Abstract:Structured illumination has been used for several decades in the field of three-dimensional（3D）shape measurement and machine vision. However，the bulky structured illumination generation system limits its poten? tial in practical applications. In this paper，a compact design method based on the silicon-on-insulator（SOI）opto? electronic integrated chip is proposed. Compared with the traditional structured light generation methods，the chip-based illumination method is simple，stable and flexible. The beam modulation and interference are achieved by on-chip devices，which efficiently avoids external disturbance and increases the portability of system. This is the first time using only on-chip devices to control the infrared beam and generate structured light pattern. The chip can provide an illumination area of about 200*200 μm2 with the chip size of 0. 5*0. 5 mm. The illumination area and structural period are related to the design of grating couplers and the wavelength of light. Moreover，dif? ferent illumination patterns can be achieved by appropriately designing the optical devices on the chip

Abstract:We propose a multichannel two-dimensional optical chemical sensor in a photonic crystal (PhC) slab. The device is composed of four cavities with a center wavelength at 3.3μm, and the channels are spaced by 10nm. The sensor is designed on an 800-nm-thick silicon-on-insulator platform for standard CMOS technology. The optical characteristics of the nanocavity structure are simulated by 3-D finite difference time-domain (FDTD) method. The transmittance of each channel is about 39%, with non-uniformity of transmittance across channels less than 0.25dB. The capability of the sensor is demonstrated by detecting carbon tetrachloride and benzene solutions, and shows a sensitivity of 209.2nm/RIU.

Abstract:A tunable feedback oscillator for 850GHz terahertz wave imaging system is proposed. The use of a folded waveguide as the slow-wave structure permits the superior performance together with the capability of UV-LIGA process. The comparison of dispersion character of FWG designed and optimized using eigenmode solver in CST Microwave Studio, respectively applied in regenerative feedback oscillator and traveling wave tube shows the key factor of tunable frequency centered at 850GHz. Additionally, a feedback circuit including T-joint structure with lossy metal is simulated and the design, including SWS and feedback circuit, is verified by 3-D particle-in-cell simulations. On varying the beam voltage, the frequency-adjustable oscillation changes from a stale single-frequency state at the beginning to multi-frequency spectra, demonstrating more than 200mW output power.

Abstract:An ion-implant technique for fabricating Si:As blocked impurity band detectors for VLWIR detection had been investigated, and the detectors with good photoelectric response performance had been demonstrated by optimizing both the processing condition and the device structural parameters together with material characteristic parameters. At 5 K temperature, with a dc bias voltage of -3.8V, the peak response wavelength of the fabricated devices is 23.8 μm, the blackbody responsivity is 3.7A/W, and the detectivity is 5.3×10¹³cm?Hz^1/2/W at 3.2V, which are comparable to （and even superior to） those reported in literatures. Especially, the device manufacturing process is compatible with that for fabrication of integrated circuit, and the detectors can be integrated with readout circuits on one chip, resulting in a remarkable reduction in produce cost and a significant improvement in the imaging performance.

Abstract:In this paper, an improved direct extraction method to extract the model parameters in InP heterojunction bipolar transistor （HBT） small-signal equivalent circuit is presented and successfully applied to small-signal equivalent circuit of InP HBT. The distributed base-collector capacitance effect is taken into consideration in the adopted model. The extracting process of this method, which extracts parameters in turn from the peripheral parasitic elements to the intrinsic internal elements, is clearer than other direct extraction methods. Except for the parasitic parameters, all other parameters are calculated without any simplified approximation. This method relies on S parameters measurement. All of the equivalent circuit parameters are extracted directly from the S parameters without using approximations based on initial values. The direct extraction method is successfully validated on InP HBT in the frequency range of 0.1 ~ 40 GHz, and excellent agreement is achieved between the measured and calculated S parameters over the whole frequency range.

Abstract:An algorithm combining frequency domain imaging algorithm and compressed sensing （CS） framework is proposed in here for millimeter-wave multi-static sparse array imaging. The algorithm consists of two major steps. Firstly, the typical fast Fourier transform （FFT） algorithm used in square boundary array with phase center approximation （PCA） is carried out. However, the residual phase error introduced by the PCA at close range cannot be compensated completely, so in the second step, the modified sparse learning via iterative minimization （SLIM） algorithm which is in the CS framework is introduced to refocus the initial images. By combining PCA and the modified SLIM algorithm, the proposed algorithm reaches a better computational efficiency, improves the image quality, and alleviates the requirement for iterations of the original SLIM algorithm. Simulation results verify the effectiveness of this algorithm.

Abstract:A 94 GHz mm-Wave LNA targeted for imaging system is designed and fabricated in 55 nm CMOS process. The dual-coupling g_m-boosting technique is proposed to achieve high gain and wide-band input matching. Meanwhile, in order to improve the gain and ensure the stability of the proposed LNA, the capacitance neutralization method and the common-gate-shorting technique are simultaneously introduced. The measurement results indicate that the LNA achieves a small signal gain of 14.2 dB, a BW-_3dB of 87.1~95 GHz, a NF of 6.7dB as well as an input-referred 1 dB compression point of -13 dBm.

Abstract:The oxygen concentration detection method based on near infrared absorption spectrum at 760.88nm was used to realize in-situ, non-contact detection of the residual oxygen concentration in the vial in the open environment on the lamp detector. The method based on Wavelength-modulated tunable diode laser absorption spectroscopy technology （TDLAS/WMS） uses the principal component extraction method （PCA） to extract the main characteristics of the second harmonic in the open light path, and then utilizes the genetic algorithm （GA） to optimize the BP neural network to build a concentration inversion model. This method can reduce the data required for calculation, suppress noise and improve the processing speed of post-processing data. The experimental results show that the average relative error of this method is reduced from 8.32% to 1.12%, and the coefficient of determination is increased by 8.86%, compared with the least square fitting method using semi-peak area. Compared with the average relative error of the single PCA-BP neural model, the mean relative error is reduced from 3.80% to 1.12%, and the coefficient of determination is increased by 2.81%.This method can effectively suppress the signal random disturbance caused by the open light path environment and improve the accuracy and stability of the detection of oxygen residual concentration in the vial.

Abstract:In order to improve the efficiency and performance of the analysis model, a wavelength selection method based on variable stability and population analysis (VSPA) is proposed. Firstly, the variables are divided into sample space and variable space, and the stability of variables is calculated in the sample space. According to the stability value, the variables are divided into useful variables and useless variables by weighted bootstrap sampling technology. Then, in the variable space, the frequency of each variable is calculated, and the exponential decline function is used to remove the variables with lower frequency from the useless variables. Finally, the proposed algorithm is applied to corn NIR data set to predict the starch content. The predicted root root-mean-square (RMSEP) and predicted correlation coefficient (R_P) is 0.0409 and 0.9974, respectively. The variables after selection are only 2.7% of the original spectral data. It shows that the proposed variable selection method can improve the operational efficiency and prediction accuracy of the model, and is proved to be an effective variable selection method.

Abstract:A new readout circuit structure for infrared focal plane array with pixel-level shared source follower and double column buses is proposed in this paper. The voltage signal of the pixel is transferred through double column buses instead of single column bus, and therefore the non-uniformity and non-linearity caused by the parasitic resistor of the column bus is eliminated. By sharing the source follower within four adjacent pixels in the same column, the aspect ratio (W/L) and area of the source follower are increased to suppress the thermal noise, flicker noise and non-uniformity caused by process. A 640×512 readout circuit using this structure is designed and fabricated in 0.35μm 2P3M CMOS process and the pixel pitch is 15μm. The test results indicate the readout circuit receives a high dynamic range (DR) of 81dB high-quality infrared images with a low power consumption of 30mW. The nonlinearity is 0.11%, and the non-uniformity is less than 1%. Medium-wave infrared detector assembly is fabricated and tested. The assembly’s non-uniformity is less than 5% and the NETD is 18mK. High-quality infrared images are obtained.

Abstract:In the dual-path FMCW laser ranging system, the length calibration accuracy of the reference optical fiber directly affects the ranging accuracy. In order to improve the ranging accuracy, a high precision reference optical fiber calibration method was proposed, which was based on hydrogen cyanide(H¹³C¹⁴N) gas cells signal splicing. In this paper, the principle of calibration method based on H¹³C¹⁴N gas cells was deeply researched. Moreover, to reduce the burden of data acquisition, H¹³C¹⁴N signal splicing was used to optimize the calibration method. The experiments show that the calibration method based on H¹³C¹⁴N signal splicing is more stable than that based on laser interferometer traditionally. After calibrating the reference optical fiber by using H¹³C¹⁴N. At the same time, within a measuring range of 3.8 m, the error between ranging value of FMCW system and measuring value of interferometer is less than 14 μm, and the standard deviation is less than 17 μm.

Abstract:It brings about both higher computational complexity and data scale augmentation, while multi-directional information of remote sensing images has the superiority in extending function and increasing accuracy in cloud detection. Directonal information on cloud detection comes from BRDF of surface combined with atmospheric effects. Although determing certain view angles in local area by analytic solution is possible, but considering the complexities of application the traverse calculation for whole angles is still carried out in POLDER officer products. Due to redundant information existed on close neighbours of angler layer, averaged joint information entropy and K-L information divergence formed a feature basis for selection of angle subset. Two algorithms of optimal and idel solutions on Pareto multi-objectives front are proposed. Experiments of cloud detection were taken on two POLDER datasets firstly, then on a dataset of directional polarimetric camera （DPC） on board GF-5 satellite. The experimental results demonstrated that the overall accuracy of cloud detection by proposed method based on 2 angle- layers combinations is 89.36%, Kappa equals 0.7845 The validation on DPC dataset also showed that in comparison with GF-5 remote sensing synthetic image the similarity among them is 86%. The computation efficiency was raised to 7 times . Thus the proposed method takes advantages on computational cost saving and retaining multi-angle image’s intrinsic information. It contributes a new insight to cloud detection with its advantages of effectiveness, satisfactory accuracy and automatic operation.

Abstract:In this paper, the solution to dislocation between adjacent rows is proposed for the advanced geosynchronous radiation imager (AGRI) of Fengyun-4A (FY-4A). Firstly, according to the scanning mode of AGRI combined with the shift theorem of the Fourier Transform, the phase difference curve of adjacent pixels in the misaligned row is calculated. Secondly, the displacement value is obtained by fitting the low frequency portion of the phase difference spectrum by the least squares method. Finally, the dislocation rows are reconstructed by means of “weight assignment” according to the displacement value, and the problem of dislocation between the rows is solved. The scheme proposed is verified by simulation experiment, and the in-orbit images of AGRI are processed. The results show that the proposed algorithm can achieve registration between adjacent rows with high precision on sub-pixel level.

Abstract:Three-axis stabilized geostationary orbit plane array camera needs to search and track the target in a certain field of view, and usually adopts the "staring" imaging technology scheme combining two-dimensional directional mirror and array detector. However, there are some problems such as image distortion and image rotation, which affect the accuracy of target tracking and positioning. The distortion correction and image rotation correction of 60 degrees two-dimensional directional mirror array camera is carried out, and a geometric model laboratory calibration method based on the combination of internal and external azimuth elements is proposed. The average error are 0.74 pixels, which satisfies the requirement of practical use, by calculating and optimizing 13 error items of the normal, north-south axis and east-west axis of the pointing mirror.

Abstract:The space-borne photon counting laser altimetry can detect continuous elevations of vegetation canopy and earth surface for its high along-orbit resolution. However, the relatively low point cloud density and low signal-to-noise ratio (SNR) of space-borne vegetation point clouds put forward new requirements for estimating vegetation canopy heights. In this paper, an adaptive directional model for estimating vegetation canopy heights using space-borne vegetation point clouds was proposed to meet the new requirements. Firstly, the range of signal elevation was roughly obtained by searching two extremums that represent the crown and ground in the statistical histogram of point cloud elevation. The land slope and average densities of crown, ground and noise were estimated as well. Then, the roughly denoised point clouds were further fine denoised by adaptive directional density-based clustering where the direction of neighborhood is along the land surface, and the thresholds related to density are adjusted adaptively according to the estimated point cloud densities. After filtering, the elevations of ground and canopy were estimated respectively by applying triangular irregular networks (TIN) where the initial points of ground and canopy in TIN were found by the densities and elevation percentage of point clouds. Vegetation point clouds of ATLAS space-borne laser altimeter are used to validate the filtering method. The experimental results show that the adaptive directional model can correctly estimate vegetation canopy heights and is fit for areas with large slope and low leaf area index. The determination coefficients R² of canopy and ground elevation between processed ATLAS data and airborne LIDAR data are 0.99 and 0.77 respectively, and RMSE are 0.28 M and 2.6 m.

Abstract:Hyperspectral laser radar combines the characteristics of LiDAR and hyperspectral information, and provides more accurate remote sensing detection methods of the extraction of vegetation physiological and biochemical parameters, but its application potential has not been fully explored. In this paper, the leaves of 10 typical tree species in Beijing are taken as samples to carry out the leaf observation experiment of indoor hyperspectral laser radar. And the tree species classification research is carried out to provide the basis of the future forestry application of hyperspectral laser radar. In this study, the hyperspectral data of tunable hyperspectral LiDAR (HSL) was carried out and compared with the data measured by ASD spectrometer. Secondly, 10 kinds of leaves were classified by random forest method. In the process, the total spectral index is obtained by combining all the bands and some sensitive bands with the spectral index. The results show that: (a) HSL is consistent with ASD spectra observed in the band 650~1 000 nm (71 channels) (R2=0.9525~0.993 2, RMSE=0.058 7); (b) The classification accuracy of the original band reflectivity is 78.31%, there into the maximum contribution rate of the classification band is 650~750 nm, and the classification accuracy is 94.18% using this band which shows that it is very effective in classify tree species by using red edge band (650~750 nm); (c) the bands sensitive to tree species classification are 680 nm, 685 nm, 690 nm, 715 nm, 720 nm, 725 nm, 730 nm; d) When we combine the spectral index and vegetation index, the classification accuracy is 82.65%. This study shows that at the single leaf level, hyperspectral LiDAR can accurately reflect the spectral characteristics of the target leaves and classify the species of different trees effectively. It is possible to extract physiological and biochemical parameters of targets for the future field applications accurately.

Abstract:Processing ground penetrating radar data to obtain well-focused images for object detection has been an active research area. Phase-shift migration (PSM) is a widely used method since it allows the wave velocity to vary with respect to multi-layer medium. However, this requires pixel-by-pixel calculation of the image, which is time-consuming. This paper presents an extended Omega-k algorithm for multi-layer medium imaging with significantly less computation complexity than the PSM algorithm. The extended Omega-k exploits fast interpolation in the wave-number domain instead of iterative calculating as done by PSM. The method of estimating the wave propagation velocity in different media is also proposed via vertex region extraction for phase compensation and image focusing. Various images of buried targets of a two-layer medium experiment are obtained, which validate the effectiveness of the proposed algorithm, and make it practical for some typical ground-based surveillance applications.

Abstract:Due to the limitations of infrared optical diffraction and infrared detectors, the noise of infrared images is relatively large and the resolution is low. Super-resolution reconstruction of infrared images improves image resolution, but at the same time enhances the noise of background. Aiming at this problem, a salience region super-resolution reconstruction algorithm for infrared images based on sparse coding is proposed. Combining the saliency detection and the super-segment reconstruction improves the target definition and reduces the background noise. Firstly, image feature is extracted by double-layer convolution, and image patches with large entropy are adaptively selected for training the joint dictionary. Sparse features are used to calculate the saliency to obtain salient regions, which reconstructs image patches in saliency region by the trained dictionary, and the background region adopts Gaussian filtering. Experimental results show that the improved reconstruction algorithm is better than ScSR and SRCNN under the same conditions, and the image signal-to-noise ratio is increased by 3-4 times.