Article
Optics
Yilong Zhang, Yaoxiang Lu, Haixia Wang, Peng Chen, Ronghua Liang
Summary: The observation, statistics, and classification of marine plankton are crucial for marine ecological research. Digital holography technology, combined with deep learning techniques, can accurately predict marine plankton species without the need for reconstruction, significantly enhancing efficiency in processing raw holograms.
OPTICS AND LASER TECHNOLOGY
(2021)
Article
Computer Science, Artificial Intelligence
David Sosa-Trejo, Antonio Bandera, Martin Gonzalez, Santiago Hernandez-Leon
Summary: Since the 19th century, scientists have tried to quantify species distributions using techniques such as direct counting and microscopes. Automatic image processing and classification methods are now being utilized to avoid manual procedures for classifying marine plankton. This article summarizes the techniques proposed for classifying marine plankton from the beginning of this field to the present day, focusing on automatic methods that utilize image processing.
ARTIFICIAL INTELLIGENCE REVIEW
(2023)
Article
Computer Science, Information Systems
Junmin Leng, Honglian Li, Fubing Li
Summary: This paper introduces a novel method to reduce speckle noise in digital holography. The method involves obtaining a single-shot digital hologram using a conventional experiment setup and generating multiple holographic patterns using binary masks designed on a computer. Reconstructing these holographic patterns effectively suppresses speckle noise in the reconstruction image.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Optics
Hwihyeong Lee, Hee Kyung Ahn
Summary: In this study, a method for rapidly locating shifted spectra with subpixel accuracy in digital holographic microscopy is proposed, resulting in high-quality synthesized images. Compared to other methods, this approach requires less data and has a simpler process.
OPTICAL ENGINEERING
(2022)
Article
Food Science & Technology
Renato Pereira Lima, Alex Sandro Bezerra de Sousa, Patrick Abeli, Randolph M. Beaudry, Silvanda de Melo Silva
Summary: Fruit surface coatings can reduce gas exchange and alter the composition of the internal atmosphere. A method for estimating the proportion of a starch-based coated surface was developed using digital image processing. The addition of a surfactant and temperature adjustment can improve the adhesion of the coating to fruit surfaces. The accumulation of internal atmosphere in coated fruits depends on the type of fruit and the starch content of the coating.
Article
Geochemistry & Geophysics
Shaobo Xia, Sheng Xu, Ruisheng Wang, Jonathan Li, Guanghui Wang
Summary: This study presents a method to extract individual buildings from ALS point clouds using widely accessible polygonal footprints. The method can achieve high instance-level building mapping accuracy around 90% and future work will focus on improving classification errors in preprocessing, shape inconsistencies between point clouds and polygons, as well as building footprint delineation and updating in postprocessing.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Environmental Sciences
Thelma Panaiotis, Louis Caray-Counil, Ben Woodward, Moritz S. Schmid, Dominic Daprano, Sheng Tse Tsai, Christopher M. Sullivan, Robert K. Cowen, Jean-Olivier Irisson
Summary: Planktonic organisms are crucial components of oceanic ecosystems, and their study has been enhanced by the development of in situ imaging instruments. By focusing on plankton early in the data processing pipeline, particularly at the segmentation stage, researchers can improve the efficiency of extracting ecological information from in situ images. Various segmentation methods, including a combination of traditional thresholding and Convolutional Neural Networks (CNN), have been compared on ISIIS data, showing that the CNN-enhanced thresholding method can reduce the number of segments and increase precision.
FRONTIERS IN MARINE SCIENCE
(2022)
Article
Optics
Pavel A. Cheremkhin, Nikolay N. Evtikhiev, Alexander Kozlov, Vitaly V. Krasnov, Vladislav G. Rodin, Rostislav S. Starikov
Summary: This paper proposes an optical-digital method of noise suppression in digital holography. It registers a set of uncorrelated holograms and applies a 3D filter to the reconstructed images, significantly improving the quality of the images.
Article
Computer Science, Artificial Intelligence
Xi Fang, Xuanang Xu, James J. Xia, Thomas Sanford, Baris Turkbey, Sheng Xu, Bradford J. Wood, Pingkun Yan
Summary: In this paper, a set of shape description losses is proposed to supervise the training of segmentation networks. These losses extract and quantify geometric features of the targets, such as volume, surface area, center of mass, and bounding box. The results demonstrate that using these shape losses alone for weakly supervised learning can achieve promising performance.
MACHINE VISION AND APPLICATIONS
(2023)
Review
Computer Science, Artificial Intelligence
Jiawei Zhang, Chen Li, Md Mamunur Rahaman, Yudong Yao, Pingli Ma, Jinghua Zhang, Xin Zhao, Tao Jiang, Marcin Grzegorzek
Summary: This study investigates the development of microorganism counting methods using digital image analysis. The research highlights the efficiency of image analysis-based microorganism counting methods compared to traditional manual counting methods.
ARTIFICIAL INTELLIGENCE REVIEW
(2022)
Article
Physics, Multidisciplinary
Zhentao Pang, Hang Zhang, Yu Wang, Letian Zhang, Yingchun Wu, Xuecheng Wu
Summary: The study introduces a deep learning method called Mo-U-net for accurate particle detection in gel secondary breakup, showing higher area prediction accuracy and boundary prediction precision compared to adaptive-threshold method. Mo-U-net is also capable of providing consistent size distribution prediction with actual data in holographic particle diagnostics.
FRONTIERS IN PHYSICS
(2021)
Article
Computer Science, Information Systems
Ruyong Ren, Shaozhang Niu
Summary: This paper proposes a new method that combines image segmentation, guided filtering, and filter reconstruction to improve the quality of digital holographic reconstructed images. The proposed method performs excellently in preserving details and suppressing noise. Furthermore, a simple and convenient holographic reconstruction image quality enhancement system is developed, providing great help for non-algorithm physics researchers. Additionally, it is the first holographic reconstruction image enhancement system with a good effect and considerable market application value.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Optics
Lakshmi Kuruguntla, Vineela Chandra Dodda, Min Wan, Karthikeyan Elumalai, Sunil Chinnadurai, Inbarasan Muniraj, John T. Sheridan
Summary: This paper demonstrates a simplified method to generate a Fourier hologram from multiple two-dimensional perspective images under low light level imaging conditions. By using a fast dictionary learning technique for integral Fourier hologram reconstruction, the proposed method achieves good hologram quality even with fewer samples.
APPLIED PHYSICS B-LASERS AND OPTICS
(2022)
Review
Chemistry, Analytical
Jose Angel Picazo-Bueno, Martin Sanz, Luis Granero, Javier Garcia, Vicente Mico
Summary: Lensless holographic microscopy (LHM) is a promising label-free technique that provides high-quality imaging and adaptive magnification in a lens-free, compact, and cost-effective way. The compact sizes and reduced prices of LHMs make them ideal for point-of-care diagnosis and increase their usability in limited-resource laboratories, remote areas, and poor countries. MISHELF microscopy, a single-shot and phase-retrieved imaging technique employing multiple illumination/detection channels and a fast-iterative phase-retrieval algorithm, can provide excellent intensity and phase imaging.
Article
Radiology, Nuclear Medicine & Medical Imaging
Lekshmi Kalinathan, Ruba Soundar Kathavarayan
Summary: This paper presents an algorithm for accurately segmenting multiple nuclei from immunohistochemically stained liver images. The algorithm addresses the challenges of stains among multi-nucleated cells, poor contrast of cell cytoplasm, and presence of mucus, blood, and inflammatory cells. By using a two-step process and extracting combination features, the algorithm achieves an accuracy of 89.76%.
JOURNAL OF DIGITAL IMAGING
(2023)