Many vision-based methods for defect detection have been proposed to replace inefficient manual inspection. However, due to the complex shape and high degree of freedom required to manipulate individual coils, an automatic optical inspection (AOI) method for cylindrical coils remains unsuccessful due to the difficulty of automatically scanning the front, rear, outer, and inner faces of complex spiral-shaped cylindrical coils. In this article, we present a turnkey integrated system that can be operated in real time for real coil manufacturers. First, a novel 1-degree-of-freedom rotational progressive system is designed to automatically transport and sort individual spiral coils. Second, a multisensor imaging system is designed to capture images of 360° panoramic views of a cylindrical coil. Then, a robust automatic defect detection algorithm consisting of crack detection and geometric measurement is developed to detect defects in cylindrical coils during manufacturing procedures. The results of a large-scale validation experiment indicate that the proposed AOI system is robust and is practically immune to variations in surface normal vectors, occlusion, and defocus blur, and it has a high sensitivity and specificity of 99.75% and 94.2%, respectively. This validation experiment proved that the software and hardware designed in this study can implement defect detection of spiral coils.
The spectral and depth (SAD) imaging method plays an important role in the field of computer vision. However, accurate depth estimation and spectral image capture from a single image without increasing the volume of the imaging sensor is still an unresolved problem. Our research finds that a snapshot narrow band imaging (SNBI) method can discern wavelength-dependent spectral aberration and simultaneously capture spectral-aberration defocused images for quantitative depth estimation. First, a micro 4D imaging (M4DI) sensor is proposed by integrating a mono-chromatic imaging sensor with a miniaturized narrow-band microarrayed spectral filter mosaic. The appearance and volume of the M4DI sensor are the same as the integrated mono-chromatic imaging sensor. A simple remapping algorithm was developed to separate the raw image into four narrow spectral band images. Then, a depth estimation algorithm is developed to generate 3D data with a dense depth map at every exposure of the M4DI sensor. Compared with existing SAD imaging method, the M4DI sensor has the advantages of simple implementation, low computational burden, and low cost. A proof-of-principle M4DI sensor was applied to sense the depth of objects and to track a tiny targets trajectory. The relative error in the three-dimensional positioning is less than 7% for objects within 1.1 to 2.8 m.
Structured illumination microscopy (SIM) with π/2 phase-shift modulation traditionally relies on frequency-domain computation, which greatly limits processing efficiency. In addition, the illumination regime inherent in structured illumination techniques often results in poor visual quality of reconstructed images. To address these dual challenges, this study introduces DM-SIM-LLIE (Differential Low-Light Image Enhancement SIM), a novel framework that integrates two synergistic innovations. First, the study pioneers a spatial-domain computational paradigm for π/2 phase-shift SIM reconstruction. Through system differentiation, mathematical derivation, and algorithm simplification, an optimized spatial-domain model is established. Second, an adaptive local overexposure correction strategy is developed, combined with a zero-shot learning deep learning algorithm, RUAS, to enhance the image quality of structured light reconstructed images. Experimental validation using specimens such as fluorescent microspheres and bovine pulmonary artery endothelial cells demonstrates the advantages of this approach: compared with traditional frequency-domain methods, the reconstruction speed is accelerated by five times while maintaining equivalent lateral resolution and excellent axial resolution. The image quality of the low-light enhancement algorithm after local overexposure correction is superior to existing methods. These advances significantly increase the application potential of SIM technology in time-sensitive biomedical imaging scenarios that require high spatiotemporal resolution.
Aiming at the problem that the existing differential confocal measurement (DCM) method lacks the ability to decouple the differential signals from over-range the depth data, which in turn leads to measurement errors, this study proposes a differential confocal effective measurement region extraction method based on dual measurement (D-DCM). The first measurement is a double blur depth measurement, which is a coarse measurement, and the initial acquisition of the depth map is realized by applying a known Gaussian kernel blurring to the focal plane image and calculating the edge gradient ratios between the original image and the blurred image. This depth map is used to determine whether the depth of the measured sample in the field of view is within the differential confocal range and to segment it. The second measurement is a DCM, which is a fine measurement. The effective measurement area of the segmented image is second measured with high precision depth measurement. It is shown by single and multiple step heights experiments that the proposed D-DCM method can effectively solve the problem of decoupling the differential signal from the axial depth in differential confocal.
2 September 2026 in Measurement Science and Technology
Advanced semiconductor packaging and heterogeneous integration demand 3D surface metrology that combines submicron precision with industrial-scale throughput. However, traditional optical methods are inherently limited by the slow speed of mechanical scanning. To overcome this, we propose a chromatic differential arrayed-confocal three-dimensional microscopic imaging (CDAC-3DMI) method. By integrating a digital micromirror device for high-speed programmable pinhole generation with a chromatic differential detection principle, the proposed system enables scanless online 3D topography reconstruction without lateral or axial mechanical scanning. Experiments on packaging substrates validate the measurement fidelity, yielding step-height and Sa relative deviations of less than 3% and 1%, respectively, compared with a commercial laser scanning confocal microscope (Olympus OLS5100), within the measurement uncertainties reported in the validation table. CDAC-3DMI reduces the end-to-end full-resolution 3D acquisition time from 20 s to 65 ms under equivalent measurement conditions, corresponding to a 308-fold improvement in throughput. CDAC-3DMI thus provides a robust, high-throughput, and non-destructive solution for submicron characterization of complex microstructures in high-volume microelectronics packaging production lines.
In view of the difficulty in simultaneously eliminating multiple systematic errors and random errors in parallel bilateral chromatic dispersion differential confocal images, this paper proposes a hybrid error simultaneous removal method that integrates its prior systematic error matrix and low-rank approximation of random errors. Firstly, the causes of multiplicative systematic errors such as uneven illumination, different reflectivity, and stage tilt in the parallel bilateral chromatic dispersion differential confocal microscopy (PBCDDCM) measurement system are analyzed, and the systematic error matrix is experimentally measured. Then, the causes of random errors such as dead pixels and Gaussian noise in confocal images are analyzed, and the prior knowledge of the low-rank matrix recovery model is utilized to simultaneously eliminate multiple types of random noise. By integrating the systematic error matrix into the low-rank matrix recovery model, designing a low-rank approximation function, and solving the low-rank matrix recovery model using the alternating direction method of multipliers, a confocal image with eliminated systematic and random errors is obtained. This method is applied to eliminate mixed errors in confocal images of standard step samples and industrial sample gold fingers, and their surface topography is measured and Poisson surface modeling is performed. Experiments show that for a standard step sample with a height of 15 μm, the average height value of the measured step surface before correction is 15.14 μm, with a relative error of 1.06%. After correction, the average height value of the measured step surface is 14.99 μm, with a relative error of 0.07%. The surface topography of the gold finger is restored, and its imaging quality is significantly improved. The average height value of the steps obtained from multiple measurements of the gold finger is 19.54 μm, which has a relative deviation of 2.51% compared to the height measurement result of 19.06 μm obtained at the same position on an Olympus laser microscope. Therefore, this method enriches the field of confocal image surface measurement and topography restoration with noise elimination in both theory and practice.