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

The high-pressure flexible hoses used in oil and gas drilling have small diameters and long dimensions. When the structured light method is used to detect the defects on their inner walls, there are problems such as the obstruction of the supporting structure and the inability to scan and measure the complete inner wall. For this purpose, a conical structure light unobstructed scanning measurement system based on the arrangement of dual cameras and lasers on the same side has been developed. Based on the pinhole imaging model and the properties of spatial conical surfaces, it is analyzed that in the lateral placement architecture where the camera and the conical structure light exciter are suitable for scanning measurement, the spatial conical light overlapped on the image plane, making it impossible to achieve a one-to-one correspondence between pixel coordinates and world coordinates. By adding a measurement camera in the vertical direction, the overlapping area of the horizontal camera's imaging is supplemented to achieve blind-spot-free scanning and detection of the inner wall of the pipeline. Based on the system calibration of the conversion between pixel distance and three-dimensional spatial distance, tests were conducted on pipes with inner diameters of 51 and 64 mm containing defects of known sizes. The test results indicated that the root mean square error of the defect size at all positions was ≤0.076mm, and its good robustness within a certain pipe bending range was verified. Meanwhile, the tilt error and the irregular error of the laser strip edge existing in the measurement system were analyzed. The results show that this system has high application value in the measurement of inner wall defects of small-diameter and long-sized pipelines and can meet the actual engineering requirements.

Optical distortion or aberration remains a vital challenge that prohibits high-resolution imaging in various applications such as space domain awareness, terrestrial remote sensing, and astronomy. However, due to the stochastic nature of these optical distortions, reducing their effect without directly measuring wavefronts is challenging. Furthermore, in the case of extreme turbulence, due to the limited size of the lenslet array in the wavefront sensor, the sensor fails to correctly quantify or minimize the image distortions of a guide star from turbulence. While numerous studies have shown effectiveness of guide star-based adaptive optics in mitigating mild turbulence, severe turbulence has remained a persistent challenge. To target this, we present TURBO-RL: TURBulence mitigatiOn using Reinforcement Learning, which uses just a single optical element (e.g., deformable mirror) to estimate and correct the wavefront errors from a guide star. TURBO-RL adopts reinforcement learning with a convolutional neural network to extract and estimate turbulence. Unlike other methods, TURBO-RL is capable of guide star imaging in severe turbulence (D/r₀=100) with only about 590 photons, making it possible to overcome the strong turbulence and possibly replace bulky and expensive wavefront sensors.

Color appearance is strongly shaped by surrounding context, yet color-quality assessments for multicolored materials, such as camouflage textiles, typically rely on judgments made against uniform backgrounds. This study systematically tests whether a homogeneous background matched to the average color of a multicolored surround can approximate the perceptual influence of the original patterned background when assessing small color differences in MARPAT camouflage. Using a center-background paradigm, pairs of stimuli with either identical centers or small lightness differences (ΔL^∗=±0.5, ±1.0) were presented on three background types: a MARPAT two-color checkerboard, its corresponding CIELAB-matched average background, and a neutral gray reference. Psychophysical experiments with 20 color-normal observers revealed systematic perceptual discrepancies between patterned and color-averaged backgrounds: even physically identical centers produced mean visual differences of ∼0.2-1.2ΔE₀₀ (1:1:1). Background composition exerted a strong effect (p<0.001), whereas checkerboard size (4×4, 8×8, 16×16) had no significant influence. The largest induced differences occurred for combinations in which the center-surround lightness ratio approached unity, consistent with enhanced chromatic induction. Sensitivity to small ΔL^∗ differences declined sharply when stimuli were spatially separated, with many observers failing to detect even ±1.0ΔL^∗ differences on gray. Substantial intra- and inter-observer variability further highlighted the difficulty of judging subtle differences in complex scenes. Baseline measurements on uniform gray backgrounds yielded near-zero visual differences, confirming that the large apparent differences observed under patterned surrounds were induction-driven rather than due to random noise. Overall, the results demonstrate that a color-averaged background does not serve as a perceptually equivalent substitute for a complex multicolored surround, underscoring the need for context-appropriate backgrounds in the visual evaluation of camouflage, textiles, displays, and other patterned materials.

This paper presents a unified analytical framework to derive the three-dimensional transfer characteristics of axially symmetric optical imaging systems operating in the far field, with emphasis on coherence scanning interferometry, confocal microscopy, and focus variation techniques. While many 3D optical instruments are treated as linear systems, in practice inconsistencies remain in how their transfer functions are derived and interpreted, particularly across forward- and back-scatter geometries. Addressing this, we develop closed-form expressions for the 3D transfer function and point spread function for generic back- and forward-scatter systems under commonly applied apodization conditions (e.g., uniform, root-cosine, and cosine). These derivations clarify the spatial frequency support and resolution trade-offs intrinsic to each geometry and validate the characteristic "bowtie" and "umbrella" structures observed experimentally in the spatial frequency domain. Our analytical results not only resolve the ambiguities of previous numerical models but provide a means to validate and understand the applicability of approximate 3D and 2D numerical models. The formalism is robust, generalizable, and appropriate to the modeling and correction of real instruments in 3D surface metrology and optical tomography.

Propagating a plane wave modulated by an amplitude grating with a specially designed transmittance function in the uniaxial crystal can lead to the generation of the so-called elliptical carpet beams. These elliptical carpet beams retain the characteristic of the traditionally radial carpet beams in free space, such as maintaining stable patterns during propagation. On the other hand, due to the anisotropic diffraction in the uniaxial crystals, the elliptical carpet beams lose some symmetry, resulting in a kind of elliptical transverse carpet pattern during propagation. The ellipticity of the carpet pattern varies with the birefringence ratio for different uniaxial crystals. In this study, we have investigated in detail the behavior of the elliptical carpet beams by varying crystal refractive indices and derived the analytical expression of elliptical carpet beams. Particularly, we show that there exists a kind of characteristic ellipse of the elliptical carpet beams and point out the properties in relation to the birefringence ratio of the crystals.

We contrast two approaches to the optics of reflection in concave spherical mirrors. In the Optics (ca. 165), Ptolemy employed the cathetus principle as a regulative rule to account qualitatively for visual appearance in concave spherical mirrors. In his Magia naturalis (1589) and De refractione (1593), Della Porta's analyses rested on the assumption of a reciprocal relation between reflection in concave mirrors and refraction in glass spheres. These two historical cases highlight two fundamentally different conceptions of optics: Ptolemy explained optical phenomena based on vision, whereas Della Porta explained the same phenomena by appealing to the geometrical action of light. The shift from perception to the geometry of light rays marks an important development in the history of optics. The elimination of vision from the equation, as it were, ushered in modern optics.

The first proposed invisibility cloaks required materials that are highly anisotropic, spatially inhomogeneous, and that possess a magnetic response. These properties are still difficult or impractical to achieve in practice, leading many researchers to explore simplified invisibility schemes that trade perfection for simplicity in design. In this paper, we investigate a traditional method by Devaney for constructing multi-angle invisibility devices, i.e., devices that are invisible for a finite number of directions of illumination in the weak scattering limit. We demonstrate that the scattering cross-section of these objects decreases dramatically as the number of invisibility directions is increased.

Based on vector diffraction theory, this paper investigates the energy flow distribution and its variation characteristics of Bessel-Gaussian beams under polarization modulation. The results demonstrate that on the focal plane, the transverse energy flow with nearly zero intensity and the longitudinal energy flow with very high intensity are obtained, with the longitudinal energy flow dominating the total energy flow field. Furthermore, by adjusting the values of the azimuthal index m and the radial index n, vortex-shaped energy flow distributions can be achieved. The degree of vorticity and the number of energy flow spots can be controlled by adjusting the magnitude of the azimuthal index m, while the direction of the vortex is manipulated by altering the signs of the azimuthal index m and the radial index n. By adjusting the magnitude of the radial index n, size-tunable energy flow rings can be obtained. This polarization modulation strategy offers a novel approach for the flexible manipulation of focused optical fields. Furthermore, the characteristics of energy flow distributions and their dynamic variations will pave the way for new approaches in optical particle manipulation and trapping.

Based on pseudo-modal expansions of partially coherent beams, an analytical expression for the cross-spectral density function of pseudo-Schell model beams through an ABCD optical system is derived. The intensity distributions calculated using analytical method show excellent agreement with those by numerical integration. The radiation forces of focused pseudo-Schell model beams on Rayleigh dielectric particles are investigated. It is found that focused Gaussian pseudo-Schell model (GPSM) beams can stably trap high-refractive-index particles at the focus, while focused vortex Gaussian pseudo-Schell model (VGPSM) beams can trap low- and high-refractive-index particles simultaneously. Compared with focused Gaussian Schell-model (GSM) beams, focused GPSM beams generate stronger radiation forces under identical beam parameters, making them more effective for particle trapping. The vortex Gaussian Schell-model (VGSM) beams with a large coherence length exhibit a particle trapping capability comparable to that of VGPSM beam. The pseudo-modal expansion method of pseudo-Schell model beams through an ABCD optical system reduces the dimensionality of the integration and facilitates the synthesis and analysis of such beams in propagation. The findings provide a theoretical basis for the use of both GPSM beams and VGPSM beams in particle trapping.

Periodic structures play a crucial role in applied optics; however, generalized forms of periodicity are becoming increasingly important in electromagnetics. Due not only to their importance but also to the high cost and great difficulty in producing these at the nanoscale required, numerical simulation of these devices is of extraordinary importance. In this contribution, the author derives, implements, and validates the generalization of a high-order perturbation of surface (HOPS) algorithm (the method of field expansions-FE) for the numerical simulation of layered media scattering to account for quasiperiodic interfaces. Due to their interfacial character, these HOPS approaches are substantially faster than their volumetric counterparts such as finite difference or finite element methods. Additionally, our approach can address structures that the classical FE method would find onerous (for interfaces with widely disparate periods) or impossible (for profiles of incommensurate periods) due to its enhanced capability of simulating quasiperiodic interfaces. Beyond validating the implementation, the author also investigates the (nonlinear) dispersion relation of surface plasmon resonances on a sequence of increasingly challenging vacuum-silver structures featuring a quasiperiodic interface.