• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.54 no.4
• /
• pp.279-286
• /
• 2018

The objective of this study was to quantitatively evaluate the influence of the netting hanging ratio on fishing performance of gillnet. A total of six tests was carried out using three kinds of gillnets with different horizontal hanging ratios (0.3, 0.5 and 0.7) on the basis commercial gillnet of mesh size of 60 mm used in Yeosu Dolsan. As a result of the test fishing operation, catches of the net with hanging ratio 0.3 and the shortest float length was 1.34 times larger than the net with hanging ratio 0.7. The caught species in experimental testes was lager in the net with hanging ratio 0.3 than the net with hanging ratio 0.7. The catch rate per unit area (CPUE) is on the silver croaker with the largest catches in tests as follows. The net with hanging ratio 0.3 net was $89.45g/m^2$, hanging ratio 0.5 net was $41.63g/m^2$, and hanging ratio 0.7 net was $24.50g/m^2$. The net with hanging ratio 0.3 (float line length of 45 m) exhibited the largest CPUE value.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.5
• /
• pp.161-168
• /
• 2021

Cracks affect the robustness of infrastructures such as buildings, bridge, pavement, and pipelines. This paper presents an automated crack detection system which detect cracks in diverse surfaces. We first constructed the combined crack dataset, consists of multiple crack datasets in diverse domains presented in prior studies. Then, state-of-the-art deep learning models in computer vision tasks including VGG, ResNet, WideResNet, ResNeXt, DenseNet, and EfficientNet, were used to validate the performance of crack detection. We divided the combined dataset into train (80%) and test set (20%) to evaluate the employed models. DenseNet121 showed the highest accuracy at 96.20% with relatively low number of parameters compared to other models. Based on the validation procedures of the advanced deep learning models in crack detection task, we shed light on the cost-effective automated crack detection system which can be applied to different surfaces and structures with low computing resources.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.42 no.3
• /
• pp.127-133
• /
• 2006

An estimation of the headline height of a bottom trammel net set across under uniform current was achieved numerically from a differential equations describing the forces of the net and compared with the measured value in a flume tank experiment. The analysis on the shape of the bottom trammel net with the headline free was based on the equilibrium equation of the bottom gill net which was modified and slack of the trammel net was varied with net depth as shown in the tank experiment. The differential equations were solved by a forth-order Runge-Kutta method. The estimated headline heights with varied slack was found to be closer than that with constant slack when compared with the actual values.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.21 no.2
• /
• pp.89-98
• /
• 1985

As a primary step in studying the effects of the netting porosity on the net-shape in the four-seam trawling net, a series of experiments were performed changing the porosity of each panel, with the simplified mo:iel of the bag net made of porous vinyl film and that ma:ie of net webbing. These models were suspended horizontally in circularly flowing water, with two pairs of susp~nding threais to four points of symetry at the border of the bag mouth in place of both wing nets. And then, the section shape of the bag mouth photographed and the tensions on both pairs of suspending threads were measured with two load cells in circularly flowing water. From the results, the auther estimated an experimental equation from the relationship between the porosity of each panel in the bag net and the section shape of the bag mouth, h/w=k (l-Pr_u/lPr_s)

• Journal of the Korea Safety Management & Science
• /
• v.5 no.2
• /
• pp.129-143
• /
• 2003

Bsusiness-to- Business transactions requires interactions between diverse application s and business. This gives rise to the need to for a common standard and protocol for exchanging busienss messages. and RosettaNet is the leading industry organization focused on developing B2B process standard for a global supply chain. This paper is intended as an investigation of Build-up Method of SCM Connection applying EC Standard(Rossetta-net) to reduce the integration effort, and improve extensibility and interoperability of an eBusiness infrastructure.-Intiative way for E-commerce applications to quickly incorporate new standards. Supply Chain Management (SCM) helps companies plan and execute supply to meet customer demand. The researches of modem scientists have thrown new light on the subject. but, The study of SCM for EC Standard has been strangely neglected by critics. Consequently, existing SCM system, which require to introduce the functions of EC, needs to be changed. Let me summarize the main points that have been made in this paper. The sharing information for connection and the Rosetta-net standard of the existing SCM system, and PIP 3A4, order management module, of the Rosetta-net specifications for connection by the Rosetta-net. Lastly, That shows the process of the gateway for connecting among the existing SCM systems. This provides felxibility in incorporating any new emerging open standard.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2016.04a
• /
• pp.633-635
• /
• 2016

Convolutional Networks (ConvNets) are powerful models that learn hierarchies of visual features, which could also be used to obtain image representations for transfer learning. The basic pipeline for transfer learning is to first train a ConvNet on a large dataset (source task) and then use feed-forward units activation of the trained ConvNet as image representation for smaller datasets (target task). Our key contribution is to demonstrate superior performance of multiple ConvNet layer features over single ConvNet layer features. Combining multiple ConvNet layer features will result in more complex feature space with some features being repetitive. This requires some form of feature selection. We use AdaBoost with single stumps to implicitly select only distinct features that are useful towards classification from concatenated ConvNet features. Experimental results show that using multiple ConvNet layer activation features instead of single ConvNet layer features consistently will produce superior performance. Improvements becomes significant as we increase the distance between source task and the target task.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.39 no.3
• /
• pp.141-148
• /
• 2021

This paper presents a method to detect zebra-crossing using deep learning which combines SegNet and ResNet. For the blind, a safe crossing system is important to know exactly where the zebra-crossings are. Zebra-crossing detection by deep learning can be a good solution to this problem and robotic vision-based assistive technologies sprung up over the past few years, which focused on specific scene objects using monocular detectors. These traditional methods have achieved significant results with relatively long processing times, and enhanced the zebra-crossing perception to a large extent. However, running all detectors jointly incurs a long latency and becomes computationally prohibitive on wearable embedded systems. In this paper, we propose a model for fast and stable segmentation of zebra-crossing from captured images. The model is improved based on a combination of SegNet and ResNet and consists of three steps. First, the input image is subsampled to extract image features and the convolutional neural network of ResNet is modified to make it the new encoder. Second, through the SegNet original up-sampling network, the abstract features are restored to the original image size. Finally, the method classifies all pixels and calculates the accuracy of each pixel. The experimental results prove the efficiency of the modified semantic segmentation algorithm with a relatively high computing speed.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.15 no.5
• /
• pp.227-234
• /
• 2020

In this paper, we propose wav-U-Net to improve speech enhancement in heavy noisy environments, and it has implemented three principal techniques. First, as input data, we use 128 modified Mel-scale filter banks which can reduce computational burden instead of 512 frequency bins. Mel-scale aims to mimic the non-linear human ear perception of sound by being more discriminative at lower frequencies and less discriminative at higher frequencies. Therefore, Mel-scale is the suitable feature considering both performance and computing power because our proposed network focuses on speech signals. Second, we add a simple ResNet as pre-processing that helps our proposed network make estimated speech signals clear and suppress high-frequency noises. Finally, the proposed U-Net model shows significant performance regardless of the kinds of noise. Especially, despite using a single channel, we confirmed that it can well deal with non-stationary noises whose frequency properties are dynamically changed, and it is possible to estimate speech signals from noisy speech signals even in extremely noisy environments where noises are much lauder than speech (less than SNR 0dB). The performance on our proposed wav-U-Net was improved by about 200% on SDR and 460% on NSDR compared to the conventional Jansson's wav-U-Net. Also, it was confirmed that the processing time of out wav-U-Net with 128 modified Mel-scale filter banks was about 2.7 times faster than the common wav-U-Net with 512 frequency bins as input values.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.48 no.4
• /
• pp.322-336
• /
• 2012

To study the catching efficiency of biodegradable trammel net for swimming crab (Portunus trituberculatus), three types (biodegradable, monofilament and multifilament) of trammel nets were used in the field test, and the tests were carried out 16 times with two different mesh sizes (105mm and 160mm) in the Yeonpyeong fishing ground of Korea, 2009~2011. The catching efficiency of three type nets was analyzed by catch in number, catch in weight and average weight per individual of small and large size swimming crab by net types and mesh sizes. Statistical T-test was also carried out to verify the efficiency between the three types of nets. The results are as follows. The catch in number of swimming crab was 24,667 and formed about 81.0% of total catch. Of all swimming crab catch, small swimming crabs with less than 64mm in carapace length which is a prohibited landing size by law formed 48.1%, larger swimming crabs with more than 64mm in carapace length which is a landing size formed 51.9%. In 105mm mesh size trammel net test, the catch share in number of small size swimming crab by biodegradable trammel net was 47.5% in comparison with multifilament trammel net and 74.2% in comparison with monofilament trammel net, so biodegradable trammel net has more protective effects on small size swimming crab than other types of trammel nets. The protective effects for small size swimming crab by biodegradable trammel net was 25.8~52.5% in comparison with other types of trammel nets. The catch share in weight of large size swimming crab by biodegradable trammel net was 98.3% in comparison with multifilament trammel net and 92.3% in comparison with monofilament trammel net, so biodegradable trammel net has a similar catch efficiency to multifilament trammel net. The results of 160mm mesh size trammel net test have shown similar results of 105mm mesh size trammel net test. This study shows that biodegradable trammel net is a more useful fishing gear than multifilament and monofilament trammel net because biodegradable trammel net has lower catch rate than other types of nets in small size swimming crab and similar catch rate than multifilament trammel net which is a well used net by fishermen.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.31 no.1
• /
• pp.1-7
• /
• 1995

The fish behavior survey in the main-net of a large scale-net was carried out in relation to the catching function of the set-net. at Kishihata set-net in Japan during the period from January 29th to February 22th 1992. The results obtained are summarized as follows; 1. The greatest number of schools observation in the main-net occurred between 17 : 00hrs and 18 : 00hrs. after which the schools suddenly disappeared. This result show that the schools did not stay in the main-net. but immediately moved through into the bag-net or left the main-net. 2. The course of movement of the small schools of sardine in the main-net was in a straight line to the slope-net. However. upon arrived at the slope-net the school turned around. away from the slope-net and went directly towards the playground. 3. The majority of small schools of sardine in the main-net were estimated to move with the mode velocity 25cm/sec. however the maximum velocity were recorded 80cm/sec.