研究目的
To propose a fast calibration technique for random phase modulation optical phased array LiDAR to address the time-consuming calibration issue in raster scanning imaging schemes and improve image quality by retrieving the point spread function.
研究成果
The proposed fast calibration for RPMI is three times faster than RSI calibration and simultaneously retrieves the PSF, enabling deconvolution to improve image quality by reducing blurring from sidelobes. Experimental results show significant improvements in SNR and SSIM after deconvolution, demonstrating the effectiveness of the method for dynamic imaging applications.
研究不足
The OPA used is 1D with only 8 elements, limiting resolution and preventing 2D imaging experiments. RPMI requires more sampling patterns due to partial correlation, and the system's performance may be affected by fabrication imperfections and modulation rate constraints.
1：Experimental Design and Method Selection:
The study compares raster scanning imaging (RSI) and random phase modulation imaging (RPMI) schemes. It uses theoretical models for OPA beam steering and image reconstruction, including the use of stochastic parallel gradient descent (SPGD) algorithm for RSI calibration and cross-correlation for PSF calculation in RPMI.
2：Sample Selection and Data Sources:
A 1D 8-element OPA with specific parameters (wavelength 1550nm, pixel width 10μm, pixel pitch 20μm) is used. Scenes include a double-slit object in simulations and experiments.
3：List of Experimental Equipment and Materials:
Equipment includes a fiber laser source, customized 8-channel D/A, camera lens, CCD, single-pixel detector, high-speed A/D, infrared camera, and linear translation stage. Materials include a lithium niobate waveguide OPA and a transparent double-slit scene.
4：Experimental Procedures and Operational Workflow:
For calibration, modulation voltages are applied, and light distributions are recorded with a CCD. For imaging, the scene is illuminated, and light intensities are detected. Data is processed for image reconstruction and deconvolution using Richardson-Lucy algorithm.
5：Data Analysis Methods:
SNR and SSIM are calculated to evaluate image quality. PSF is derived from cross-correlation of light distributions for deconvolution.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容-
single-pixel detector
PDA20CS-EC
Thorlabs
Detects light intensities transmitted through the scene for image reconstruction.
-
fiber laser source
NP, 100mw
NP
Provides continuous wave laser light at 1550nm wavelength for illumination in the imaging system.
-
D/A
customized 8-channel
Generates and applies modulation voltages to the OPA electrodes for phase modulation.
-
camera lens
Nikon AF-S Nikkor, f=50mm, F/1.4G
Nikon
Focuses the modulated beam onto the scene for illumination.
-
CCD
Measures far-field light distributions during calibration and records light patterns.
-
A/D
PicoScope 6404D
PicoScope
Acquires and digitizes light intensity data from the detector for processing.
-
infrared camera
Records light distributions at the imaging area for calibration purposes.
-
linear translation stage
Moves the scene dynamically for dynamic imaging experiments.
-
登录查看剩余6件设备及参数对照表
查看全部
