• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.7
• /
• pp.1706-1725
• /
• 2024

The unique U-shaped structure of U-Net network makes it achieve good performance in image segmentation. This network is a lightweight network with a small number of parameters for small image segmentation datasets. However, when the medical image to be segmented contains a lot of detailed information, the segmentation results cannot fully meet the actual requirements. In order to achieve higher accuracy of medical image segmentation, a novel improved U-Net network architecture called multi-scale encoder-decoder U-Net+ (MEDU-Net+) is proposed in this paper. We design the GoogLeNet for achieving more information at the encoder of the proposed MEDU-Net+, and present the multi-scale feature extraction for fusing semantic information of different scales in the encoder and decoder. Meanwhile, we also introduce the layer-by-layer skip connection to connect the information of each layer, so that there is no need to encode the last layer and return the information. The proposed MEDU-Net+ divides the unknown depth network into each part of deconvolution layer to replace the direct connection of the encoder and decoder in U-Net. In addition, a new combined loss function is proposed to extract more edge information by combining the advantages of the generalized dice and the focal loss functions. Finally, we validate our proposed MEDU-Net+ MEDU-Net+ and other classic medical image segmentation networks on three medical image datasets. The experimental results show that our proposed MEDU-Net+ has prominent superior performance compared with other medical image segmentation networks.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.11 no.2
• /
• pp.104-113
• /
• 2008

Due to the increasing needs of land supply, green space has been continuously encroached and resulted considerable naturality deterioration in Korea. In order to overcome the problem, it is necessary to introduce a integrated system for managing the quantitative and qualitative aspects of green space. With this regard, this study is to derive the feasibility for the adoption of the 'no net loss of green' policy to Korea from German and Japanese experiences. In Germany, natural resources are protected and inevitable damages were compensated through the adoption of 'interference regulation'. In Japanese case, future green space and artificial green foundation should be specified on the green basic plan which pursuit the improvement of naturality. In order to introduce the 'no net loss of green' policy to Korea, not only awareness raising on integrated green management system but also amendment of related regulations are needed. In detail, restoration and recovery on the damaged nature should be clearly elucidated and indicators to evaluate the current naturality and calculation methods for the restoration should be developed. Should this integrated green management system introduced, deterioration of naturality by development activities could be minimized thorough the adoption of land development permission criteria and green space management methodologies.

• Journal of Wetlands Research
• /
• v.26 no.3
• /
• pp.227-235
• /
• 2024

Inland wetlands provide important ecosystem services or functions, including biodiversity conservation, water purification, flood control, and carbon sequestration. However, many inland wetlands are being degraded by urbanization and industrialization, raising the need for conservation and restoration. This study proposes the adoption of No Net Loss(NLL) of Wetland for wetland conservation and restoration in Korea by the review of the literatures. To this end,, we analyzed the wetland conservation and management systems of the United States, Canada, Germany, Japan, and Australia, and identified institutional features and success factors of each country, and sought ways to adapt them to domestic conditions of Korea. The case of the United States highlights the importance of legal frameworks and public-private collaboration, Canada demonstrates systematic management approaches, Germany maintains a balance between development and conservation, Japan underscores the need for cooperative conservation efforts, and Australia emphasizes the efficiency of offset markets. The introduction and operation of these schemes are expected to contribute to climate change mitigation, biodiversity conservation, and water quality improvement, but ongoing research and policy efforts are needed to ensure that the schemes are well established.

• Korea Journal of Hospital Management
• /
• v.8 no.3
• /
• pp.118-142
• /
• 2003

This study is aimed at evaluating the financial structure of hospitals before and after the separation of prescription and drug dispensing policy started to be implemented in July 2000 and at making a suitable hospital managerial strategy through the verification of the factors which have effect on their profitability. This study investigated the hospitals which have passed the accredition review to be designated as a accredited training hospital each year for three years from 1999 to 2001. Those hospitals were selected from members of the Korea Hospital Association. 106 hospitals were targeted for analysis except for the hospitals whose financial statements and managerial performance were not reported faithfully. The financial indicators used in this study were stability indicators(liability to total assets, ratio of debt to fund balance, fixed ratio), liquidity indicators(current ratio, quick ratio), activity indicators(total assets turnover, fixed assets turnover), profitability indicators(net profit to total assets, net profit to net worth, operating margin), and operating expenses to patient revenues indicators(drug and supplies costs/payroll/overhead expenses). The result of this study are as follows: First, the analysis of the increase of loss-making hospitals before and after. The separation of prescription and drug dispensing policy shows that the number of loss-making hospitals increase after the separation(22.6% before the separation; 31.1% after the separation). However, there was no significant statistical difference. Second, the analysis of operating expenses to patient revenues indicators showed that the ratio of drug and supplies cost became lower in all hospitals but the ratio of payroll/overhead expenses became higher. Additionally, the factor which have the greatest effect on profitability was operating expenses to patient revenues indicators (drug and supplies costs/payroll/overhead expenses). Third, the analysis of managerial performance by four types of loss-loss, loss-profit, profit-loss and profit-profit compared the results before the separation with those after the separation revealed as follows : Reliance on liability to total assets became higher in the profit-loss type($56.2%{\rightarrow}66.4%$), lower in the loss-profit type($82.7%{\rightarrow}74.5%$). Total assets turnover became higher in the profit-profit type($1.3{\rightarrow}1.5$), but lower in the loss-profit type($0.8{\rightarrow}0.7$). Operating margin decreased to minus 5.9% from 4.3% in the profit-loss type, but increased to 7.2% from minus 7.8% in the loss-profit type. Forth, operating expenses to revenues indicators showed that the increase of payroll was the biggest in the profit-loss type($39.2%{\rightarrow}49.9%$) and that overhead cost decreased in the loss-profit type but that rather increased in other types.

• Smart Structures and Systems
• /
• v.32 no.3
• /
• pp.135-151
• /
• 2023

Using 3D asphalt pavement surface data, a deep and multiscale network named CrackNet-M is proposed in this paper for pixel-level crack detection for improvements in both accuracy and robustness. The CrackNet-M consists of four function-specific architectural modules: a central branch net (CBN), a crack map enhancement (CME) module, three pooling feature pyramids (PFP), and an output layer. The CBN maintains crack boundaries using no pooling reductions throughout all convolutional layers. The CME applies a pooling layer to enhance potential thin cracks for better continuity, consuming no data loss and attenuation when working jointly with CBN. The PFP modules implement direct down-sampling and pyramidal up-sampling with multiscale contexts specifically for the detection of thick cracks and exclusion of non-crack patterns. Finally, the output layer is optimized with a skip layer supervision technique proposed to further improve the network performance. Compared with traditional supervisions, the skip layer supervision brings about not only significant performance gains with respect to both accuracy and robustness but a faster convergence rate. CrackNet-M was trained on a total of 2,500 pixel-wise annotated 3D pavement images and finely scaled with another 200 images with full considerations on accuracy and efficiency. CrackNet-M can potentially achieve crack detection in real-time with a processing speed of 40 ms/image. The experimental results on 500 testing images demonstrate that CrackNet-M can effectively detect both thick and thin cracks from various pavement surfaces with a high level of Precision (94.28%), Recall (93.89%), and F-measure (94.04%). In addition, the proposed CrackNet-M compares favorably to other well-developed networks with respect to the detection of thin cracks as well as the removal of shoulder drop-offs.

• The Korean Journal of Malacology
• /
• v.24 no.2
• /
• pp.131-136
• /
• 2008

Growth and survival of the spat of arkshell, Scapharca broughtonii were investigated for the improvement of survival in intermediate culture with different shape of protective net and type of preventive net of spat loss. The growth performance of the spat of arkshell was observed for 60 days by using different forms of protective net, as the exposure form with an average shell length of $12.8{\pm}3.2$ mm(P<0.05), the fish pot form with $12.2{\pm}3.5$ mm, the cylinder form with $11.9{\pm}3.8$ mm and the last one is the rectangular form with $10.9{\pm}3.7$ mm. Their numerical value did not show any marked difference with each other. But in case of survival rate, the fish pot form showed highest survival rate which is 43.7% and the significantly lowest was showed by the exposure form, which is 5.4%(P<0.05). The cylinder form and rectangular form showed 41.2 and 31.6% respectively. And then the growth of the spat of arkshell was observed for 163 days by rearing in a sort of preventive net, the first group was in a blackout curtain with average shell length of $13.9{\pm}3.1$ mm, the next group was in a balsam pear net with $12.9{\pm}3.0$ mm, in a polyethylene net with $11.8{\pm}3.1$ mm and the control with $12.6{\pm}3.3$ mm which was not installed by preventive net of spat loss. The survival rate was 91.5% in a blackout curtain, 90.1% in a polyethylene net, 88.5% in a balsam pear net and 61.5% in a controlled group. It is seen that the highest growth and survival rate were observed in a fish pot form and cylinder form. These results were a little bit difference from those of the spat of arkshell cultured in the form of different preventive nets. There was no significant difference in survival rate in relation to the sorts of preventive net against carrying-away, but there was a difference in survival rate by more than 30% as against the non-installed controlled group. We expect that survival rate would be highly improved in intermediate culture carried out with fish pot form and cylinder form of protective net and preventive net of spat loss.

• Horticultural Science & Technology
• /
• v.29 no.4
• /
• pp.293-297
• /
• 2011

This experiment was conducted to elucidate the effects of mulching materials and removing time of the transparent polyethylene (PE) film on the growth of garlic at Uiseong experimental field, Korea. The experimental mulching materials comprised of transparent polyethylene film (0.025 mm) and net polyethylene (NPE). Plant height and leaf number of garlic were highest at PE treatment when the PE removing date was March 18 and this treatment also promoted the no. of cloves. Length of leaf sheath and bolting rate were highest and bulb weight loss rate was lowest at PE + NPE treatment when the PE removing date was March 18. But clove number was the lowest in this treatment compared to conventional PE film treatment. Conventional mulching method accelerated secondary growth rate but bulb weight loss was vice versa. There were statistically no differences in bulb diameter among treatments but conventional treatment positively focused on bulb diameter. Whenever PE film remove can suppress weeds compared to no mulching treatment but the dry weight of weeds were increasing trends as the removal dates of PE film were delayed. Transparent PE or PE + NPE treatments can be recommended to grow best quality garlic when PE film removing date is March 18.

• The Journal of Industrial Distribution & Business
• /
• v.11 no.1
• /
• pp.19-28
• /
• 2020

Purpose: This research aims to empirically investigate the motivation of corporate voluntary disclosure by exploring the impact of historical loss on corporate social responsibility disclosure (CSRD). Research design, data, and methodology: This paper takes Chinese A-share listed firms that issued standalone social responsibility reports during the period of 2009-2017 as a sample. Drawing on extant literature, this paper defines historical loss firms as firms with net profit greater than or equal to 0 and undistributed profit less than 0. The tendency score matching method (PSM) is used to find matching samples for historical loss firms. Then OLS regression is conducted to investigate the relationship between historical loss and corporate social disclosure. Results: The results show that historical loss has a significant positive impact on the quality of corporate social responsibility disclosure. After changing the measurement of independent and dependent variables as well as adopting different matching methods to screen the control group, the results still hold. Further research indicates that the relationship between historical loss and CSRD is influenced by corporate financing constraints and industry competition. Conclusions: This research supports the resource motivation hypothesis of corporate social responsibility disclosure, and provides empirical evidence for regulators to strengthen supervision on corporate disclosure.

• Ecology and Resilient Infrastructure
• /
• v.2 no.4
• /
• pp.311-316
• /
• 2015

This study defined and classified no-net-loss-of-greenness (NNLG) based on the law, and then assessed the NNLG index by metropolitan cities and provinces in Korea after estimating NNLG evaluation indicators for the introduction of NNLG for health improvement of urban ecosystems. The results are as follows. First, NNLG was the comprehensive meaning that was included in the greenbelt and park greenbelt and the green area which was defined by the Act on Urban Parks, Greenbelts, etc. and the National Land Planning and Utilization Act respectively. Second, NNLG was classified as a park greenbelt which was included urban parks and greenbelts such as buffer greenbelts, scenic greenbelts, and connecting greenbelts, green areas which was included in green conservation areas, green production areas, green natural areas, and green coverage which is included forests, grasslands, and wetlands that were occupied by vegetation such as trees, shrubs, and plants. Third, NNLG index by cities and provinces was assessed based on the estimation of NNLG evaluation indicators, which included parks and greenbelt areas per capita, green areas per capita, green coverage per capita, ratio of parks and greenbelts, ratio of green areas, and ratio of green coverage. As a result, Sejong city got the highest point of NNLG index and Seoul and Daegu got lowest points of NNLG index among metropolitan cities in Korea. Chungbuk got the highest point of NNLG index and Kyonggi and Jeju got lowest points of NNLG index among provinces in Korea.

• Journal of Broadcast Engineering
• /
• v.26 no.6
• /
• pp.778-789
• /
• 2021

Recently, various lightweight methods for using Convolutional Neural Network(CNN) models in mobile devices have emerged. Weight quantization, which lowers bit precision of weights, is a lightweight method that enables a model to be used through integer calculation in a mobile environment where GPU acceleration is unable. Weight quantization has already been used in various models as a lightweight method to reduce computational complexity and model size with a small loss of accuracy. Considering the size of memory and computing speed as well as the storage size of the device and the limited network environment, this paper proposes a method of compressing integer weights after quantization using a video codec as a method. To verify the performance of the proposed method, experiments were conducted on VGG16, Resnet50, and Resnet18 models trained with ImageNet and Places365 datasets. As a result, loss of accuracy less than 2% and high compression efficiency were achieved in various models. In addition, as a result of comparison with similar compression methods, it was verified that the compression efficiency was more than doubled.