• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.4
• /
• pp.980-997
• /
• 2024

To improve the brightness of images and reveal hidden information in dark areas is the main objective of low-light image enhancement (LLIE). LLIE methods based on deep learning show good performance. However, there are some limitations to these methods, such as the complex network model requires highly configurable environments, and deficient enhancement of edge details leads to blurring of the target content. Single-scale feature extraction results in the insufficient recovery of the hidden content of the enhanced images. This paper proposed an edge detection-based multi-scale feature enhancement network for LLIE (EDMFEN). To reduce the loss of edge details in the enhanced images, an edge extraction module consisting of a Sobel operator is introduced to obtain edge information by computing gradients of images. In addition, a multi-scale feature enhancement module (MSFEM) consisting of multi-scale feature extraction block (MSFEB) and a spatial attention mechanism is proposed to thoroughly recover the hidden content of the enhanced images and obtain richer features. Since the fused features may contain some useless information, the MSFEB is introduced so as to obtain the image features with different perceptual fields. To use the multi-scale features more effectively, a spatial attention mechanism module is used to retain the key features and improve the model performance after fusing multi-scale features. Experimental results on two datasets and five baseline datasets show that EDMFEN has good performance when compared with the stateof-the-art LLIE methods.

• FOCUS: LIFE SCIENCE
• /
• no.1
• /
• pp.3.1-3.7
• /
• 2024

The article discusses the critical role of chromatography in the analysis and purification of proteins in biopharmaceuticals, emphasizing the importance of comprehensive characterization for ensuring their safety and efficacy. It highlights the use of Avantor® ACE® HPLC columns for the separation and purification of proteins, focusing on the analysis of intact proteins using reversed-phase liquid chromatography (RPLC) with fully porous particles. This article also details the application of different mobile phase additives, such as TFA and formic acid, and emphasizes the advantages of using type B ultra-pure silica-based columns for efficiency and peak shape in biomolecule analysis. Additionally, it addresses the challenges of analyzing intact proteins due to slow molecular diffusion and introduces the concept of solid-core (or superficially porous) particles, emphasizing their benefits over traditional porous particles for the analysis of therapeutic proteins. Furthermore, it discusses the development of Avantor® ACE® UltraCore BIO columns, specifically designed for the high-efficiency separation of large biomolecules, such as proteins, and demonstrates their effectiveness in achieving high-resolution separations, even for higher molecular weight proteins like monoclonal antibodies (mAbs). In addition, it underscores the complexity of analyzing and characterizing intact protein biopharmaceuticals, requiring a range of analytical techniques and the use of wide-pore stationary phases, operated at elevated temperatures and with relatively shallow gradients. It highlights the comprehensive range of options offered by Avantor® ACE® wide pore columns, including both fully porous and solid-core particles, bonded with a variety of complementary stationary phase chemistries to optimize selectivity during method development. The use of ultrapure and highly inert base silica is emphasized for enabling the use of lower concentrations of mobile phase modifiers without compromising analyte peak shape, particularly beneficial for LC-MS applications. Then the article concludes by emphasizing the significance of reversed-phase liquid chromatography and its compatibility with mass spectrometry as a valuable tool for the separation and analysis of intact proteins and their closely related variants in biopharmaceuticals.

• Archives of Plastic Surgery
• /
• v.51 no.5
• /
• pp.504-509
• /
• 2024

The treatment of breast cancer has seen great success in the recent decade. With longer survivorship, more attention is paid to function and aesthetics as integral treatment components. However, breast cancer-related lymphedema (BCRL) remains a significant complication. Immediate lymphatic reconstruction is an emerging technique to reduce the risk of BCRL, the Lymphatic Microsurgical Preventive Healing Approach (LYMPHA) being the most widely used approach. Despite promising results, it is often difficult to find suitably sized recipient venules and perform the micro-anastomoses between mismatched vessels deep in the axilla. Moreover, high axillary venous pressure gradients and potential damage from radiotherapy may affect the long-term patency of the anastomoses. From an ergonomic point of view, performing lymphaticovenular anastomosis in the deep axilla may be challenging for the microsurgeon. In response to these limitations, we modified the technique by moving the lymphatic reconstruction distally-terming it distally based LYMPHA (dLYMPHA). A total of 113 patients underwent mastectomy with axillary clearance in our institution from 2018 to 2021. Of these, 26 underwent subsequent dLYMPHA (Group 2), whereas 87 did not (Group 1). In total, 17.2% (15 patients) and 3.84% (1 patient) developed BCRL in Groups 1 and 2, respectively (p = 0.018). Lymphatics and recipient venules suitable for anastomoses can be reliably found in the distal upper limb with better size match. A distal modification achieves a more favorable lymphaticovenular pressure gradient, vessel match, and ergonomics while ensuring a comparably low BCRL rate.

• Annual Conference of KIPS
• /
• 2007.05a
• /
• pp.333-336
• /
• 2007

We propose a simple method for tracking a nonparameterized subject contour in a single video stream with a moving camera and changing background. Then we present a method to eliminate the tracked contour object by replacing with the background scene we get from other frame. Our method consists of two parts: first we track the object using LOD (Level-of-Detail) canny edge maps, then we generate background of each image frame and replace the tracked object in a scene by a background image from other frame that is not occluded by the tracked object. Our tracking method is based on level-of-detail (LOD) modified Canny edge maps and graph-based routing operations on the LOD maps. To reduce side-effects because of irrelevant edges, we start our basic tracking by using strong Canny edges generated from large image intensity gradients of an input image. We get more edge pixels along LOD hierarchy. LOD Canny edge pixels become nodes in routing, and LOD values of adjacent edge pixels determine routing costs between the nodes. We find the best route to follow Canny edge pixels favoring stronger Canny edge pixels. Our accurate tracking is based on reducing effects from irrelevant edges by selecting the stronger edge pixels, thereby relying on the current frame edge pixel as much as possible. This approach is based on computing camera motion. Our experimental results show that our method works nice for moderate camera movement with small object shape changes.

• Nuclear Engineering and Technology
• /
• v.56 no.10
• /
• pp.4237-4246
• /
• 2024

Using strong electromagnetic fields generated by lasers to interact with electrons for precise diagnosis and manipulation of electron beams represents a recent focal point in accelerator technology. This approach surpasses the limitations of conventional RF technology, such as low electric field gradients and timing jitters, effectively enhancing the accuracy of ultrafast electron beam diagnostics and manipulations. As demands for precision continue to rise, the precise diagnosis of crucial parameters of ultrafast electron beams remains challenging. This study delves into the electromagnetic behavior of THz-driven devices and proposes an all-optical method utilizing single-cycle THz radiation to compress and characterize a 3 MeV electron beam. Particle tracking simulations demonstrate an astonishing compression effect, reducing the bunch length from 54.0 fs to 4.3 fs, and achieving sub-femtosecond bunch length measurement resolution. Moreover, when combined with an orthogonal THz streak camera, this method shows even greater potential in multi-bunch scenarios.

• Animal Bioscience
• /
• v.37 no.12
• /
• pp.2189-2201
• /
• 2024

Objective: Despite the considerable body of research on the effects of heat stress coupled with water scarcity (either through restriction or deprivation) on goats, aimed at enhancing their welfare, there remains a notable gap in the literature regarding the subsequent period following water restoration, during which the cumulative impact is fully alleviated. In response to this gap, we propose a strategy grounded in the assessment of body-thermal status to improve the welfare of heat-stressed and water-deprived goats. Specifically, our strategy seeks to determine the minimally required recovery interval necessary to completely mitigate the residual effects of water deprivation endured for a duration of 72 hours. Methods: Eight healthy Aardi bucks, aged 10 months and weighing 30 kg, were subjected to three distinct stages: euhydration, dehydration, and rehydration. Each stage spanned for 72 hours except for the rehydration stage, which was left unrestricted. Various meteorological, biophysiological, and thermophysiological measurements were subsequently recorded. Results: Exposure of heat-stressed goats, as indicated by the temperature-humidity index values, to a 72 hours deprivation period resulted in noticeable (p<0.05) alterations in their biophysiological (daily feed intake, body weight, and feces water content) and thermophysiological responses (core, rectal, skin, and surface temperatures, respiratory and heart rates, internal, external, and total body-thermal gradients, heat tolerance and adaptability coefficients, heterothermial total body-heat storage, and total water conservation). Remarkably, our findings demonstrate that all assessed variables, whether measured or estimated, returned to their baseline euhydration levels within 10 days of commencing the rehydration phase. Conclusion: In order to improve the welfare of heat-stressed and 72 hours water-deprived goats, it is imperative to allow a recovery period of no less than 10 days following the restoration of water access prior to initiating any subsequent experiments involving these animals. Such experiments, addressing these critical aspects, serve to advance our understanding of goat welfare and obviously hold promise for contributing to future food security and economic viability.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.36 no.2
• /
• pp.105-116
• /
• 2000

In order to investigate the seasonal variation of watermass in the central coast of the southern sea of korea, oceanographic observation on the fishing grounds were carried out by the trainingship of Yosu University on May, Aug. and Nov. in 1998 and Feb. in 1999. The resultes obtained are summerized as follows : 1). The watermass in the fishing ground were divided into the coastal water(30.0~31.6$\textperthousand$ ), mixing water(31.7~33.4$\textperthousand$) and the offshore water(33.5~35.0$\textperthousand$) according to the distribution of salinity from T-S diagram plotted all salinity data observed on May, Aug. and Nov. in 1998 and Feb. in 1999. 2) The ranges of temperature and salinity were from 14.1$^{\circ}C$ to 18.8$^{\circ}C$ and from 32.2$\textperthousand$ to 34.9$\textperthousand$ in spring(May), from 14.2$^{\circ}C$ to 27.7$^{\circ}C$ and from 29.0$\textperthousand$ to 34.7$\textperthousand$ in summer(August), from 13.4$^{\circ}C$ to 21.3$^{\circ}C$ and from 31.45$\textperthousand$ to 34.5$\textperthousand$ in autumn(November) and from 8.2$^{\circ}C$ to 14.8$^{\circ}C$ and from 33.9$\textperthousand$ to 34.6$\textperthousand$ in winter(February), respectively. 3) The distribution of watermass in the fishing ground varied largely each seasons, but a general tendency on the distribution was obtained. That is, in spring and autumm the offshore water was distributed most widely and in summer the coastal and mixing water occupied the fishing ground but in winter the offshore water prevailed. 4) Variation of temperature and salinity were appeared between the surface and 30m in the coastal region and between the surface and 50m in the open ocaen region. Therefore, in the summer the thermocline and halocline were made between surface and 30m layer with vertical gradients of 10.5$^{\circ}C$/30m and 4.0$\textperthousand$/30m in the coastal region and in the open ocean region the thermocline and halocline were made between surface and 50m layer with vertical gradients of 13.$0^{\circ}C$/50m and 3.8$\textperthousand$/50m.

• Journal of Korean Society of Forest Science
• /
• v.81 no.4
• /
• pp.363-371
• /
• 1992

This study was aimed at developing a computer model to determine rational road networks in mountainous forests. The computer model is composed of two major subroutines for digital terrain analyses and route selection. The digital terrain model(DTM) provides various information on topographic and vegetative characteristics of forest stands. The DTM also evaluates the effectiveness of road construction based on slope gradients. Using the results of digital terrain analyses, the route selection subroutine, heuristically, determines the optimal road layout satisfying the predefined road densities. The route selection subroutine uses the area-partitioning method in order to fully of roads. This method leads to unbiased road layouts in forest areas. The size of the unit partitiones area can be calculated as a function of the predefined road density. In addition, the user-defined road density of the area-partitioning method provides flexibility in applying the model to real situations. The rational road network can be easily achived for varying road densities, which would be an essential element for network design of forest roads. The optimality conditions are evaluated in conjuction with longitudinal gradients, investment efficiency earthwork quantity or the mixed criteria of these three. The performance of the model was measured and, then, compared with those of conventional ones in terns of average skidding distance, accessibility of stands, development index and circulated road network index. The results of the performance analysis indicate that selection of roading routes for network design using the digital terrain analysis and the area-partitioning method improves performance of the network design medel.

• Korean Journal of Ecology and Environment
• /
• v.40 no.2
• /
• pp.285-293
• /
• 2007

The study was to determine zonal characteristics of nutrients and chlorophyll and evaluate their trophic relations in Daechung Reservoir. For this study, we compared longterm water quality data among three zones along with trophic state using 1993 to 2002 dataset, obtained from the Ministry of Environment, Korea. Total phosphorous (TP), Secchi depth (SD) and chlorophyll (CHL) showed typical longitudinal declines from the riverine to lacustrine zone, but total nitrogen (TN) was not evident. Largest seasonal variations in TP and CHL occurred during the summer monsoon from July to August. In the reservoir, ambient TN averaged 1.67 mg $L^{-1}$ and ratios of TN : TP averaged 88.04, indicating that nitrogen is not likely limited but phosphorus limitation was evident. Trophic State Index (TSI), based on CHL, TP, and SD, varied depending on the zones and seasons. Mean TSI (TP) in the riverine zone was 62 during the monsoon, indicating a hypertrophic condition, whereas the mean was 40 in the lacustrine, indicating a nearly oligotrophic. Values of TSI (CHL) showed maximum in the transition zone during the monsoon. The deviation analysis of TSI showed that about 65% of TSI (CHL)-TSI (TP) and TSI (CHL)-TSI (SD) values were less than zero and the lowest values were -42, indicating an effect of inorganic turbidity on algal growth in the reservoir. Correlation analysis of CHL vs. SD shewed greater correlation coefficient (p<0.001, r=-0.47) in the transition than other two zones (p<0.001, $r{\leq}-0.40$). Correlation analysis of TP vs. CHL was greatest in the lacustrine and TP was minimum in the lacustrine zone, indicating a lowest yield of algal biomass in the lacustrine. Overall data suggests that zonal response of chlorophyll yield at a given nutrient unit is clearly differed among the longitudinal gradients, so the management strategy such as cross sectional modelling should be provided in each zone.

• Journal of Chest Surgery
• /
• v.31 no.8
• /
• pp.749-755
• /
• 1998

Background: Twelve patients with acyanotic tetralogy of Fallot(TOF), characterized by the combination of a malaligned ventricular septal defect(VSD) and infundibular pulmonic stenosis with the clinical finding of acyanosis at rest, underwent surgical correction between January 1988 and July 1997. Materials and methods: 9.92% of patients with the diagnosis of TOF were acyanotic TOF in the same period. Ages ranged from 12 to 42 months(mean 25.2 months). 2D-echocardiographic studies, cardiac catheterization, and angiocardiograms were performed in all patients before operation. The preoperative mean systemic arterial oxygen saturation was 93.5%. According to the 2D-echocardiographic analysis, there was Lt-to-Rt shunt through VSD in 4 patients, bidirectional shunt in 2 patients, and no shunt in 6 patients. Results: The preoperative mean right ventricle to pulmonary artery(RV-PA) pressure gradients were 52.3 mmHg on 2D- echocardiogram and 48.4 mmHg on cardiac catheterization. The repair of ventricular septal defect was performed through a right atrial approach and the hypertrophic infundibular muscle bundles were resected by the transatrial and transpulmonary approach. Six patients(50%) received a transannular patch. The mean cardiopulmonary bypass time was 135.0 minutes, and the aortic crossclamp time was 87.8 minutes. Postoperative complications included bleeding necessitating reentry in one and chylothorax in one. No patient died after operation and there were no late deaths. Postoperative 2D-echocardiograms revealed tiny patch dehiscence in 5 cases and a moderate RV-PA pressure gradients(mean 15.3 mmHg). All patients were in New York Heart Association functional class 1 after operation. Conclusions: acyanotic TOF is the uncommon form of TOF, and acyanotic TOF can be repaired with a good outcome.