Article
Energy & Fuels
Leyla Ismailova, Victoria Dochkina, Mustafa Al Ibrahim, Mokhles Mezghani
Summary: A practical automated method for estimating drill cuttings particle sizes using digital images is proposed. The method combines several well-known techniques and has been tested on images from wells with varying lithologies and particle sizes. The results show that the method is robust and provides accurate measurements comparable to manual measurements, with a fast processing speed suitable for near real-time applications.
JOURNAL OF PETROLEUM SCIENCE AND ENGINEERING
(2022)
Article
Chemistry, Analytical
Nan Wu, Shinichiro Haruyama
Summary: Using a one-dimensional laser speckle image to measure acoustic vibrations, this study introduces a fast and sub-pixel accuracy algorithm for estimating displacement, achieving real-time reconstruction of audio signals with increased frequency bandwidth and robustness compared to previous research. Experimental results demonstrate a sampling rate of 20k samples-per-second and high-quality real-time audio signal reconstruction.
Article
Computer Science, Information Systems
Obed Appiah, Michael Asante, James Benjamin Hayfron-Acquah
Summary: Median filter is a predominant filter used to suppress impulse noise with variations focusing on generating quality outputs or reducing running time. The proposed Improved Approximation Median Filtering Algorithms show advantages in both output quality and running time.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Li Guo, Yonghong Peng, Rui Qin, Bingyu Liu
Summary: This paper presents the use of deep learning models for direct ore feed load estimation from ore pellet images in order to overcome the challenges in obtaining ore characteristics in production environments. Through a weakly supervised learning approach and a two-stage model training algorithm, competitive model performance has been achieved.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Automation & Control Systems
Levent Subasi, Mustafa Gokler, Ulas Yaman
Summary: Waterjet-guided laser machining is a novel process for precision cutting and drilling, but noise factors affect the characteristics of the laser beam. This study proposes two real-time measurement methods that can accurately measure the laser beam characteristics.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2022)
Article
Chemistry, Multidisciplinary
Ricardo Oliva-Garcia, Sabato Ceruso, Jose G. Marichal-Hernandez, Jose M. Rodriguez-Ramos
Summary: This work presents a real-time full-resolution depth estimation device that enables the generation of high-resolution depth maps in real time, without the need for custom hardware. The method is compared to other algorithms and shows advantages in computational time and precision.
APPLIED SCIENCES-BASEL
(2022)
Article
Mathematics, Interdisciplinary Applications
SungHyok Ro, JinHyok Jon, KumSong Ryu
Summary: This paper introduces an improved method of microscopic image analysis to enhance the accuracy of estimating the particle size distribution of comminuted minerals. The image processing methods used include microscopic image stitching, individual particle segmentation with watershed and deep learning, and individual particle weight estimation with shape factor. Through comparison with sieving test, it is confirmed that this method improves the estimation ability of particle size distribution for samples such as ferruginous quartzite, coal, and magnetite.
COMPUTATIONAL PARTICLE MECHANICS
(2023)
Article
Computer Science, Artificial Intelligence
Huitong Yang, Liang Lei, Haiwei Sang
Summary: In this paper, a novel stereo network GAMNet is proposed to enhance the accuracy of stereo-based disparity estimation. The network consists of a lightweight attention module, an MPF module, and a stacked encoder-decoder with the DCA module and 3D convolutions. Experimental results demonstrate that GAMNet outperforms previous methods and a new 3D scene reconstruction evaluation strategy is proposed.
IET IMAGE PROCESSING
(2023)
Article
Immunology
Guolan Lu, Marc A. Baertsch, John W. Hickey, Yury Goltsev, Andrew J. Rech, Lucas Mani, Erna Forgo, Christina Kong, Sizun Jiang, Garry P. Nolan, Eben L. Rosenthal
Summary: Highly multiplexed single-cell imaging has advanced our understanding of spatial cellular interactions. RAPID, a real-time, GPU-accelerated image processing software, improves data processing efficiency and image quality for large-scale multiplexed fluorescence microscopy data.
FRONTIERS IN IMMUNOLOGY
(2022)
Article
Engineering, Electrical & Electronic
Anders Astrom, Robert Forchheimer
Summary: We propose a method to estimate the time-to-impact (TTI) by detecting and tracking local extremal points in a sequence of images. This method, different from the ordinary optical flow technique, allows for fast and low-complex processing, inspired by the TTI capability of insects. It is also suitable for near-sensor image processing architectures.
JOURNAL OF ELECTRONIC IMAGING
(2022)
Article
Computer Science, Information Systems
Xiangyu Zhou, Zhongjie Xu, Xiangai Cheng, Zhongyang Xing
Summary: This study proposes a deep learning model and countermeasure network for restoring laser interference images, and demonstrates its effectiveness and superiority on various datasets through experiments.
Article
Construction & Building Technology
Mohammadreza Pourebrahimi, Vahid Shahhosseini, Ali Akbar Ramezanianpour
Summary: Aggregate size distribution is crucial for concrete strength and workability. A new method utilizing macrostructure and microstructure images of hardened concrete was developed to estimate this distribution accurately. The method analyzes macrostructure images to extract coarser aggregates and simulates mechanical sieving based on relative motions of the aggregates and apertures. Geometric rules are applied to optimize computational costs. Microstructure images are used to calculate air voids area ratio and hardened cement components area ratios, enabling the estimation of the volume ratio of finer aggregates. The proposed method performs brilliantly in estimating aggregate size distribution, as confirmed by comparison with actual grading.
CONSTRUCTION AND BUILDING MATERIALS
(2023)
Article
Engineering, Civil
Qingfeng Tang, Xiukun Wei, Dehua Wei, Xing Shen, Xinqiang Yin, Diqing Wang, Limin Jia, Qitian Zhong
Summary: This paper proposes an innovative real-time high-precision lightweight approach to estimate the wear of pantograph sliders. In the first stage, a method based on image processing and deep learning is used to locate the region of the slider, taking into account its unique characteristics. In the second stage, a neural network is proposed to estimate the wear of the slider, calculating the residual thickness by counting pixels and analyzing the error.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Dai Duong Nguyen, Abdelhafid El Ouardi, Sergio Rodriguez, Samir Bouaziz
Summary: Feature extraction algorithm dedicated for SLAM applications is implemented on SoC-FPGA using high-level synthesis, integrating HOOFR extraction algorithm and bucketing detection method to improve system performance without simplifying the original algorithm. The design achieves a throughput of 54 fps at 1226 x 370 pixels or 14 fps at 1920 x 1080 pixels on Intel Arria 10 SoC-FPGA.
JOURNAL OF REAL-TIME IMAGE PROCESSING
(2021)
Article
Computer Science, Information Systems
Juan Wang, Chang Ding, Minghu Wu, Yuanyuan Liu, Guanhai Chen
Summary: This study introduces a lightweight and fast dehazing network based on a multiple scale-patch framework, which effectively removes haze with superior performance in speed and effectiveness.
KSII TRANSACTIONS ON INTERNET AND INFORMATION SYSTEMS
(2021)