• Ocean Systems Engineering
• /
• v.7 no.2
• /
• pp.143-155
• /
• 2017

For complex flexible structures such as nets, the determination of drag forces and its deformation is a challenging task. The accurate prediction of loads on cages is one of the key steps in designing fish farm facilities. The basic physics with a simple cage, can be addressed by the use of experimental studies. However, to design more complex cage system for various environmental conditions, a reliable numerical simulation tool is essential. In this work, the current load on a cage is calculated using a Morison-force model applied at instantaneous positions of equivalent-net modeling. Variations of solidity ratio ($S_n$) of the net and current speed are considered. An equivalent array of cylinders is built to represent the physical netting. Based on the systematic comparisons between the published experimental data for Raschel nets and the current numerical simulations, carried out using the commercial software OrcaFlex, a new formulation for $C_d$ values, used in the equivalent-net model, is presented. The similar approach can also be applied to other netting materials following the same procedure. In case of high solidity ratio and current speed, the hybrid model defines $C_d$ as a function of Re (Reynolds number) and $S_n$ to better represent the corresponding weak diffraction effects. Otherwise, the conventional $C_d$ values depending only on Re can be used with including shielding effects for downstream elements. This new methodology significantly improves the agreement between numerical and experimental data.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.43 no.2
• /
• pp.101-115
• /
• 2007

The pound net fishery is very important one in Korean coastal fishery and it need to grasp the characteristics of the net affected by many factors. It is considered that the structure and the shape of the pound net can be changed by the direction and speed of current, wave height, depth and conditions of sea bed. However, most of all, the speed of current and wave height influence more upon the pound net than any other factors to deform and flutter. In this study, author carried out the experiments with a model of double one-side pound net made by the similarity law as 1:100 scales at a real experimental area, and additionally the model net experiments were conducted in the circulating water channel in Pukyong National University. The author analyzed the data of transformation of shape and tension of the model pound net to recognize the characteristics of the current and wave acting on it. Regardless of the direction of flow affecting on the fish court net or bag net, the deformed angle and depth to the side panel and bottom of box nets becomes bigger as the wave gets higher and the period of wave is faster. The tension in both upward or downward tends to be changed by the speed of wave. Those value of changes occurred similarly in either fish court net or bag net. Generally, when bag net is located at upward of flow, the value of tension was bigger 10% than any other location or nets. Regardless of the setting direction, the tension of the pound net is increased in proportion to flow speed, wave height and period of wave, and it becomes bigger about 15-30% at upward to flow than downward. Where the flow is upward in the court net, the tension in the wave increased to 37% compared to the one in the flow only in the condition of flow of 0.1-0.3m/s. Where the flow is upward in the bag net, the tension in the wave increased to 52% in the flow of 0.1m/s, and the tension increased to 48% in the flow of 0.2-0.3m/s.

• Journal of Internet Computing and Services
• /
• v.18 no.5
• /
• pp.61-67
• /
• 2017

In this paper we introduce two experimental results from classification of Instagram images and some valuable lessons from them. We have tried some experiments for evaluating the competitive power of Convolutional Neural Network(CNN) in classification of real social network images such as Instagram images. We used AlexNet and ResNet, which showed the most outstanding capabilities in ImageNet Large Scale Visual Recognition Challenge(ILSVRC) 2012 and 2015, respectively. And we used 240 Instagram images and 12 pre-defined categories for classifying social network images. Also, we performed fine-tuning using Inception V3 model, and compared those results. In the results of four cases of AlexNet, ResNet, Inception V3 and fine-tuned Inception V3, the Top-1 error rates were 49.58%, 40.42%, 30.42%, and 5.00%. And the Top-5 error rates were 35.42%, 25.00%, 20.83%, and 0.00% respectively.

• Journal of Broadcast Engineering
• /
• v.28 no.3
• /
• pp.267-274
• /
• 2023

There is a problem that the size of the dataset is insufficient due to the limitation of the collection of the medical image public data, so there is a possibility that the existing studies are overfitted to the public dataset. In this paper, we compare the performance of eight (Unet, X-Net, HarDNet, SegNet, PSPNet, SwinUnet, 3D-ResU-Net, UNETR) medical image semantic segmentation models to revalidate the superiority of existing models. Anatomical Tracings of Lesions After Stroke (ATLAS) V1.2, a public dataset for stroke diagnosis, is used to compare the performance of the models and the performance of the models in ATLAS V2.0. Experimental results show that most models have similar performance in V1.2 and V2.0, but X-net and 3D-ResU-Net have higher performance in V1.2 datasets. These results can be interpreted that the models may be overfitted to V1.2.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.9 no.3
• /
• pp.76-81
• /
• 2006

The reinforcement of soil shear strength by nylon net as substitute materials simulating a fine root system was evaluated by soil strength parameters(apparent cohesion(c) and internal friction angle(tan${\phi}$), using simple shear tester which clearly depicts shear deformation and controls soil suction. And the results of shear test by using bamboo as a substitute materials simulating a main root system and using nylon net as a substitute materials simulating a fine root system were compared. The reinforcement of soil strength by nylon net are expressed by apparent cohesion more than internal friction angle. In addition the increment of apparent cohesion by nylon net reached a peak in suction 60 $cmH_2O$. Different from with bamboo, the possibility of the change on internal friction angle(tan${\phi}$) caused by the soil water condition was shown in shear strain 20% condition. These results show that the mechanism of reinforcement by substitute materials simulating root system may be different in the condition of various soil water content.

• Journal of Ocean Engineering and Technology
• /
• v.22 no.4
• /
• pp.51-58
• /
• 2008

The objective of this study was to investigate the mechanical characteristics of fiber-reinforced lightweight soil using waste fishing net or monofilament for recycling both dredged soils and bottom ash. Reinforced lightweight soil consists of dredged soil, cement, air foam, and bottom ash. Waste fishing net or monoiament was added the mixture in order to increase the shear strength of the lightweight soil. Test specimens were fabricated with various mixing conditions, including waste fishing net content and monofilament content. Several series of unconfined compression tests and direct shear tests were carried out. From the experimental results, it was found that the unconfined compressive strength, as well as the stress-strain behavior of reinforced lightweight soil was strongly influenced by mixing conditions. In this study, the maximum increase in shear strength was obtained with either a 0.5% content of monofilament or 0.25% waste fishing net. The unconfined compressive strength of reinforced lightweight soil with monofilament was greater than that of reinforced lightweight soil with waste fishing net.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.41 no.3
• /
• pp.227-233
• /
• 2005

In order to estimate the effective horse power(EHP) in towing net of a bottom trawl ship, the ship's resistance was calculated by using a series data of Yamagata and Wigley formula. Also the effective horse power for a ship(EHPs) was estimated versus the ship speed in sailing and the propulsive efficiency was calculated with the brake horse power and the effective horse power. Then the effective horse power for a ship and a trawl net were estimated in the application of the propulsive efficiency in towing net. The total effective horse power($EHP_T$) was average 187.6kW and the effective horse power for a 1.awl net($EHP_n$) was average 176.7kW at a smooth sea state in towing net. The ratio of $EHP_n$ to $EHP_T$ was about 94.0% and the value was higher slightly than was already informed at a smooth sea state. The power for keeping up a townet speed was required more about 20% of a maximum continuous power at a rather rough sea state than a smooth sea state. In the future, if the residual resistance is considered with a sea state, $EHP_n$ will be estimated more correctly Also the data of EHP estimated by this method will be used as the basic data to design a trawl net.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.2 no.4
• /
• pp.137-147
• /
• 1998

The neural net application was tried to develop the technique for monitoring the health status of a steel truss bridge which was scaled down to 1/15 of the real bridge for the laboratory experiments. The damage scenarios were chosen as 7 cases. The dynamic behavior, which was changed due to the breakage of the members, of the bridge was investigated by finite element analysis. The bridge consists of single spam, and eight (8) main structural subsystems. The loading vehicle, which weighs as 100 kgf, was operated by the servo-motor controller. The accelerometers were bonded on the surface of 7 cross-beams to measure the dynamic behavior induced by the abnormal structural condition. Artificial neural network technique was used to determine the severity of the damage. At first, the neural net was learnt by the results of finite element analysis, and also, the maximum detection error was 3.65 percents. Another neural net was also learnt, and verified by the experimental results, and in this case, the maximum detection error was 1.05 percents. In future study, neural net is necessary to be learnt and verified by various data from the real bridge.

• Journal of IKEEE
• /
• v.27 no.4
• /
• pp.456-463
• /
• 2023

In this paper, we introduce the SSResUNet network model, which integrates the SPADE structure with the U-Net network model for accurate land cover classification using high-resolution satellite imagery without requiring user intervention. The proposed network possesses the advantage of preserving the spatial characteristics inherent in satellite imagery, rendering it a robust classification model even in intricate environments. Experimental results, obtained through training on KOMPSAT-3A satellite images, exhibit superior performance compared to conventional U-Net and U-Net++ models, showcasing an average Intersection over Union (IoU) of 76.10 and a Dice coefficient of 86.22.

• Journal of Biomedical Engineering Research
• /
• v.45 no.4
• /
• pp.179-186
• /
• 2024

This paper investigates the impact of Riemannian Procrustes Analysis (RPA) on enhancing the classification performance of SPD-Net when applied to EEG signals across different sessions and subjects. EEG signals, known for their inherent individual variability, are initially transformed into Symmetric Positive Definite (SPD) matrices, which are naturally represented on a Riemannian manifold. To mitigate the variability between sessions and subjects, we employ RPA, a method that geometrically aligns the statistical distributions of these matrices on the manifold. This alignment is designed to reduce individual differences and improve the accuracy of EEG signal classification. SPD-Net, a deep learning architecture that maintains the Riemannian structure of the data, is then used for classification. We compare its performance with the Minimum Distance to Mean (MDM) classifier, a conventional method rooted in Riemannian geometry. The experimental results demonstrate that incorporating RPA as a preprocessing step enhances the classification accuracy of SPD-Net, validating that the alignment of statistical distributions on the Riemannian manifold is an effective strategy for improving EEG-based BCI systems. These findings suggest that RPA can play a role in addressing individual variability, thereby increasing the robustness and generalization capability of EEG signal classification in practical BCI applications.