High-Resolution Tolerance Against Noise Imaging Technique Based on Active Shift of Optical Axis

Optical imaging resolution is determined by both optical parameters such as light wavelength and the objective lens Numerical Aperture (NA) and by spatial sampling such as Charge-Coupled Device (CCD) camera pixel size. Focused on improving optical imaging pixel resolution, we reviewed multiframe Super-Resolution (SR) implemented in previous measurement to improve image resolution, but found its computational cost to high and calculation too long. Pixeldisplacement estimation also adversely affects resulting image quality and is difficult to apply practically due to SR registration calculation sensitivity to noise. The resolution improvement we have proposed uses active subpixel shifting of the optical axis based on actively controlling spatial image displacement using a glass-plate-parallel substrate and a galvano scanner. Multiframe SR registration is used but without pixel-displacement estimation. Theoretically, our proposal may improve image resolution stably at higher speed but to clarify this, we developed optics and rebuilt the multiframe SR.We then computationally analyzed improved resolution and noise tolerance using a Modulation Transfer Function (MTF). Optical and aliasing noise have been suppressed and image resolution improved in high measurement, making higherresolution imaging faster and cheaper thanks to our proposal.

Download Full-text

Development of a Portable Optical Interferometry Microscope System

Key Engineering Materials ◽

10.4028/www.scientific.net/kem.656-657.561 ◽

2015 ◽

Vol 656-657 ◽

pp. 561-566

Author(s):

Hiroyuki Nishimoto ◽

Kenji Yamaguchi ◽

Katsunori Kimura ◽

Yoshiaki Suzuki ◽

Kazutake Uehara ◽

...

Keyword(s):

Optical Axis ◽

Low Cost ◽

Optical Filter ◽

Numerical Aperture ◽

Video Camera ◽

Optical Interferometry ◽

Objective Lens ◽

Basic Performance ◽

Manufacturing Applications ◽

Interference Image

Optical interferometry methods are widely used for measuring microdisplacement with nanometer accuracy. However, most commercially available optical interferometry systems are large and expensive for manufacturing applications. In this study, we report the development of a low-cost portable optical interferometry microscope for factory use. The light source was a tungsten–halogen white lamp with an optical filter. The microscope has an objective lens with a numerical aperture of 0.3, a magnifying power of 10, and field depth of 3.056 μm. Interference images were collected with an NTSC CCD-video camera. The resolution of the interference image is 320 × 240 pixels and stored in BMP format. To obtain phase shifted interferometry images, a piezoelectric actuator was used to monitor the table movement along the optical axis. The total cost of all system parts is approximately 7000 to 8000 US dollars. To evaluate the basic performance of the developed interferometry microscope, we measured a steel ball, the penetration mark of a Rockwell scale hardness indenter, and a gauge block surface with a bump. The developed interferometry microscope can measure continuous and gently sloping surfaces. The processing time is approximately 10–20 s.

Download Full-text

X-ray diffraction microscopy based on refractive optics

Journal of Applied Crystallography ◽

10.1107/s1600576717011037 ◽

2017 ◽

Vol 50 (5) ◽

pp. 1441-1456 ◽

Cited By ~ 26

Author(s):

H. F. Poulsen ◽

A. C. Jakobsen ◽

H. Simons ◽

S. R. Ahl ◽

P. K. Cook ◽

...

Keyword(s):

Axial Strain ◽

Three Dimensional ◽

Numerical Aperture ◽

Dark Field ◽

Reciprocal Space ◽

Objective Lens ◽

Optical Parameters ◽

X Ray ◽

Integrated Intensity ◽

Dimensional Reconstruction

A formalism is presented for dark-field X-ray microscopy using refractive optics. The new technique can produce three-dimensional maps of lattice orientation and axial strain within millimetre-sized sampling volumes and is particularly suited toin situstudies of materials at hard X-ray energies. An objective lens in the diffracted beam magnifies the image and acts as a very efficient filter in reciprocal space, enabling the imaging of individual domains of interest with a resolution of 100 nm. Analytical expressions for optical parameters such as numerical aperture, vignetting, and the resolution in both direct and reciprocal spaces are provided. It is shown that the resolution function in reciprocal space can be highly anisotropic and varies as a function of position in the field of view. Inserting a square aperture in front of the objective lens facilitates disjunct and space-filling sampling, which is key for three-dimensional reconstruction and analysis procedures based on the conservation of integrated intensity. A procedure for strain scanning is presented. Finally the formalism is validated experimentally at an X-ray energy of 17 keV.

Download Full-text

45.5X Infinity Corrected Schwarzschild Microscope Objective Lens Design

International Journal on Measurement Technologies and Instrumentation Engineering ◽

10.4018/ijmtie.2018010102 ◽

2018 ◽

Vol 7 (1) ◽

pp. 17-37

Author(s):

Sami D. Alaruri

Keyword(s):

Sensitivity Analysis ◽

Transfer Function ◽

Modulation Transfer Function ◽

Numerical Aperture ◽

Objective Lens ◽

Lens Design ◽

Modulation Transfer ◽

Microscope Objective ◽

Spherical Mirrors ◽

Microscope Objective Lens

In this article, the design of a 45.5X (numerical aperture (NA) =0.5) infinity corrected, or infinite conjugate, Schwarzschild reflective microscope objective lens is discussed. Fast Fourier transform modulation transfer function (FFT MTF= 568.4 lines/mm at 50% contrast for the on-axis field-of-view), root-mean-square wavefront error (RMS WFE= 0.024 waves at 700 nm), point spread function (PSF, Strehl ratio= 0.972), encircled energy (0.88 µm spot radius at 80% fraction of enclosed energy), optical path difference (OPD=-0.644 waves) and Seidel coefficients calculated with Zemax® are provided to show that the design is diffraction-limited and aberration-free. Furthermore, formulas expressing the relationship between the parameters of the two spherical mirrors and the Schwarzschild objective lens focal length are given. In addition, tolerance and sensitivity analysis for the Schwarzschild objective lens, two spherical mirrors indicate that tilting the concave mirror (or secondary mirror) has a higher impact on the modulation transfer function values than tilts introduced by the convex mirror (or primary mirror). Finally, the performed tolerance and sensitivity analysis on the lens design suggests that decentering any of the mirrors by the same distance has the same effect on the modulation transfer function values.

Download Full-text

Errors in the Vickers diagonal length measurement caused by different designs of microscopes

ACTA IMEKO ◽

10.21014/acta_imeko.v9i5.981 ◽

2020 ◽

Vol 9 (5) ◽

pp. 261

Author(s):

S. Takagi ◽

Y. Tanaka ◽

Y. Seino

Keyword(s):

Numerical Aperture ◽

Image Resolution ◽

Point Of View ◽

Length Measurement ◽

Resolving Power ◽

Measuring System ◽

Objective Lens ◽

Depth Of Focus ◽

The Relationship ◽

Diagonal Length

The errors in the Vickers diagonal length measurement are discussed from the point of view of the design of microscope. Effects of the numerical aperture (NA) of an objective lens are evaluated through the experiment. With smaller NA, the resolving power of the microscope gets worse and the depth of focus gets deeper as predicted Rayleigh’s criterion. The relationship between the repeatability of diagonal length and the objective NA follows the theory. In addition, the variation of NA relates the measured diagonal length. Those effects are much significant than the image resolution or the minimum count of a measuring system.

Download Full-text

Magnetic Domain Observation of an Amorphous Fe-Si-B Ribbon by Means of a High Voltage Scanning Electron Microscope

Proceedings, annual meeting, Electron Microscopy Society of America ◽

10.1017/s0424820100098113 ◽

1981 ◽

Vol 39 ◽

pp. 320-321

Author(s):

K. Tsuno ◽

Y. Harada ◽

T. Sato

Keyword(s):

Electron Microscope ◽

Scanning Electron Microscope ◽

High Voltage ◽

Amorphous Ribbon ◽

Image Resolution ◽

Objective Lens ◽

Magnetic Domains ◽

Scattered Electron ◽

Voltage Scanning ◽

Scanning Electron

Magnetic domains of ferromagnetic amorphous ribbon have been observed using Bitter powder method. However, the domains of amorphous ribbon are very complicated and the surface of ribbon is not flat, so that clear domain image has not been obtained. It has been desired to observe more clear image in order to analyze the domain structure of this zero magnetocrystalline anisotropy material. So, we tried to observe magnetic domains by means of a back-scattered electron mode of high voltage scanning electron microscope (HVSEM).HVSEM method has several advantages compared with the ordinary methods for observing domains: (1) high contrast (0.9, 1.5 and 5% at 50, 100 and 200 kV) (2) high penetration depth of electrons (0.2, 1.5 and 8 μm at 50, 100 and 200 kV). However, image resolution of previous HVSEM was quite low (maximum magnification was less than 100x), because the objective lens cannot be excited for avoiding the application of magnetic field on the specimen.

Download Full-text

Towards 1-Ångstrom-resolution STEM

Proceedings, annual meeting, Electron Microscopy Society of America ◽

10.1017/s0424820100150812 ◽

1993 ◽

Vol 51 ◽

pp. 996-997

Author(s):

H.S. von Harrach ◽

D.E. Jesson ◽

S.J. Pennycook

Keyword(s):

High Resolution ◽

Field Emission ◽

Phase Contrast ◽

Spherical Aberration ◽

A Priori ◽

Dark Field ◽

Image Resolution ◽

Objective Lens ◽

Annular Dark Field ◽

First Results

Phase contrast TEM has been the leading technique for high resolution imaging of materials for many years, whilst STEM has been the principal method for high-resolution microanalysis. However, it was demonstrated many years ago that low angle dark-field STEM imaging is a priori capable of almost 50% higher point resolution than coherent bright-field imaging (i.e. phase contrast TEM or STEM). This advantage was not exploited until Pennycook developed the high-angle annular dark-field (ADF) technique which can provide an incoherent image showing both high image resolution and atomic number contrast.This paper describes the design and first results of a 300kV field-emission STEM (VG Microscopes HB603U) which has improved ADF STEM image resolution towards the 1 angstrom target. The instrument uses a cold field-emission gun, generating a 300 kV beam of up to 1 μA from an 11-stage accelerator. The beam is focussed on to the specimen by two condensers and a condenser-objective lens with a spherical aberration coefficient of 1.0 mm.

Download Full-text

Computer-Assisted High-Re Solution TEM

Proceedings, annual meeting, Electron Microscopy Society of America ◽

10.1017/s0424820100179567 ◽

1990 ◽

Vol 48 (1) ◽

pp. 162-163

Author(s):

F.A. Ponce ◽

H. Hikashi

Keyword(s):

Signal To Noise Ratio ◽

Accurate Determination ◽

Computer Software ◽

Video Camera ◽

Image Contrast ◽

Objective Lens ◽

Computer Assisted ◽

Optical Parameters ◽

Astigmatism Correction

The determination of the atomic positions from HRTEM micrographs is only possible if the optical parameters are known to a certain accuracy, and reliable through-focus series are available to match the experimental images with calculated images of possible atomic models. The main limitation in interpreting images at the atomic level is the knowledge of the optical parameters such as beam alignment, astigmatism correction and defocus value. Under ordinary conditions, the uncertainty in these values is sufficiently large to prevent the accurate determination of the atomic positions. Therefore, in order to achieve the resolution power of the microscope (under 0.2nm) it is necessary to take extraordinary measures. The use of on line computers has been proposed [e.g.: 2-5] and used with certain amount of success.We have built a system that can perform operations in the range of one frame stored and analyzed per second. A schematic diagram of the system is shown in figure 1. A JEOL 4000EX microscope equipped with an external computer interface is directly linked to a SUN-3 computer. All electrical parameters in the microscope can be changed via this interface by the use of a set of commands. The image is received from a video camera. A commercial image processor improves the signal-to-noise ratio by recursively averaging with a time constant, usually set at 0.25 sec. The computer software is based on a multi-window system and is entirely mouse-driven. All operations can be performed by clicking the mouse on the appropiate windows and buttons. This capability leads to extreme friendliness, ease of operation, and high operator speeds. Image analysis can be done in various ways. Here, we have measured the image contrast and used it to optimize certain parameters. The system is designed to have instant access to: (a) x- and y- alignment coils, (b) x- and y- astigmatism correction coils, and (c) objective lens current. The algorithm is shown in figure 2. Figure 3 shows an example taken from a thin CdTe crystal. The image contrast is displayed for changing objective lens current (defocus value). The display is calibrated in angstroms. Images are stored on the disk and are accessible by clicking the data points in the graph. Some of the frame-store images are displayed in Fig. 4.

Download Full-text

Microspectroscopy Using a Solid Immersion Lens

Microscopy and Microanalysis ◽

10.1017/s1431927600026817 ◽

2001 ◽

Vol 7 (S2) ◽

pp. 148-149

Author(s):

C.D. Poweleit ◽

J Menéndez

Keyword(s):

Spatial Resolution ◽

Focal Plane ◽

Numerical Aperture ◽

Normal Incidence ◽

Spot Size ◽

Index Of Refraction ◽

Objective Lens ◽

Improvement Factor ◽

Oil Immersion ◽

Long Time

Oil immersion lenses have been used in optical microscopy for a long time. The light’s wavelength is decreased by the oil’s index of refraction n and this reduces the minimum spot size. Additionally, the oil medium allows a larger collection angle, thereby increasing the numerical aperture. The SIL is based on the same principle, but offers more flexibility because the higher index material is solid. in particular, SILs can be deployed in cryogenic environments. Using a hemispherical glass the spatial resolution is improved by a factor n with respect to the resolution obtained with the microscope’s objective lens alone. The improvement factor is equal to n2 for truncated spheres.As shown in Fig. 1, the hemisphere SIL is in contact with the sample and does not affect the position of the focal plane. The focused rays from the objective strike the lens at normal incidence, so that no refraction takes place.

Download Full-text

Retinal vessel segmentation with constrained-based nonnegative matrix factorization and 3D modified attention U-Net

EURASIP Journal on Image and Video Processing ◽

10.1186/s13640-021-00546-6 ◽

2021 ◽

Vol 2021 (1) ◽

Author(s):

Yang Yu ◽

Hongqing Zhu

Keyword(s):

Matrix Factorization ◽

Nonnegative Matrix Factorization ◽

Computational Cost ◽

Three Dimensional ◽

Nonnegative Matrix ◽

Retinal Vessel ◽

Classification Problem ◽

Image Resolution ◽

Retinal Vessels ◽

Segmentation Task

AbstractDue to the complex morphology and characteristic of retinal vessels, it remains challenging for most of the existing algorithms to accurately detect them. This paper proposes a supervised retinal vessels extraction scheme using constrained-based nonnegative matrix factorization (NMF) and three dimensional (3D) modified attention U-Net architecture. The proposed method detects the retinal vessels by three major steps. First, we perform Gaussian filter and gamma correction on the green channel of retinal images to suppress background noise and adjust the contrast of images. Then, the study develops a new within-class and between-class constrained NMF algorithm to extract neighborhood feature information of every pixel and reduce feature data dimension. By using these constraints, the method can effectively gather similar features within-class and discriminate features between-class to improve feature description ability for each pixel. Next, this study formulates segmentation task as a classification problem and solves it with a more contributing 3D modified attention U-Net as a two-label classifier for reducing computational cost. This proposed network contains an upsampling to raise image resolution before encoding and revert image to its original size with a downsampling after three max-pooling layers. Besides, the attention gate (AG) set in these layers contributes to more accurate segmentation by maintaining details while suppressing noises. Finally, the experimental results on three publicly available datasets DRIVE, STARE, and HRF demonstrate better performance than most existing methods.

Download Full-text