screen formed by a line light source coupled with a lens-type receiving device was studied. Considering the off-axis effect of optical lens and the attenuation of line light source with the distance, the illuminance at different positions in the detection light screen was equivalent to the line light source. Assuming that the thickness of the light screen was uniform and the projectile length was always greater than the thickness of the screen, the ratio of the area formed by the projectile blocking the detection light screen to the cross-sectional area of the light screen in the current area was equivalent to the ratio of the projectile diameter to the width of the light screen at the current position. The theoretical simulation and live ammunition test verification were carried out on the 4.8 m×2.4 m triangular detection light screen. The simulation analysis results are consistent with the live ammunition test results. The sensitivity is higher near the lens and lower far away from the lens, and the standard deviation of the analog voltage amplitude in the key detection area was 0.05 V, from which the uniformity meets the test requirements. The research results can provide references for the engineering design of the triangular detection light screen.
2021, 42(6): 963-968.
doi: 10.5768/JAO202142.0601003
Abstract:
In the process of transmission and interaction with substance, the vector light field with spatiotemporal evolution of light field state is different from the traditional polarized light. It has the unique vectorization characteristics. The regulate and control characteristics of crystal tellurium dioxide (TeO2) for radial polarized vector light field was utilized to study. According to the measurement of Stokes parameters, the evolution laws of polarization state of three crystal samples were analyzed, and the optical rotation characteristics of TeO2 crystal were revealed. It showed that the polarization regulate and control effect could be considered as same as double λ/2 wave plate. The research results are helpful to understand the transmission characteristics of vector light field in birefringent crystals, which can promote the crystal device design and application in connection with vector light field.
In the process of transmission and interaction with substance, the vector light field with spatiotemporal evolution of light field state is different from the traditional polarized light. It has the unique vectorization characteristics. The regulate and control characteristics of crystal tellurium dioxide (TeO2) for radial polarized vector light field was utilized to study. According to the measurement of Stokes parameters, the evolution laws of polarization state of three crystal samples were analyzed, and the optical rotation characteristics of TeO2 crystal were revealed. It showed that the polarization regulate and control effect could be considered as same as double λ/2 wave plate. The research results are helpful to understand the transmission characteristics of vector light field in birefringent crystals, which can promote the crystal device design and application in connection with vector light field.
2021, 42(6): 969-974.
doi: 10.5768/JAO202142.0601004
Abstract:
Aiming at the problem of long time-consumption for automatic focus, by analyzing the change rule of image definition during the focusing process, the effective interval was determined within the focusing range. The drive rate was adjusted according to the slope of the definition change, so that a variable speed control method for focusing mechanism limited interval was improved and realized. Compared with the conventional control methods, the proposed method has the advantages of short stroke, fast speed and high precision. The experimental results show that the shortest focus time of the proposed method is less than 1.5 s, which can be used as a common automatic focus control method for similar optical lenses.
Aiming at the problem of long time-consumption for automatic focus, by analyzing the change rule of image definition during the focusing process, the effective interval was determined within the focusing range. The drive rate was adjusted according to the slope of the definition change, so that a variable speed control method for focusing mechanism limited interval was improved and realized. Compared with the conventional control methods, the proposed method has the advantages of short stroke, fast speed and high precision. The experimental results show that the shortest focus time of the proposed method is less than 1.5 s, which can be used as a common automatic focus control method for similar optical lenses.
2021, 42(6): 975-981.
doi: 10.5768/JAO202142.0601005
Abstract:
The accuracy of stability control is an important indicator of photoelectric pod system. In order to reduce the effect of gyro drift on stability accuracy in stabilized sighting control, the gyro drift needs to be compensated. A method of drift measurement and compensation was introduced for the photoelectric pod based on the navigation system, which was designed to calculate the component of earth rotation in the platform by the position and attitude data of airborne navigation system and pod angle during drift measurement phase of photoelectric pod, so as to obtain a more accurate gyro error. And in the stable phase, the gyro error and earth rotation component were compensated respectively by the navigation system. This method could effectively separate the earth rotation component and gyro error in the gyro drift measurement stage and adjust earth rotation component in the stable stage, so that the stability precision of stabilized sighting system could be improved. The experimental results show that, by comparing the results of 10-minute traditional drift measurement and drift measurement based on the navigation system, the accumulated drift error is stably controlled, the pitch angle is reduced from 1.80° in traditional method to 0.04°, and the heading angle is reduced from 0.77° to 0.04°.
The accuracy of stability control is an important indicator of photoelectric pod system. In order to reduce the effect of gyro drift on stability accuracy in stabilized sighting control, the gyro drift needs to be compensated. A method of drift measurement and compensation was introduced for the photoelectric pod based on the navigation system, which was designed to calculate the component of earth rotation in the platform by the position and attitude data of airborne navigation system and pod angle during drift measurement phase of photoelectric pod, so as to obtain a more accurate gyro error. And in the stable phase, the gyro error and earth rotation component were compensated respectively by the navigation system. This method could effectively separate the earth rotation component and gyro error in the gyro drift measurement stage and adjust earth rotation component in the stable stage, so that the stability precision of stabilized sighting system could be improved. The experimental results show that, by comparing the results of 10-minute traditional drift measurement and drift measurement based on the navigation system, the accumulated drift error is stably controlled, the pitch angle is reduced from 1.80° in traditional method to 0.04°, and the heading angle is reduced from 0.77° to 0.04°.
2021, 42(6): 982-988.
doi: 10.5768/JAO202142.0601006
Abstract:
Aiming at the difficulty in extracting the electromagnetic parameters of highly absorbing samples by using the transmission measurement system, the time-domain spectrum of polar and non-polar liquids were measured by the reflection time-domain spectroscopy system. The propagation characteristics of terahertz wave in the sample were modeled by finite integration time domain algorithm, and the reflection spectrum was reconstructed by using the Gaussian mixture model to restore the original terahertz time-domain signal from the noise, which solved the false fluctuations in electromagnetic parameters caused by the interference of signal in reflection system. The experimental results show that compared with the results calculated by using the original signal, the refractive index and extinction coefficient of water and liquid paraffin in the 0.4 THz~2 THz band calculated by using the restored signal, effectively eliminate the false fluctuations, and the detrended fluctuation analysis (DFA) scale index is increased by 7%, 3%, 29%, and 31%, respectively, which verifies the validity and correctness of the method.
Aiming at the difficulty in extracting the electromagnetic parameters of highly absorbing samples by using the transmission measurement system, the time-domain spectrum of polar and non-polar liquids were measured by the reflection time-domain spectroscopy system. The propagation characteristics of terahertz wave in the sample were modeled by finite integration time domain algorithm, and the reflection spectrum was reconstructed by using the Gaussian mixture model to restore the original terahertz time-domain signal from the noise, which solved the false fluctuations in electromagnetic parameters caused by the interference of signal in reflection system. The experimental results show that compared with the results calculated by using the original signal, the refractive index and extinction coefficient of water and liquid paraffin in the 0.4 THz~2 THz band calculated by using the restored signal, effectively eliminate the false fluctuations, and the detrended fluctuation analysis (DFA) scale index is increased by 7%, 3%, 29%, and 31%, respectively, which verifies the validity and correctness of the method.
2021, 42(6): 989-996.
doi: 10.5768/JAO202142.0601007
Abstract:
Aiming at the instability problem caused by the nonlinear constraint between the driving axis and the sensitive axis of the azimuth line-of-sight (LOS) control with the two-dimensional gyro platform, the variation rules of influence of azimuth line-of-sight (LOS) by different types of disturbance sources were analyzed, and a passing zenith stabilization method based on the disturbance sources classification control was proposed. This method used the feedback and feedforward dual channel compound control structure, eliminated the endogenous torque disturbance caused by gyro measurement noise amplification in the zenith position based on the control structure itself, and the torque disturbance caused by the high frequency component of the roll disturbance was suppressed by adding the filtering link of the feedforward channel, then the vibration problem of the azimuth drive axis in zenith position was solved. At the same time, the azimuth and the low frequency component of the roll disturbance were effectively isolated by the feedback channel and the feedforward channel. The simulation results show that this method can effectively suppress the amplitude of azimuth motor torque disturbance caused by the gyro measurement noise and high frequency component of the roll disturbance, and enhance the system stability. Finally, an experiment was carried out on a two-dimensional gyro platform. The stabilization precision of the line-of-sight bearing subjected to vibration in zenith position was reduced from 82.4 μrad to 44.6 μrad, and the isolation ratio was increased from −14.54 dB to −27.85 dB under the swing condition. The experimental results show that this method can effectively improve the disturbance isolation performance of azimuth line-of-sight in zenith position.
Aiming at the instability problem caused by the nonlinear constraint between the driving axis and the sensitive axis of the azimuth line-of-sight (LOS) control with the two-dimensional gyro platform, the variation rules of influence of azimuth line-of-sight (LOS) by different types of disturbance sources were analyzed, and a passing zenith stabilization method based on the disturbance sources classification control was proposed. This method used the feedback and feedforward dual channel compound control structure, eliminated the endogenous torque disturbance caused by gyro measurement noise amplification in the zenith position based on the control structure itself, and the torque disturbance caused by the high frequency component of the roll disturbance was suppressed by adding the filtering link of the feedforward channel, then the vibration problem of the azimuth drive axis in zenith position was solved. At the same time, the azimuth and the low frequency component of the roll disturbance were effectively isolated by the feedback channel and the feedforward channel. The simulation results show that this method can effectively suppress the amplitude of azimuth motor torque disturbance caused by the gyro measurement noise and high frequency component of the roll disturbance, and enhance the system stability. Finally, an experiment was carried out on a two-dimensional gyro platform. The stabilization precision of the line-of-sight bearing subjected to vibration in zenith position was reduced from 82.4 μrad to 44.6 μrad, and the isolation ratio was increased from −14.54 dB to −27.85 dB under the swing condition. The experimental results show that this method can effectively improve the disturbance isolation performance of azimuth line-of-sight in zenith position.
2021, 42(6): 997-1005.
doi: 10.5768/JAO202142.0601008
Abstract:
According to the requirement of stabilization precision of 0.2 mrad for artillery aiming line in upper anti-stabilization aiming system, a multi-closed control loop was formed by adding high gain acceleration closed-loop correction in traditional inertia rate stable closed-loop, and the PID parameters of the system were adjusted by integral of time-weight absolute error (ITAE) optimal control. The structure of the upper anti-stabilization aiming system was introduced and analyzed, and the closed-loop control system was simulated, such as load, gyro and brushless torque motor. The ITAE optimal controller was used to adjust the PID control parameters of the multi-closed-loop control system. Added random interference and unit step response to the system in order to test the related performance. The test results show that compared with the traditional adjustment of PID parameters and single-speed-loop control system, the multi-closed-loop ITAE optimal controller based on acceleration can improve the system anti-disturbance performance by 78%, the overshoot is reduced by about 23%, and the swing stability accuracy is improved by about 29%, which can better meet the stability requirements of the upper anti-stabilization aiming system.
According to the requirement of stabilization precision of 0.2 mrad for artillery aiming line in upper anti-stabilization aiming system, a multi-closed control loop was formed by adding high gain acceleration closed-loop correction in traditional inertia rate stable closed-loop, and the PID parameters of the system were adjusted by integral of time-weight absolute error (ITAE) optimal control. The structure of the upper anti-stabilization aiming system was introduced and analyzed, and the closed-loop control system was simulated, such as load, gyro and brushless torque motor. The ITAE optimal controller was used to adjust the PID control parameters of the multi-closed-loop control system. Added random interference and unit step response to the system in order to test the related performance. The test results show that compared with the traditional adjustment of PID parameters and single-speed-loop control system, the multi-closed-loop ITAE optimal controller based on acceleration can improve the system anti-disturbance performance by 78%, the overshoot is reduced by about 23%, and the swing stability accuracy is improved by about 29%, which can better meet the stability requirements of the upper anti-stabilization aiming system.
2021, 42(6): 1006-1010.
doi: 10.5768/JAO202142.0601009
Abstract:
According to the use requirements of closed circuit television (CCTV) lens, a dual band CCTV fish-eye lens based on the principle of non-similar imaging was designed. This system works in the wavelength range of 480 nm~850 nm, with the dual band imaging of visible light and near-infrared light, which could realize day and night monitoring. The F-number is 1.8, the field angle is 180º, the focal length is 1 mm, and the total optical length is 7.76 mm. The lens has the characteristics of large relative aperture, large field angle and miniaturization. The system adopts 7 groups of 9-piece anti long distance structure, without special glass and aspheric surface, which greatly reduced the complexity of the system and the manufacturing cost. The optical design software Zemax was used to design its optical system, selecting the 1/3 inch CCD as the detector. When the Nyquist frequency is 120 lp/mm, the values of meridional modulation transfer function curves and the sagittal modulation transfer function curves of each field of view are more than 0.5, close to the diffraction limit, and the imaging quality is good. The field curvature of the whole field is less than 1 mm, the relative distortion is less than 25%, and the relative illumination is about 95%, which meet the use requirements of CCTV lens, and can be widely used in monitoring and reconnaissance fields.
According to the use requirements of closed circuit television (CCTV) lens, a dual band CCTV fish-eye lens based on the principle of non-similar imaging was designed. This system works in the wavelength range of 480 nm~850 nm, with the dual band imaging of visible light and near-infrared light, which could realize day and night monitoring. The F-number is 1.8, the field angle is 180º, the focal length is 1 mm, and the total optical length is 7.76 mm. The lens has the characteristics of large relative aperture, large field angle and miniaturization. The system adopts 7 groups of 9-piece anti long distance structure, without special glass and aspheric surface, which greatly reduced the complexity of the system and the manufacturing cost. The optical design software Zemax was used to design its optical system, selecting the 1/3 inch CCD as the detector. When the Nyquist frequency is 120 lp/mm, the values of meridional modulation transfer function curves and the sagittal modulation transfer function curves of each field of view are more than 0.5, close to the diffraction limit, and the imaging quality is good. The field curvature of the whole field is less than 1 mm, the relative distortion is less than 25%, and the relative illumination is about 95%, which meet the use requirements of CCTV lens, and can be widely used in monitoring and reconnaissance fields.
2021, 42(6): 1011-1016.
doi: 10.5768/JAO202142.0602001
Abstract:
In order to achieve the parallel, non-contacting and real-time measurement of filament diameter, a real-time and fast detection system of arrayed filament diameter was designed based on the complex programmable logic device (CPLD). Arraying the sensing system and solving diameters by CPLD, the multiple filaments were measured in parallel. The load of the data transmission of system was significantly reduced, the solving time of diameter was shortened, the detection frequency of diameter was increased to 5 kHz and the parallel measurement of 24 arrayed filaments was realized. The repeated accuracy of the dynamic measurement of system is 3.5 μm and 8.3 μm within the diameter range of 0.1 mm~0.3 mm and 0.3 mm~1.0 mm, respectively, and the absolute value of percentage of repeatability error in dynamic measurement is less than 1.21% and 1.43%, respectively, which meets the requirements of real-time and fast detection of arrayed filament diameter.
In order to achieve the parallel, non-contacting and real-time measurement of filament diameter, a real-time and fast detection system of arrayed filament diameter was designed based on the complex programmable logic device (CPLD). Arraying the sensing system and solving diameters by CPLD, the multiple filaments were measured in parallel. The load of the data transmission of system was significantly reduced, the solving time of diameter was shortened, the detection frequency of diameter was increased to 5 kHz and the parallel measurement of 24 arrayed filaments was realized. The repeated accuracy of the dynamic measurement of system is 3.5 μm and 8.3 μm within the diameter range of 0.1 mm~0.3 mm and 0.3 mm~1.0 mm, respectively, and the absolute value of percentage of repeatability error in dynamic measurement is less than 1.21% and 1.43%, respectively, which meets the requirements of real-time and fast detection of arrayed filament diameter.
2021, 42(6): 1017-1024.
doi: 10.5768/JAO202142.0602002
Abstract:
Due to the influence of atmospheric turbulence, the images obtained by the remote optical imaging equipment will appear serious degradation phenomena, such as geometric distortion, motion blur and defocus blur, etc. At present, the mainstream turbulence restoration algorithms usually rely on the non-rigid registration, reconstruction or finding rare lucky regions from long-term video sequences. These methods require huge calculated amount or complete video data in advance, which cannot meet the real-time requirements of actual application scenarios. Therefore, a real-time turbulence restoration algorithm that could be implemented in field programmable gate array (FPGA) was proposed. This method used the randomness of atmospheric turbulence to filter the image in the time domain through the continuous frames and solved the problem of image geometric distortion. Then, the Wiener filter in the frequency domain was converted into a convolution in the spatial domain, which was easier to implement, and solved the problem of image blur. The experimental results show that the proposed algorithm not only meets the real-time requirements, but also effectively realizes the restoration of turbulence images.
Due to the influence of atmospheric turbulence, the images obtained by the remote optical imaging equipment will appear serious degradation phenomena, such as geometric distortion, motion blur and defocus blur, etc. At present, the mainstream turbulence restoration algorithms usually rely on the non-rigid registration, reconstruction or finding rare lucky regions from long-term video sequences. These methods require huge calculated amount or complete video data in advance, which cannot meet the real-time requirements of actual application scenarios. Therefore, a real-time turbulence restoration algorithm that could be implemented in field programmable gate array (FPGA) was proposed. This method used the randomness of atmospheric turbulence to filter the image in the time domain through the continuous frames and solved the problem of image geometric distortion. Then, the Wiener filter in the frequency domain was converted into a convolution in the spatial domain, which was easier to implement, and solved the problem of image blur. The experimental results show that the proposed algorithm not only meets the real-time requirements, but also effectively realizes the restoration of turbulence images.
2021, 42(6): 1025-1033.
doi: 10.5768/JAO202142.0602003
Abstract:
Aiming at the problems that the current image segmentation algorithm has low accuracy in realizing the internal defect segmentation of industrial castings and the algorithm is not lightweight enough, an improved industrial casting internal defect detection algorithm Effi-DeepLab based on DeepLabv3+ was proposed. This method used MBConv in EfficientNet to replace the original Xception module for feature extraction, making the feature extraction network more efficient and lighter. Aiming at the problem of small internal defects in industrial castings, the expansion rate of the hollow convolution in the atrous spatial pyramid pooling (ASPP) layer was redesigned. The convolution block was more robust to small targets; the low-order semantic information in the feature extraction stage was fully utilized at the decoding end to perform multi-scale feature fusion to improve the accuracy of small target defect segmentation. The experimental results show that the segmentation accuracy and mIoU of the Effi-DeepLab model in the internal defect image data set of the automobile wheel are 93.58% and 89.39%, respectively, which are improved by 2.65% and 2.24%, respectively, compared with DeepLabv3+, and has better segmentation effect. In addition, it is experimentally verified that the proposed algorithm has good generalization.
Aiming at the problems that the current image segmentation algorithm has low accuracy in realizing the internal defect segmentation of industrial castings and the algorithm is not lightweight enough, an improved industrial casting internal defect detection algorithm Effi-DeepLab based on DeepLabv3+ was proposed. This method used MBConv in EfficientNet to replace the original Xception module for feature extraction, making the feature extraction network more efficient and lighter. Aiming at the problem of small internal defects in industrial castings, the expansion rate of the hollow convolution in the atrous spatial pyramid pooling (ASPP) layer was redesigned. The convolution block was more robust to small targets; the low-order semantic information in the feature extraction stage was fully utilized at the decoding end to perform multi-scale feature fusion to improve the accuracy of small target defect segmentation. The experimental results show that the segmentation accuracy and mIoU of the Effi-DeepLab model in the internal defect image data set of the automobile wheel are 93.58% and 89.39%, respectively, which are improved by 2.65% and 2.24%, respectively, compared with DeepLabv3+, and has better segmentation effect. In addition, it is experimentally verified that the proposed algorithm has good generalization.
2021, 42(6): 1034-1039.
doi: 10.5768/JAO202142.0602004
Abstract:
In order to improve the detection and identification ability of the photoelectric system for the target, and achieve the fusion processing of single-photon detection of 3D point cloud data and 2D visible light images, the fusion processing algorithm of single-photon detection imaging system was proposed. A direct linear transformation method was used, and the parameter calibration problems in the process of fusion processing were solved by selection of homonymic feature points and indirect adjustment algorithm. The fusion processing algorithm was verified by the experimental data, and the pixel-level fusion processing with 1 024×768 resolution single-photon detection of 3D point cloud data and 2D visible light images was achieved. The experimental results show that the proposed fusion processing algorithm can effectively achieve the fusion of 3D and 2D images.
In order to improve the detection and identification ability of the photoelectric system for the target, and achieve the fusion processing of single-photon detection of 3D point cloud data and 2D visible light images, the fusion processing algorithm of single-photon detection imaging system was proposed. A direct linear transformation method was used, and the parameter calibration problems in the process of fusion processing were solved by selection of homonymic feature points and indirect adjustment algorithm. The fusion processing algorithm was verified by the experimental data, and the pixel-level fusion processing with 1 024×768 resolution single-photon detection of 3D point cloud data and 2D visible light images was achieved. The experimental results show that the proposed fusion processing algorithm can effectively achieve the fusion of 3D and 2D images.
2021, 42(6): 1040-1047.
doi: 10.5768/JAO202142.0602005
Abstract:
In order to accurately adjust the brightness of the overlapping area of multi-channel projection images, a projection image edge fusion method based on B-spline curve was designed. Firstly, the parameters of attenuation function and gamma correction were set according to the projection screen, external light environment and the nature of the projector itself. Then, the overlap area of the projection image was acquired by the camera and the color intensity of each pixel was calculated, and the peak signal-to-noise ratio was calculated by the ideal value. Then, the correction accuracy and flexibility of quadratic B-spline curve and cubic B-spline curve were compared. Finally, the edge brightness of the projection image was optimized by cubic B-spline until the peak signal-to-noise ratio reached the standard value. The experimental results show that the peak signal-to-noise ratio of the channel R in the overlapped area of the projection image is increased by 9.13 dB, that of the channel G by 6.09 dB, and that of the channel B by 7.53 dB. The edge brightness optimization technology of projection image based on B-spline curve improves the accuracy of brightness adjustment in overlapping area of projection image.
In order to accurately adjust the brightness of the overlapping area of multi-channel projection images, a projection image edge fusion method based on B-spline curve was designed. Firstly, the parameters of attenuation function and gamma correction were set according to the projection screen, external light environment and the nature of the projector itself. Then, the overlap area of the projection image was acquired by the camera and the color intensity of each pixel was calculated, and the peak signal-to-noise ratio was calculated by the ideal value. Then, the correction accuracy and flexibility of quadratic B-spline curve and cubic B-spline curve were compared. Finally, the edge brightness of the projection image was optimized by cubic B-spline until the peak signal-to-noise ratio reached the standard value. The experimental results show that the peak signal-to-noise ratio of the channel R in the overlapped area of the projection image is increased by 9.13 dB, that of the channel G by 6.09 dB, and that of the channel B by 7.53 dB. The edge brightness optimization technology of projection image based on B-spline curve improves the accuracy of brightness adjustment in overlapping area of projection image.
2021, 42(6): 1048-1055.
doi: 10.5768/JAO202142.0602006
Abstract:
Aiming at the problem that common image feature matching algorithms will produce a large number of mismatched points in the image feature matching stage for images with parallax, a image feature matching algorithm combining accelerated-KAZE (AKAZE) algorithm with adaptive local affine matching was proposed. Firstly, the AKAZE algorithm was used to extract the feature points. Then, the binary descriptor M-LDB was used for description and the brute force matching was performed to generate coarse matching point pairs. Finally, the image-based affine transformation could provide the characteristic of strong geometric constraints, and adopted adaptive local affine matching to complete fine matching. The experimental results show that the algorithm has the effects of uniform extraction of feature points and accurate matching for image matching with rotation change, scale change, and viewing angle change at the same time. The number of correct feature points extracted is increased by average 1.66 times compared with SIFT algorithm, 1.08 times compared with SURF algorithm, 6.92 times compared with ORB algorithm and 1.23 times compared with GMS algorithm, respectively. It can meet the needs of image matching with large disparity.
Aiming at the problem that common image feature matching algorithms will produce a large number of mismatched points in the image feature matching stage for images with parallax, a image feature matching algorithm combining accelerated-KAZE (AKAZE) algorithm with adaptive local affine matching was proposed. Firstly, the AKAZE algorithm was used to extract the feature points. Then, the binary descriptor M-LDB was used for description and the brute force matching was performed to generate coarse matching point pairs. Finally, the image-based affine transformation could provide the characteristic of strong geometric constraints, and adopted adaptive local affine matching to complete fine matching. The experimental results show that the algorithm has the effects of uniform extraction of feature points and accurate matching for image matching with rotation change, scale change, and viewing angle change at the same time. The number of correct feature points extracted is increased by average 1.66 times compared with SIFT algorithm, 1.08 times compared with SURF algorithm, 6.92 times compared with ORB algorithm and 1.23 times compared with GMS algorithm, respectively. It can meet the needs of image matching with large disparity.
2021, 42(6): 1056-1061.
doi: 10.5768/JAO202142.0602007
Abstract:
To improve the detection accuracy for armored targets in complex environment, a salient target detection algorithm was proposed. The low-level features and multi-scale pooling semantic features constrained by visual attention mechanism were respectively obtained by visual attention mechanism and joint pyramid upsampling module. Then the aggregation strategy was used to fuse, so as to improve the ability of target representation in low contrast or occlusion conditions. The experimental results show that the proposed algorithm obtains good detection results for multi-scale targets in complex scenes, the precision, recall rate and mean average precision (mAP) are 72.2%, 71.4% and 77.1%, respectively, which can meet the practical application requirements.
To improve the detection accuracy for armored targets in complex environment, a salient target detection algorithm was proposed. The low-level features and multi-scale pooling semantic features constrained by visual attention mechanism were respectively obtained by visual attention mechanism and joint pyramid upsampling module. Then the aggregation strategy was used to fuse, so as to improve the ability of target representation in low contrast or occlusion conditions. The experimental results show that the proposed algorithm obtains good detection results for multi-scale targets in complex scenes, the precision, recall rate and mean average precision (mAP) are 72.2%, 71.4% and 77.1%, respectively, which can meet the practical application requirements.
2021, 42(6): 1062-1066.
doi: 10.5768/JAO202142.0603001
Abstract:
During the flight test, the high-speed photogrammetry stations are characterized by large number, scattered distribution, and separation of man and machine. In order to obtain the real-time measurement data of multiple targets in two separate areas of the take-off and landing section at the same time, a distributed high-speed photogrammetry network system was proposed based on the multicast concept, and an efficient uncorrelated signal high-speed camera trigger criterion algorithm was designed. Through the real-time acquisition of a variety of airborne and moving platform data, and through the image processing methods to complete the real-time precise intelligent control and the real-time unloading of massive data of multidimensional and stereoscopic high-speed photogrammetry system, the intelligent acquisition of test images from 8 different types of high-speed cameras was realized. The false trigger rate of the measurement system is reduced to 0.76%, which effectively improves the work efficiency and ensures the accuracy and efficiency of flight test data.
During the flight test, the high-speed photogrammetry stations are characterized by large number, scattered distribution, and separation of man and machine. In order to obtain the real-time measurement data of multiple targets in two separate areas of the take-off and landing section at the same time, a distributed high-speed photogrammetry network system was proposed based on the multicast concept, and an efficient uncorrelated signal high-speed camera trigger criterion algorithm was designed. Through the real-time acquisition of a variety of airborne and moving platform data, and through the image processing methods to complete the real-time precise intelligent control and the real-time unloading of massive data of multidimensional and stereoscopic high-speed photogrammetry system, the intelligent acquisition of test images from 8 different types of high-speed cameras was realized. The false trigger rate of the measurement system is reduced to 0.76%, which effectively improves the work efficiency and ensures the accuracy and efficiency of flight test data.
2021, 42(6): 1067-1071.
doi: 10.5768/JAO202142.0603002
Abstract:
Ammonia is one of the common pollutants in the atmosphere, and the temperature has a significant impact on ammonia concentration detection results. In order to obtain the accurate concentration of ammonia, the influence of the temperature must be eliminated. The genetic algorithm was combined with differential optical absorption spectroscopy to explore the influence of temperature on ammonia concentration detection. The ammonia detection system at 296 K~328 K was developed based on ultraviolet differential optical absorption spectroscopy, and the compensated detection results were obtained by using genetic algorithm. The experimental results show that the temperature compensation model can effectively eliminate the non-linear influence of temperature on ammonia detection and improve the detection accuracy. The error of 44×10−6 ammonia detection results at 328 K is reduced by 26.97%. With the change of temperature, the linear correlation coefficients are above 0.998 16. For 6×10−6 ammonia, the detection limit of the system is 0.198×10−6 and 0.278×10−6, respectively, before and after temperature compensation.
Ammonia is one of the common pollutants in the atmosphere, and the temperature has a significant impact on ammonia concentration detection results. In order to obtain the accurate concentration of ammonia, the influence of the temperature must be eliminated. The genetic algorithm was combined with differential optical absorption spectroscopy to explore the influence of temperature on ammonia concentration detection. The ammonia detection system at 296 K~328 K was developed based on ultraviolet differential optical absorption spectroscopy, and the compensated detection results were obtained by using genetic algorithm. The experimental results show that the temperature compensation model can effectively eliminate the non-linear influence of temperature on ammonia detection and improve the detection accuracy. The error of 44×10−6 ammonia detection results at 328 K is reduced by 26.97%. With the change of temperature, the linear correlation coefficients are above 0.998 16. For 6×10−6 ammonia, the detection limit of the system is 0.198×10−6 and 0.278×10−6, respectively, before and after temperature compensation.
2021, 42(6): 1072-1079.
doi: 10.5768/JAO202142.0603003
Abstract:
Aiming at the photoelectric test requirements of the ground minimum control speed flight test, a photoelectric test method based on the combination of airborne photogrammetry and ground photoelectric tracking measurement was proposed. Firstly, the design of photoelectric test system for ground minimum control speed flight test was briefly introduced. Secondly, the methods and principles of system calibration, automatic ground tracking target detection, aircraft eccentricity measurement based on airborne photography and aircraft track and speed measurement based on photoelectric tracking measurement were discussed in detail. Then, the implementation process of flight test was summarized. Finally, the flight test verification was carried out. The test results show that the method is effective and reliable. It can provide an intuitive video of the aircraft side deviation process, and realize the positioning and measurement accuracy of ground static verification with aircraft deviation from runway direction better than 2 mm, elevation direction better than 2 cm and heading better than 5 cm. It fully meets the requirements for data integrity, intuitiveness, accuracy and reliability of the photoelectric test for the ground minimum control speed flight test.
Aiming at the photoelectric test requirements of the ground minimum control speed flight test, a photoelectric test method based on the combination of airborne photogrammetry and ground photoelectric tracking measurement was proposed. Firstly, the design of photoelectric test system for ground minimum control speed flight test was briefly introduced. Secondly, the methods and principles of system calibration, automatic ground tracking target detection, aircraft eccentricity measurement based on airborne photography and aircraft track and speed measurement based on photoelectric tracking measurement were discussed in detail. Then, the implementation process of flight test was summarized. Finally, the flight test verification was carried out. The test results show that the method is effective and reliable. It can provide an intuitive video of the aircraft side deviation process, and realize the positioning and measurement accuracy of ground static verification with aircraft deviation from runway direction better than 2 mm, elevation direction better than 2 cm and heading better than 5 cm. It fully meets the requirements for data integrity, intuitiveness, accuracy and reliability of the photoelectric test for the ground minimum control speed flight test.
2021, 42(6): 1080-1085.
doi: 10.5768/JAO202142.0603004
Abstract:
The residual polarization of optical lens is a key parameter that affects the accuracy of polarization remote sensing detection, polarization military target recognition and precision optical instruments measurement. In order to solve the problem of accurate measurement of polarization degree of large aperture optical lens in visible light band, a measuring device of residual polarization degree of optical lens was studied. The large aperture off-axis parabolic mirror with a surface metal film was used as the collimating light source, by controlling the incident angle of the incident light and reducing the residual polarization degree of the incident light, the consistent Stokes parameters were obtained in the horizontal, vertical, 45° and 135° directions. In order to verify the performance of the device, the residual polarization of the optical lens whose aperture was less than 160 mm in the visible light band were measured by the device. The measuring results show that the residual polarization of the measuring device is less than 0.2%, which can meet the needs of high-precision measurement of residual polarization of the optical lens.
The residual polarization of optical lens is a key parameter that affects the accuracy of polarization remote sensing detection, polarization military target recognition and precision optical instruments measurement. In order to solve the problem of accurate measurement of polarization degree of large aperture optical lens in visible light band, a measuring device of residual polarization degree of optical lens was studied. The large aperture off-axis parabolic mirror with a surface metal film was used as the collimating light source, by controlling the incident angle of the incident light and reducing the residual polarization degree of the incident light, the consistent Stokes parameters were obtained in the horizontal, vertical, 45° and 135° directions. In order to verify the performance of the device, the residual polarization of the optical lens whose aperture was less than 160 mm in the visible light band were measured by the device. The measuring results show that the residual polarization of the measuring device is less than 0.2%, which can meet the needs of high-precision measurement of residual polarization of the optical lens.
2021, 42(6): 1086-1091.
doi: 10.5768/JAO202142.0603005
Abstract:
The micro-electro-mechanical system (MEMS) acoustic film has extremely high requirements for tape-out, storage and packaging environments, and its surface defects will affect the quality and performance of MEMS devices. The image defects detection is an effective non-contact optical detection means that can effectively improve the yield rate of MEMS production. However, the periodic structure texture of the MEMS devices surface will interfere with defect detection. A acoustic film defect detection algorithm based on frequency domain transformation was proposed. By calculating the gradient distribution of spectrogram and establishing the Boolean mask, the dominant frequency components corresponding to the periodic structure texture were eliminated. The residual spectrograms were subjected to a Fourier inversion to reconstruct the defect images. The reconstructed images were decomposed by single-layer Haar wavelet to obtain the low-frequency sub-band image and the defect information was extracted by simple threshold segmentation. The defect detection effects of different types of MEMS acoustic film were showed. The experimental results show that it is reasonable to set the zoom constant in the range of 0.7~1.0.
The micro-electro-mechanical system (MEMS) acoustic film has extremely high requirements for tape-out, storage and packaging environments, and its surface defects will affect the quality and performance of MEMS devices. The image defects detection is an effective non-contact optical detection means that can effectively improve the yield rate of MEMS production. However, the periodic structure texture of the MEMS devices surface will interfere with defect detection. A acoustic film defect detection algorithm based on frequency domain transformation was proposed. By calculating the gradient distribution of spectrogram and establishing the Boolean mask, the dominant frequency components corresponding to the periodic structure texture were eliminated. The residual spectrograms were subjected to a Fourier inversion to reconstruct the defect images. The reconstructed images were decomposed by single-layer Haar wavelet to obtain the low-frequency sub-band image and the defect information was extracted by simple threshold segmentation. The defect detection effects of different types of MEMS acoustic film were showed. The experimental results show that it is reasonable to set the zoom constant in the range of 0.7~1.0.
2021, 42(6): 1092-1101.
doi: 10.5768/JAO202142.0604001
Abstract:
With the continuous development of low-level-light (LLL) night vision technology, the LLL night vision devices show the characteristics of many varieties, large performance differences, and wide application fields. How to sort out the main line of development from numerous LLL night vision devices has become a research hot spot for practitioners of low-light night vision devices. The development history of the zero-generation, first-generation, second-generation, third-generation, and ultra-second generation LLL night vision devices was reviewed. The industry-recognized methods for distinguishing the zero-generation, first-generation, second-generation, ultra-second generation, and third-generation low-light night vision devices were summarized. The concept of the fourth-generation low-light night vision device was put forward, and at the same time the position of several atypical low-light night vision devices in the technical field of low-light night vision devices was explained, and a tree diagram was drew, so that the technical characteristics of each low-light night vision device and their position in the field of low-light night vision device technology could be intuitively, accurately, and comprehensively understood. The proposed perspectives have a certain reference role for researchers engaged in low-light night vision device technology.
With the continuous development of low-level-light (LLL) night vision technology, the LLL night vision devices show the characteristics of many varieties, large performance differences, and wide application fields. How to sort out the main line of development from numerous LLL night vision devices has become a research hot spot for practitioners of low-light night vision devices. The development history of the zero-generation, first-generation, second-generation, third-generation, and ultra-second generation LLL night vision devices was reviewed. The industry-recognized methods for distinguishing the zero-generation, first-generation, second-generation, ultra-second generation, and third-generation low-light night vision devices were summarized. The concept of the fourth-generation low-light night vision device was put forward, and at the same time the position of several atypical low-light night vision devices in the technical field of low-light night vision devices was explained, and a tree diagram was drew, so that the technical characteristics of each low-light night vision device and their position in the field of low-light night vision device technology could be intuitively, accurately, and comprehensively understood. The proposed perspectives have a certain reference role for researchers engaged in low-light night vision device technology.
2021, 42(6): 1102-1106.
doi: 10.5768/JAO202142.0604002
Abstract:
Metalens is an imaging device based on metasurfaces, which can accurately regulate the phase, polarization, and amplitude of light. It has advantages of lightweight, easy-integration and planarizartion. For miniaturization and integration of near-infrared imaging devices, a near-infrared polarized insensitive metalens based on amorphous silicon was designed and fabricated. For selecting basic structures, the focusing efficiency of metalenses designed by 8-order and 6-order primitives were compared. To improve the processing tolerance and reduce the processing difficulty, the 8-order primitive was used to design metalens, and the maximum depth-width ratio was 6. For large-span between simulation and processing of metalens, the 600 nm-thick polysilicon by chemical vapor deposition was grew. The metalens was made by electron beam etching. The measured results show that the metalens has a good appearance and the focusing efficiency is 65%.
Metalens is an imaging device based on metasurfaces, which can accurately regulate the phase, polarization, and amplitude of light. It has advantages of lightweight, easy-integration and planarizartion. For miniaturization and integration of near-infrared imaging devices, a near-infrared polarized insensitive metalens based on amorphous silicon was designed and fabricated. For selecting basic structures, the focusing efficiency of metalenses designed by 8-order and 6-order primitives were compared. To improve the processing tolerance and reduce the processing difficulty, the 8-order primitive was used to design metalens, and the maximum depth-width ratio was 6. For large-span between simulation and processing of metalens, the 600 nm-thick polysilicon by chemical vapor deposition was grew. The metalens was made by electron beam etching. The measured results show that the metalens has a good appearance and the focusing efficiency is 65%.
2021, 42(6): 1107-1114.
doi: 10.5768/JAO202142.0605001
Abstract:
Aiming at the characteristics of hyper-spectral imaging, a scheme for detecting tiny cameras based on three-dimensional features was proposed. The results were accurately determined in the spectral dimension after the suspected target was screened out by the cat's eye effect in the space dimension. According to the camera structure, the reflection spectral characteristics of the visible light camera were analyzed. Based on the geometric optics and radiometry, the detection distance of the system was calculated and simulated. The results show that if the equipment is working properly, the optical power affects the minimum detection distance, while the target size affects the maximum detection distance. A tiny-camera spectral feature verification system was built. The results show that the non-reflective light ratio curve of the target using the absorptive infrared cut-off filter changes smoothly and the value is high. As for the curve of the target using the reflective infrared cut-off filter, the value is high in visible light part while it is low in infrared part, and the curve begins to drop near 700 nm and even mutates where the absolute slope value is more than 10 times as it in the infrared band according to the experimental data. The experimental results are consistent with the expected results of the analysis, which proves the feasibility of hyper-spectral imaging technology to detect tiny cameras.
Aiming at the characteristics of hyper-spectral imaging, a scheme for detecting tiny cameras based on three-dimensional features was proposed. The results were accurately determined in the spectral dimension after the suspected target was screened out by the cat's eye effect in the space dimension. According to the camera structure, the reflection spectral characteristics of the visible light camera were analyzed. Based on the geometric optics and radiometry, the detection distance of the system was calculated and simulated. The results show that if the equipment is working properly, the optical power affects the minimum detection distance, while the target size affects the maximum detection distance. A tiny-camera spectral feature verification system was built. The results show that the non-reflective light ratio curve of the target using the absorptive infrared cut-off filter changes smoothly and the value is high. As for the curve of the target using the reflective infrared cut-off filter, the value is high in visible light part while it is low in infrared part, and the curve begins to drop near 700 nm and even mutates where the absolute slope value is more than 10 times as it in the infrared band according to the experimental data. The experimental results are consistent with the expected results of the analysis, which proves the feasibility of hyper-spectral imaging technology to detect tiny cameras.
2021, 42(6): 1115-1121.
doi: 10.5768/JAO202142.0605002
Abstract:
The data matrix (DM) code has large information storage capacity, wide coding range and high reliability, which is an important carrier of people’s daily information exchange. In order to realize the parallel processing method of DM code, the influence of the distance value and gray value of DM code on the recognition time was analyzed experimentally. The DM codes under different conditions were scanned several times, and the reading time was recorded. The critical dot diameter value of DM code was determined to be 11 pixels, and the critical gray value was 220. On this basis, a parallel processing method of DM code based on spatial light modulator was designed, and experimental verification of laser parallel processing of DM code was carried out in quartz glass. Compared with the traditional single-beam laser marking two-dimensional code, the processing efficiency was improved.
The data matrix (DM) code has large information storage capacity, wide coding range and high reliability, which is an important carrier of people’s daily information exchange. In order to realize the parallel processing method of DM code, the influence of the distance value and gray value of DM code on the recognition time was analyzed experimentally. The DM codes under different conditions were scanned several times, and the reading time was recorded. The critical dot diameter value of DM code was determined to be 11 pixels, and the critical gray value was 220. On this basis, a parallel processing method of DM code based on spatial light modulator was designed, and experimental verification of laser parallel processing of DM code was carried out in quartz glass. Compared with the traditional single-beam laser marking two-dimensional code, the processing efficiency was improved.
2021, 42(6): 1122-1126, 1132.
doi: 10.5768/JAO202142.0605003
Abstract:
Aiming at the large aperture optical window working in the complex environment such as pressure difference and temperature gradient, a combination scheme of optical window was proposed, which was composed of optical glass and acrylic plate. Based on the thermal-optical analysis, the overall strength and thermal environment of optical window were analyzed and calculated theoretically, and the minimum thickness of optical window glass was obtained. The pressure field and axial temperature field were mapped to three-dimensional structure model by finite element software. The surface shape change and imaging quality evaluation index of 380 mm diameter optical window under different glass thickness were calculated. The simulation results were verified by corresponding environmental tests. The results show that the thickness of large aperture optical window with K9 optical glass as raw material is not less than 32.5 mm under this working environment; when the thickness of optical window is 35 mm, the thermodynamic effect can be ignored. Therefore, the large aperture combined optical window of 35 mm can not only meet the intensity, but also meet the imaging quality requirements of multi-spectral camera, which provides a basis for the design of this kind of window.
Aiming at the large aperture optical window working in the complex environment such as pressure difference and temperature gradient, a combination scheme of optical window was proposed, which was composed of optical glass and acrylic plate. Based on the thermal-optical analysis, the overall strength and thermal environment of optical window were analyzed and calculated theoretically, and the minimum thickness of optical window glass was obtained. The pressure field and axial temperature field were mapped to three-dimensional structure model by finite element software. The surface shape change and imaging quality evaluation index of 380 mm diameter optical window under different glass thickness were calculated. The simulation results were verified by corresponding environmental tests. The results show that the thickness of large aperture optical window with K9 optical glass as raw material is not less than 32.5 mm under this working environment; when the thickness of optical window is 35 mm, the thermodynamic effect can be ignored. Therefore, the large aperture combined optical window of 35 mm can not only meet the intensity, but also meet the imaging quality requirements of multi-spectral camera, which provides a basis for the design of this kind of window.
2021, 42(6): 1127-1132.
doi: 10.5768/JAO202142.0605004
Abstract:
Based on the theory of matrix optics, the first-order properties of the zoom system of the three-element variable focal power devices were analyzed, and the control equation of the focal power based on the system magnification was obtained. Combined with the initial parameters of the three-element system and taking the minimum spherical aberration as the objective, the initial structure of the system was designed, the control function of the shape parameters of the lens was obtained, and the focal power and image quality were evaluated and calculated. Finally, the design example was compared and verified. The results show that the solution of the optimal value of the minimum spherical aberration is reliable. The starting point and method of the design provide a theoretical research reference for the further design of zoom system by using variable focal power devices.
Based on the theory of matrix optics, the first-order properties of the zoom system of the three-element variable focal power devices were analyzed, and the control equation of the focal power based on the system magnification was obtained. Combined with the initial parameters of the three-element system and taking the minimum spherical aberration as the objective, the initial structure of the system was designed, the control function of the shape parameters of the lens was obtained, and the focal power and image quality were evaluated and calculated. Finally, the design example was compared and verified. The results show that the solution of the optimal value of the minimum spherical aberration is reliable. The starting point and method of the design provide a theoretical research reference for the further design of zoom system by using variable focal power devices.
New website launched!
陕公网安备 61011302001501号