• Journal of Korea Multimedia Society
• /
• v.25 no.1
• /
• pp.1-9
• /
• 2022

Ship detection in synthetic aperture radar (SAR) images is an important application in marine monitoring for the military and civilian domains. Over the past decade, object detection has achieved significant progress with the development of convolutional neural networks (CNNs) and lot of labeled databases. However, due to difficulty in collecting and labeling SAR images, it is still a challenging task to solve SAR ship detection CNNs. To overcome the problem, some methods have employed conventional data augmentation techniques such as flipping, cropping, and affine transformation, but it is insufficient to achieve robust performance to handle a wide variety of types of ships. In this paper, we present a novel and effective approach for deep SAR ship detection, that exploits label-rich Electro-Optical (EO) images. The proposed method consists of two components: a data augmentation network and a ship detection network. First, we train the data augmentation network based on conditional generative adversarial network (cGAN), which aims to generate additional SAR images from EO images. Since it is trained using unpaired EO and SAR images, we impose the cycle-consistency loss to preserve the structural information while translating the characteristics of the images. After training the data augmentation network, we leverage the augmented dataset constituted with real and translated SAR images to train the ship detection network. The experimental results include qualitative evaluation of the translated SAR images and the comparison of detection performance of the networks, trained with non-augmented and augmented dataset, which demonstrates the effectiveness of the proposed framework.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.164-164
• /
• 2023

This study investigates the importance of flood-influencing factors on the accuracy of flood risk mapping using the integration of remote sensing-based and machine learning techniques. Here, the Extreme Gradient Boosting (XGBoost) and Random Forest (RF) algorithms integrated with GIS-based techniques were considered to develop and generate flood risk maps. For the study area of NAPA County in the United States, rainfall data from the 12 stations, Sentinel-1 SAR, and Sentinel-2 optical images were applied to extract 13 flood-influencing factors including altitude, aspect, slope, topographic wetness index, normalized difference vegetation index, stream power index, sediment transport index, land use/land cover, terrain roughness index, distance from the river, soil, rainfall, and geology. These 13 raster maps were used as input data for the XGBoost and RF algorithms for modeling flood-prone areas using ArcGIS, Python, and R. As results, it indicates that XGBoost showed better performance than RF in modeling flood-prone areas with an ROC of 97.45%, Kappa of 93.65%, and accuracy score of 96.83% compared to RF's 82.21%, 70.54%, and 88%, respectively. In conclusion, XGBoost is more efficient than RF for flood risk mapping and can be potentially utilized for flood mitigation strategies. It should be noted that all flood influencing factors had a positive effect, but altitude, slope, and rainfall were the most influential features in modeling flood risk maps using XGBoost.

• Geomechanics and Engineering
• /
• v.33 no.2
• /
• pp.175-182
• /
• 2023

In general, the design of structures and its construction processes are fundamentally dependent on their foundation and supporting ground. Thus, it is imperative to understand the behavior of the soil under certain stress and drainage conditions. As it is well known that certain characteristics and behaviors of soils with fines are highly dependent on water content, it is critical to accurately measure and identify the status of the soils in terms of water contents. Liquid limit is one of the important soil index properties to define such characteristics. However, liquid limit measurement can be affected by the proficiency of the operator. On the other hand, dynamic properties of soils are also necessary in many different applications and current testing methods often require special equipment in the laboratory, which is often expensive and sensitive to test conditions. In order to address these concerns and advance the state of the art, this study explores a novel method to determine the liquid limit of cohesive soil by employing video-based vibration analysis. In this research, the modal characteristics of cohesive soil columns are extracted from videos by utilizing phase-based motion estimation. By utilizing the proposed method that analyzes the optical flow in every pixel of the series of frames that effectively represents the motion of corresponding points of the soil specimen, the vibration characteristics of the entire soil specimen could be assessed in a non-contact and non-destructive manner. The experimental investigation results compared with the liquid limit determined by the standard method verify that the proposed method reliably and straightforwardly identifies the liquid limit of clay. It is envisioned that the proposed approach could be applied to measuring liquid limit of soil in practical field, entertaining its simple implementation that only requires a digital camera or even a smartphone without the need for special equipment that may be subject to the proficiency of the operator.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.7
• /
• pp.1807-1822
• /
• 2023

Innovation and rapid increased functionality in user friendly smartphones has encouraged shutterbugs to have picturesque image macros while in work environment or during travel. Formal signboards are placed with marketing objectives and are enriched with text for attracting people. Extracting and recognition of the text from natural images is an emerging research issue and needs consideration. When compared to conventional optical character recognition (OCR), the complex background, implicit noise, lighting, and orientation of these scenic text photos make this problem more difficult. Arabic language text scene extraction and recognition adds a number of complications and difficulties. The method described in this paper uses a two-phase methodology to extract Arabic text and word boundaries awareness from scenic images with varying text orientations. The first stage uses a convolution autoencoder, and the second uses Arabic Character Segmentation (ACS), which is followed by traditional two-layer neural networks for recognition. This study presents the way that how can an Arabic training and synthetic dataset be created for exemplify the superimposed text in different scene images. For this purpose a dataset of size 10K of cropped images has been created in the detection phase wherein Arabic text was found and 127k Arabic character dataset for the recognition phase. The phase-1 labels were generated from an Arabic corpus of quotes and sentences, which consists of 15kquotes and sentences. This study ensures that Arabic Word Awareness Region Detection (AWARD) approach with high flexibility in identifying complex Arabic text scene images, such as texts that are arbitrarily oriented, curved, or deformed, is used to detect these texts. Our research after experimentations shows that the system has a 91.8% word segmentation accuracy and a 94.2% character recognition accuracy. We believe in the future that the researchers will excel in the field of image processing while treating text images to improve or reduce noise by processing scene images in any language by enhancing the functionality of VGG-16 based model using Neural Networks.

• Advances in nano research
• /
• v.15 no.1
• /
• pp.35-48
• /
• 2023

Nanoparticles are known for their outstanding properties such as particle size, surface area, optical and electrical properties. These properties have significantly boasted their applications in various surface phenomena. In this work, calcium oxide nanoparticles were synthesized from periwinkle shells as an approach towards waste management through resource recovery. The sol gel method was used for the synthesis. The nanoparticles were characterized using X-Ray diffractometer (XRD), Fourier Transformed Infra-Red Spectrophotometer (FTIR), Brunauer Emmett Teller (BET), scanning electron microscopy (SEM), transmission electron microscopy (TEM) and ultra violet visible spectrophotometer (UV-visible). While DLS and SEM underestimate the particle diameter, the BET analysis reveals surface area of 138.998 m²/g, pore volume = 0.167 m³/g and pore diameter of 2.47 nm. The nanoparticles were also employed as an adsorbent for the purification of dye (methyl orange) contaminated water. The adsorbent showed excellent removal efficiency (up to 97 %) for the dye through the mechanism of physical adsorption. The adsorption of the dye fitted the Langmuir and Temkin models. Analysis of FTIR spectrum after adsorption complemented with computational chemistry modelling to reveal the imine nitrogen group as the site for the adsorption of the dye unto the nanomaterials. The synthesized nanomaterials have an average particle size of 24 nm, showed a unique XRD peak and is thermally and mechanically stable within the investigated temperature range (30 to 70 ℃).

• Journal of Convergence for Information Technology
• /
• v.12 no.2
• /
• pp.206-212
• /
• 2022

This study suggests art makeup design by utilizing the characteristics of psychedelic works. The study analyzed the concept of being psychedelic and art makeup based on precedent studies and publications, and classified the psychedelic characteristics as being those of mysterious, optical illusions, and playfulness. After selecting the work featuring the characteristics of being psychedelic, six works were created with this motif, by integrating the airbrush, the pictorial, and the object techniques of the art makeup expression techniques. As a result, it was demonstrated that the design that has utilized the characteristics of psychedelic works could be applicable to art makeup. Next, the confluence of art works and art makeup can serve as an approach to drawing creative designs. Thus, this study suggests a direction for extending the creativity and variety of art makeup design, and a continued development for art makeup design by further confluences with various areas is expected

• Korean Journal of Materials Research
• /
• v.33 no.7
• /
• pp.315-322
• /
• 2023

In the present work, we address the new route for the green synthesis of manganese dioxide (MnO₂) by an innovative method named the solution plasma process (SPP). The reaction mechanism of both colloidal and nanostructured MnO₂ was investigated. Firstly, colloidal MnO₂ was synthesized by plasma discharging in KMnO₄ aqueous solution without any additives such as reducing agents, acids, or base chemicals. As a function of the discharge time, the purple color solution of MnO₄^- (oxidation state +7) was changed to the brown color of MnO₂ (oxidation state +4) and then light yellow of Mn²⁺ (oxidation state +2). Based on the UV-vis analysis we found the optimal discharging time for the synthesis of stable colloidal MnO₂ and also reaction mechanism was verified by optical emission spectroscopy (OES) analysis. Secondly, MnO₂ nanoparticles were synthesized by SPP with a small amount of reducing sugar. The precipitation of brown color was observed after 8 min of plasma discharge and then completely separated into colorless solution and precipitation. It was confirmed layered type of nanoporous birnessite-MnO₂ by X-ray powder diffraction (XRD), fourier-transform infrared spectroscopy (FT-IR), and electron microscopes. The most important merits of this approach are environmentally friendly process within a short time compared to the conventional method. Moreover, the morphology and the microstructure could be controllable by discharge conditions for the appropriate potential applications, such as secondary batteries, supercapacitors, adsorbents, and catalysts.

• Advances in nano research
• /
• v.15 no.5
• /
• pp.451-466
• /
• 2023

Gastrointestinal cancer (GC) is a prevalent malignant tumor of the digestive system that poses a severe health risk to humans. Due to the specific organ structure of the gastrointestinal system, both endoscopic and MRI diagnoses of GIC have limited sensitivity. The primary factors influencing curative efficacy in GIC patients are drug inefficacy and high recurrence rates in surgical and pharmacological therapy. Due to its unique optical features, good biocompatibility, surface effects, and small size effects, nanotechnology is a developing and advanced area of study for the detection and treatment of cancer. Because of its deep location and complex surgery, diagnosing and treating gastrointestinal cancer is very difficult. The early diagnosis and urgent treatment of gastrointestinal illness are enabled by nanotechnology. As diagnostic and therapeutic tools, nanoparticles directly target tumor cells, allowing their detection and removal. XGBoost was used as a classification method known for achieving numerous winning solutions in data analysis competitions, to capture nonlinear relations among many input variables and outcomes using the boosting approach to machine learning. The research sample included 300 GC patients, comprising 190 males (72.2% of the sample) and 110 women (27.8%). Using convolutional neural networks (CNN) and artificial neural networks (ANN)-EXtreme Gradient Boosting (XGBoost), the patients mean± SD age was 50.42 ± 13.06. High-risk behaviors (P = 0.070), age at diagnosis (P = 0.037), distant metastasis (P = 0.004), and tumor stage (P = 0.015) were shown to have a statistically significant link with GC patient survival. AUC was 0.92, sensitivity was 81.5%, specificity was 90.5%, and accuracy was 84.7 when analyzing stomach picture.

• Current Optics and Photonics
• /
• v.8 no.1
• /
• pp.45-55
• /
• 2024

A fiber spool with ultra-low vibration sensitivity has been demonstrated for the ultra-narrow-linewidth fiber-stabilized laser by the multi-object orthogonal experimental design method, which can achieve the optimization object and analysis of influence levels without extensive computation. According to a test of 4 levels and 4 factors, an L₁₆ (4⁴) orthogonal table is established to design orthogonal experiments. The vibration sensitivities along the axial and radial directions and the normalized sums of the vibration sensitivities are determined as single objects and comprehensive objects, respectively. We adopt the range analysis of object values to obtain the influence levels of the four design parameters on the single objects and the comprehensive object. The optimal parameter combinations are determined by both methods of comprehensive balance and evaluation. Based on the corresponding fractional frequency stability of ultra-narrow-linewidth fiber-stabilized lasers, we obtain the final optimal parameter combination A3B1C2D1, which can achieve the fiber spool with vibration sensitivities of 10^-12/g magnitude. This work is the first time to use an orthogonal experimental design method to optimize the vibration sensitivities of fiber spools, providing an approach to design the fiber spool with ultra-low vibration sensitivity.

• Microbiology and Biotechnology Letters
• /
• v.51 no.4
• /
• pp.403-415
• /
• 2023

This study aimed to isolate and identify L-(+)-lactic acid-producing bacteria from tree barks collected in Thailand and evaluate the potential strain as probiotics. Twelve strains were isolated and characterized phenotypically and genotypically. The strains exhibited a rod-shaped morphology, high-temperature tolerance, and the ability to ferment different sugars into lactic acid. Based on 16S rRNA gene analysis, all strains were identified as belonging to Weizmannia coagulans. Among the isolated strains, BKMTCR2-2 demonstrated exceptional lactic acid production, with 96.41% optical purity, 2.33 g/l of lactic acid production, 1.44 g/g of lactic acid yield (per gram of glucose consumption), and 0.0049 g/l/h of lactic acid productivity. This strain also displayed a wide range of pH tolerance, suggesting suitability for the human gastrointestinal tract and potential probiotic applications. The whole-genome sequence of BKMTCR2-2 was assembled using a hybridization approach that combined long and short reads. The genomic analysis confirmed its identification as W. coagulans and safety assessments revealed its non-pathogenic attribute compared to type strains and commercial probiotic strains. Furthermore, this strain exhibited resilience to acidic and bile conditions, along with the presence of potential probiotic-related genes and metabolic capabilities. These findings suggest that BKMTCR2-2 holds promise as a safe and effective probiotic strain with significant lactic acid production capabilities.