Journal of The Optical Society of America A-optics Image Science and Vision最新文献

Accurate estimation of the noise type and noise level in color images is crucial for tasks such as denoising, segmentation, and super-resolution. However, existing approaches often rely on the assumption that the noise type is known, and they tend to suffer from significant deviations when dealing with complex textures or strong inter-channel correlations in color images. To address these limitations, this paper proposes a quaternion-based framework for estimating noise type and noise level under two representative and widely used noise families: additive noise and multiplicative-additive noise (non-additive noise). By leveraging quaternion matrix modeling, the proposed method effectively captures cross-channel correlations, thereby enhancing the accuracy of both type discrimination between these two noise categories and noise-level estimation. On this basis, a classification model is developed by combining statistical features with logistic regression. Furthermore, differentiated noise-level estimation strategies based on weak-texture extraction are designed for the identified additive or non-additive noise models. Extensive experimental results demonstrate that the proposed method can accurately identify noise types and significantly improve the precision of noise-level estimation across diverse color image datasets and complex noise conditions, outperforming state-of-the-art techniques.

Aiming at the problems of low accuracy and slow speed of existing point cloud weld extraction algorithms in 3D vision-based robotic intelligent welding, this study proposes a novel, to our knowledge, three-stage automatic point cloud weld extraction method. In the plane segmentation stage, the random sample consensus (RANSAC) algorithm is improved: by narrowing the selection range of sampling points to the local neighborhood and optimizing neighborhood construction with dynamic curvature detection, the efficiency of plane fitting is enhanced. In the feature point extraction stage, based on plane parameters, the plane intersection line method and distance threshold method are adopted to obtain weld seam feature points. In the curve fitting stage, farthest point sampling (FPS) is used to denoise and resample the feature points, and then the weld curve is fitted to achieve high-precision contour reconstruction. Experiments show that the method exhibits high efficiency, robustness, and engineering adaptability.

We present a rigorous frequency-domain finite-element formulation for modeling the diffraction of transverse magnetic polarized electromagnetic waves through metallic subwavelength slits. The approach incorporates realistic boundary conditions, open-domain truncation via perfectly matched layers, and material dispersion. Numerical simulations reproduce Fabry-Perot- like resonances for a reference slit and demonstrate that embedding symmetric internal ridges produces measurable spectral shifts, redistributes near-field hot spots, and modulates transmitted power by factors exceeding two compared to the plain slit. The proposed model provides compact, quantitative design rules for tuning resonance wavelengths and quality factors, enabling the engineering of subwavelength photonic components for filtering, sensing, and light manipulation.

The wave superposition model of the geometric phase shows how the addition of waves creates a shift in the resulting wave position. While previous work focused on a basis of linearly polarized light waves and the Pancharatnam-Berry phase, we show how the spin-redirection phase (Rytov-Vladimirsky-Berry phase) can also be derived from the same approach of wave superposition, using rotating vectors to represent the superposing oscillations. The result is the first derivation of the spin-redirection phase using wave superposition. We illustrate this approach with two classic examples of the geometric phase of rotations in space: a system of three fold mirrors and the helically coiled fiber.

The maintenance of railway rails relies heavily on accurate profiling and wear assessment. In this research, a rail profile detection system using line-structured light machine vision technology is developed. Traditional image processing algorithms for rail profile measurement involve Zhang's camera calibration, radial distortion correction, Gaussian filtering, and the iterative closest point (ICP) algorithm for point cloud registration. Building upon these conventional algorithms, an integrated error correction framework comprising projective transformation and system offset compensation is proposed. We introduce a method to dynamically determine the direction vector of the rail alignment and the angle between the laser plane and the rail cross-section for the projective transformation. Compared to the same hardware system without error correction, this method improves measurement accuracy from 0.0686 to ±0.015mm at the lateral wear measurement points and from 0.0678 to ±0.020mm at the vertical wear measurement points in profile detection.

Traditional prism imaging spectrometers suffer from poor direct vision, low spectral linearity, and a large natural smile. Spurred by these limitations, this paper proposes a spectral imaging system based on a double Amici prism, for which the vector refraction law yields expressions for the prism's smile and keystone. Accurate calculations of the smile and keystone for the single and compound prisms are derived using MATLAB. The characteristic curve of a smile is drawn and verified through simulations, with the designed system's waveband ranging from 400 to 1000 nm, spectral resolution across the full waveband exceeding 12 nm, and the maximum smile and keystone values of 3.117 and 3.955 µm, respectively. Notably, the developed system is completely coaxial and meets the requirements of direct vision and smile correction. Furthermore, the new, to our knowledge, method, which combines the vector refraction law with MATLAB programming, is adopted, enabling quick and accurate determination of the smile characteristic of compound prisms. Overall, this paper has general significance for the design of the prism spectral imaging system for smile correction.

As optical systems are expected to meet increasingly strict demands, controlling pupil aberrations is increasingly important in optical design. Building on prior work on image aberration theory and leveraging the connection between pupil and image aberrations, we derive analytical expressions for 50 intrinsic pupil aberration coefficients for plane-symmetric systems, supported by both algebraic and numerical validation.

Displacement integration in background-oriented Schlieren (BOS) is a critical step in the reconstruction of physical fields. This process typically employs high-order fitting integration or discrete Poisson solvers. This paper examines the strengths and limitations of these conventional approaches and proposes a novel physics-informed neural network (PINN) framework constrained by the Poisson equation, termed the adaptively balanced Poisson-constrained PINN (AB-PoissonPINN). The proposed method incorporates relative loss balancing with random backtracking (ReLoBRaLo) to dynamically balance the contributions of different loss components. The integration performance of AB-PoissonPINN is evaluated through both simulated and experimental numerical integration and is benchmarked against established techniques, including weighted cubic spline least squares integration (WCSLI), discrete Poisson solvers, and standard PINN. Experimental results demonstrate that AB-PoissonPINN consistently achieves higher accuracy than WCSLI, discrete Poisson solvers, and standard PINN under both noise-free conditions and various noise levels.

Starbursts are entoptic visual phenomena perceived as radial spikes around point light sources, frequently reported under scotopic conditions or following ocular surgeries such as LASIK or intraocular lens implantation. While starbursts have traditionally been attributed to high-order optical aberrations, in particular spherical aberration, we hypothesize that crystalline lens suture patterns may play a dominant role in the formation of starbursts through light diffraction. This study investigates the contribution of lens suture diffraction to starburst formation and compares it to the influence of high-order spherical aberrations. Using ex vivo porcine crystalline lenses mounted in a mechanical expansion unit, we performed through-focus imaging and aberrometric analysis with a custom optical system. Suture patterns were segmented and used to simulate diffraction-based point-spread functions (PSFs) by convolving them with a Gaussian point source. Simulated PSFs were compared to those generated using fourth- and sixth-order spherical aberration parameters. A cross-correlation analysis and angular detection of diffraction spikes were performed to assess the similarity between experimental and simulated PSFs. Results demonstrate that diffraction by crystalline lens sutures generates starburst-like PSFs, independent of high-order aberrations. Through-focus simulations revealed asymmetric Strehl ratio distributions when suture diffraction was included, suggesting that sutures significantly degrade optical quality over a broader dioptric range. Cross-correlation coefficients (τ) between experimental and simulated PSFs exceeded 0.7 in all cases, and diffraction spike orientations showed high angular consistency, confirming the predictive value of the suture diffraction model. Our findings support the hypothesis that lens sutures are the primary source of starburst formation via diffraction, with spherical aberration acting as a secondary factor.

The process for manufacturing wavefront-guided scleral lenses includes polishing both the anterior and posterior lens surfaces. The purpose of this study was to determine if extended front-surface polishing durations induce aberration changes in wavefront-guided scleral lenses that exceed manufacturing tolerances. Two sets of six scleral lenses were manufactured. Each set contained one reference lens with only sphere correction and five wavefront-guided scleral lenses, all of which shared the same wavefront-correcting optics. After cutting, the anterior surfaces of the five wavefront-guided lenses in both sets were polished for differing durations. Set 1 lenses were polished with a used polishing cloth, and Set 2 were polished with a brand-new polishing cloth. Aberrations in each polished lens were measured with a wavefront sensor and examined as a function of polishing time. Changes in aberrations that would be clinically significant and/or greater than manufacturing tolerances were not observed for the material or polishing times studied or the remaining lifetime of the polishing cloth.