• Journal of Bio-Environment Control
• /
• v.30 no.3
• /
• pp.175-182
• /
• 2021

This experiment was conducted to find out the possibility of use of reused rockwool and comparison of growth, productivity and quality of tomatoes according to the use of rockwool and coir medium. The experiment was conducted in an automatic controlled greenhouse at Andong National University, College of Life Science, located in Andong, Gyeongsangbuk-do.. As a result of the experiment, there was no difference in the number of leaves, plant height, and leaf area between treatments, and the crown diameter was slightly higher in rockwool medium, also there was no difference between reused rockwool and coir medium. Fruit productivity showed different responses depending on the cultivation environment, but there was no significant difference between rockwool, reused rockwool and coir medium. In addition, the quality of fruit was observed to be different according to the concentration of EC in the medium. Therefore, in tomato hydroponic cultivation, there was no difference in the type of medium in growth, productivity, fruit quality and the environmental and water management had a great effect, and it is expected that the reuse of rockwool will have a positive effect on the economic point of view.

• Journal of the Korean Wood Science and Technology
• /
• v.47 no.2
• /
• pp.229-238
• /
• 2019

Various wood defects occur during tree growing or wood processing. Thus, to use wood practically, it is necessary to objectively assess their quality based on the usage requirement by accurately classifying their defects. However, manual visual grading and species classification may result in differences due to subjective decisions; therefore, computer-vision-based image analysis is required for the objective evaluation of wood quality and the speeding up of wood production. In this study, the SIFT+k-NN and CNN models were used to implement a model that automatically classifies knots and analyze its accuracy. Toward this end, a total of 1,172 knot images in various shapes from five domestic conifers were used for learning and validation. For the SIFT+k-NN model, SIFT technology was used to extract properties from the knot images and k-NN was used for the classification, resulting in the classification with an accuracy of up to 60.53% when k-index was 17. The CNN model comprised 8 convolution layers and 3 hidden layers, and its maximum accuracy was 88.09% after 1205 epoch, which was higher than that of the SIFT+k-NN model. Moreover, if there is a large difference in the number of images by knot types, the SIFT+k-NN tended to show a learning biased toward the knot type with a higher number of images, whereas the CNN model did not show a drastic bias regardless of the difference in the number of images. Therefore, the CNN model showed better performance in knot classification. It is determined that the wood knot classification by the CNN model will show a sufficient accuracy in its practical applicability.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.11
• /
• pp.319-325
• /
• 2018

The use of 3D printing (Additive Manufacturing) technology has expanded from initial model production to the mass production of parts in the industrial field based on the continuous research and development of materials and process technology. As a representative polymer material for 3D printing, the polyamide-based material, which is one of the high-strength engineering plastics, is used mainly for manufacturing parts for automobiles because of its light weight and durability. In this study, the specimens were fabricated using Selective Laser Sintering, which has excellent mechanical properties, and the flexural characteristics were analyzed according to the temperature of the two types of polyamide 12 and glass bead reinforced PA12 materials. The test specimens were prepared in the directions of $0^{\circ}$, $45^{\circ}$, and $90^{\circ}$ based on the work platform, and then subjected to a flexural test in three test temperature environments of $-25^{\circ}C$, $25^{\circ}C$, and $60^{\circ}C$. As a result, PA12 had the maximum flexural strength in the direction of $90^{\circ}$ at $-25^{\circ}C$ and $0^{\circ}$ at $25^{\circ}C$ and $60^{\circ}C$. The glass bead-reinforced PA12 exhibited maximum flexural strength values at all test temperatures in the $0^{\circ}$ fabrication direction. The tendency of the flexural strength changes of the two materials was different due to the influence of the plane direction of the lamination layer depending on the type of stress generated in the bending test.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.6
• /
• pp.251-256
• /
• 2019

A capsule endoscopy is a medical device that can capture an entire digestive organ from the esophagus to the anus at one time. It produces a vast amount of images consisted of about 8~12 hours in length and more than 50,000 frames on a single examination. However, since the analysis of endoscopic images is performed manually by a medical imaging specialist, the automation requirements of the analysis are increasing to assist diagnosis of the disease in the image. Among them, this study focused on automatic detection of polyp images. A polyp is a protruding lesion that can be found in the gastrointestinal tract. In this paper, we propose a weighted-image generation method to enhance the polyp image learning by multi-scale analysis. It is a way to extract the suspicious region of the polyp through the multi-scale analysis and combine it with the original image to generate a weighted image, that can enhance the polyp image learning. We experimented with SVM and RF which is one of the machine learning methods for 452 pieces of collected data. The F1-score of detecting the polyp with only original images was 89.3%, but when combined with the weighted images generated by the proposed method, the F1-score was improved to about 93.1%.

• 재활복지
• /
• v.21 no.3
• /
• pp.129-146
• /
• 2017

In an event of an emergency, such as fire in a building, visually impaired and blind people are prone to exposed to a level of danger that is greater than that of normal people, for they cannot be aware of it quickly. Current fire detection methods such as smoke detector is very slow and unreliable because it usually uses chemical sensor based technology to detect fire particles. But by using vision sensor instead, fire can be proven to be detected much faster as we show in our experiments. Previous studies have applied various image processing and machine learning techniques to detect fire, but they usually don't work very well because these techniques require hand-crafted features that do not generalize well to various scenarios. But with the help of recent advancement in the field of deep learning, this research can be conducted to help solve this problem by using deep learning-based object detector that can detect fire using images from security camera. Deep learning based approach can learn features automatically so they can usually generalize well to various scenes. In order to ensure maximum capacity, we applied the latest technologies in the field of computer vision such as YOLO detector in order to solve this task. Considering the trade-off between recall vs. complexity, we introduced two convolutional neural networks with slightly different model's complexity to detect fire at different recall rate. Both models can detect fire at 99% average precision, but one model has 76% recall at 30 FPS while another has 61% recall at 50 FPS. We also compare our model memory consumption with each other and show our models robustness by testing on various real-world scenarios.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.17 no.6
• /
• pp.1-13
• /
• 2018

The number of deaths in Korea is higher than the OECD average. Therefore, an e-Call system is being developed as a vehicle ICT-based emergency rescue system that automatically detects an accident in the event of a vehicle accident and transmits related information to the center. In order to overcome the limitations of social acceptability and function of e-Call system, we propose a model that allows users to be aware of the necessity of service voluntarily. We predicted the market share of e-call services according to the proposed business model and analyzed it through B/C analysis. Benefits are calculated on a penetration basis, and device purchase and communications costs are calculated for each period. B/C analysis shows that pessimistic scenarios are 0.98 in 2025 and 1.01 in 2030. In an optimistic scenario, it is 1.05 in 2025 and 1.20 in 2030, which is more economical.

• Korean Chemical Engineering Research
• /
• v.57 no.2
• /
• pp.274-282
• /
• 2019

In recent years, big data analysis has been expanded to include automatic control through reinforcement learning as well as prediction through modeling. Research on the utilization of image data is actively carried out in various industrial fields such as chemical, manufacturing, agriculture, and bio-industry. In this paper, we applied NASNet, which is an AutoML reinforced learning algorithm, to DeepU-Net neural network that modified U-Net to improve image semantic segmentation performance. We used BRATS2015 MRI data for performance verification. Simulation results show that DeepU-Net has more performance than the U-Net neural network. In order to improve the image segmentation performance, remove dropouts that are typically applied to neural networks, when the number of kernels and filters obtained through reinforcement learning in DeepU-Net was selected as a hyperparameter of neural network. The results show that the training accuracy is 0.5% and the verification accuracy is 0.3% better than DeepU-Net. The results of this study can be applied to various fields such as MRI brain imaging diagnosis, thermal imaging camera abnormality diagnosis, Nondestructive inspection diagnosis, chemical leakage monitoring, and monitoring forest fire through CCTV.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.9
• /
• pp.517-524
• /
• 2019

The overdrive hub clutch is attached to a 6-speed automatic transmission to reduce fuel consumption by using the additional power of the engine. This paper proposes a means to minimize the load and roll-over ratio on the punch during the piercing process for the overdrive hub clutch product. Die clearance, shear angle, and friction coefficient, which can affect the load and roll-over ratio of the punch during processing, were set as the design variables. Sensitivity analysis was also conducted to determine the influence of each design variable on the punch load and roll-over ratio. As a result, shear angle, friction coefficient and die clearance were found to be sensitive to load and roll-over ratio. The punch load and roll-over ratio were set as the objective function and the equation of each design variable and objective function was derives using the Response Surface Method. Finally, the optimal value of the design variables was derived using the Response Surface Method. Application of this model to finite element analysis resulted in 22.14% improvement in the roll-over ratio of the punch load and material.

• Science of Emotion and Sensibility
• /
• v.24 no.2
• /
• pp.13-24
• /
• 2021

In level 5 (mind-off) of autonomous driving, the autonomous vehicle passengers are expected to have various activities such as face-to-face meetings, working, relaxing, and watching movies. In particular, various changes in the interior space of the vehicle are expected. Moreover, according to the survey conducted by the American Automobile Association, 73% of the respondents reported that they were afraid to board autonomous vehicles. In level 5 of autonomous driving, the subject of safety was expected to be transferred to autonomous vehicles; thus, research should be conducted from the user's perspective. Recently, various studies have been conducted to secure the safety of fully autonomous vehicles. However, there are limited studies addressing the psychological safety of actual passengers. Therefore, this study conducted a questionnaire based on the AHP technique. Consequently, the automobile safety system's priority for securing passengers' psychological safety according to each type of indoor behavior was derived, and the interior space for securing the psychological stability of passengers was suggested based on the obtained results. This study offers a new direction for interior space design, satisfying the psychological safety of passengers. This study is important because it advocates that the interior environment of fully autonomous driving cars is expected to be designed to secure the user's psychological safety.

• Korean Chemical Engineering Research
• /
• v.59 no.2
• /
• pp.209-218
• /
• 2021

First principle-based modeling studies have been performed to improve the heat exchange efficiency of ORV and optimize operation, but the heat transfer coefficient of ORV is an irregular system according to time and location, and it undergoes a complex modeling process. In this study, FNN, LSTM, and AutoML-based modeling were performed to confirm the effectiveness of data-based modeling for complex systems. The prediction accuracy indicated high performance in the order of LSTM > AutoML > FNN in MSE. The performance of AutoML, an automatic design method for machine learning models, was superior to developed FNN, and the total time required for model development was 1/15 compared to LSTM, showing the possibility of using AutoML. The prediction of NG and seawater discharged temperatures using LSTM and AutoML showed an error of less than 0.5K. Using the predictive model, real-time optimization of the amount of LNG vaporized that can be processed using ORV in winter is performed, confirming that up to 23.5% of LNG can be additionally processed, and an ORV optimal operation guideline based on the developed dynamic prediction model was presented.