Blockchain enabled Decentralized Autonomous Organization (DAO), a new form of organization with conveying its core value - trust. Token holders who are participating DAO's governance share their thoughts, information, and ideas in online forum. But it is problem that chronological form of DAO's online forum makes token holders hard to find crucial information, meaning that many of them might not understand what is happening discussion. In this paper, we studied not only a decision making process which feature is iteration, visualization, and applicable to DAO with 6 steps in total but also a decision making tool which is based on the process of this paper. The tool has features to help participants such as voting model, visualization features which gives guidance to them for their decision during the process. Our experiment showed that the process and tool is somewhat reasonable, and the information during the process is effective for participants. This work is expected to be applied to current DAOs to make a decision among the token holders.

In a typical initial setup of a PCB component inspection system, operators should manually input various information such as category, position, and inspection area for each component to be inspected, thus causing much inconvenience and longer setup time. Although there are many deep learning based object detectors, RetinaNet is regarded as one of best object detectors currently available. In this paper, a method using an extended RetinaNet is proposed that automatically detects its component category and position for each component mounted on PCBs from a high-resolution color input image. We extended the basic RetinaNet feature pyramid network by adding a feature pyramid layer having higher spatial resolution to the basic feature pyramid. It was demonstrated by experiments that the extended RetinaNet can detect successfully very small components that could be missed by the basic RetinaNet. Using the proposed method could enable automatic generation of inspection areas, thus considerably reducing the setup time of PCB component inspection systems.

How accurately reproducible energy is delivered to the wafer in the process of making thin films using PE-CVD (Plasma enhanced chemical vapor deposition) during the semiconductor process. This is the most important technique, and most of the reaction on the wafer surface is made by thermal energy. In this study, we studied the method of monitoring the change of thermal energy transferred to the wafer surface by monitoring the temperature change according to the change of the thin film formed on the showerhead facing the wafer. Through this research, we could confirm the monitoring of wafer thin-film which is changed due to abnormal operation and accumulation of equipment, and we can expect improvement of semiconductor quality and yield through process reproducibility and equipment status by real-time monitoring of problem of deposition process equipment performance.

Deep red EuSi₂O₂N₂ phosphors were synthesized under various sintering gas pressures (1 atm, 2 atm, and 3 atm). They were in good agreement with the standard EuSi₂O₂N₂ ICSD card # 41-6046 (a monoclinic crystal system with space group of P2₁/a). Their photoluminescence intensities were significantly increased with increasing the gas pressures. They showed a broad band emission peaking at 680 nm due to 4f⁶5d¹ - 4f⁷ of Eu²⁺ ion, which can be efficiently excited in the visible range up to 550 nm. The best one at 3 atm was applied for red LED based on blue chip, which showed the strong deep red emission.

In this paper, we studied the problem if the experiment number occurring in order to identify defect in scaffold. We need to change each of the 5 print factor to predict defect when printing disk type scaffold using FDM 3d printer. So then the number of scaffold print will be more than 100,000 times. This experiment number is difficult to perform in the field. In order to solve this problem, we have produced a prediction model based on machine learning multiple linear regression using print conditions and defect scaffold data for print conditions. The prediction model produced was verified through experiments. The verification confirmed that the error was less than 0.5 %. We have confirmed that satisfied within the target margin of error 5 %.

Most of the background subtraction algorithms show good performance in static scenes. In the case of dynamic scenes, they frequently cause false alarm to "temporal clutter", a repetitive motion within a certain area. In this paper, we propose a robust technique for the multiple interval pixel sampling (MIS) algorithm to handle highly dynamic scenes. An adaptive threshold scheme is used to suppress false alarms in low-confidence regions. We also utilize multiple background models in the foreground segmentation process to handle repetitive background movements. Experimental results revealed that our approach works well in handling various temporal clutters.

In improving laser cutting of optical plastic films for mass production of optoelectronics display units, it is important to understand particle contamination over optical film surface due to fume particle generation and dispersion. This numerical study investigates the effects of downward and upward air flow motions on fume particle dispersion around laser cut line. The simulations employ random particle sampling of up to one million fume particles by probabilistic distributions of particle size, ejection velocity and angle, and fume particle dispersion and surface landing are predicted using Basset-Boussinesq-Oseen model of low Reynolds number flows. The numerical results show that downward air flow scatters fume particles of a certain size range farther away from laser cut line and aggravate surface contamination. However, upward air flow pushes fume particles of this size range back toward laser cut line or sucks them up with rising air motion, thus significantly alleviating surface contamination.

We investigated the Si_1-xGe_x surface properties when dry cleaning the films using NF₃ / H₂O remote plasma. After the dry cleaning process, it was found that about 80-250 nm wide bumps were formed on the SiGe surface regardless of Ge concentration in the rage of x = 0.1 ~ 0.3. In addition, effects of the dry cleaning processing parameters such as pressure, substrate temperature, and H₂O flow rates were examined. It was found that the surface bump is significantly dependent on the flow rate of H₂O. Based on these observations, we would like to provide additional guidelines for implementing the dry cleaning process to SiGe materials.

To develop gas-distinguishing sensor system, it is highly required to integrate multiple sensors for effective detection of a single targeted gas or mixture of gases. In addition, it is important to collect the reliable data from individual sensors into one integrated measuring device. Collecting the data of toxic gases on the spot should be done without inhalation. We suggest simple wirelessly running system for data collection that guarantees both reliability of data sources and safety. Here, we made a multi-gas measuring instrument(device) combined with Bluetooth module which provides a safe and precise big data accumulation system.

Copper Oxide (CuO) films were deposited on the n-type silicon wafer by rf magnetron sputtering for heterojunction solar cells. And then the samples were treated as a function of the annealing temperature (300-600℃) in a vacuum. Their electrical, optical and structural properties of the fabricated heterojunction solar cells were then investigated and the power conversion efficiencies (PCE) of the fabricated p-type copper oxide/n-type Si heterojunction cells were measured using solar simulator. After being treated at temperature of 500℃, the solar cells with CuO film have PCE of 0.43%, Current density of 5.37mA/㎠, Fill Factor of 39.82%.

Recently, artificial intelligence related technologies including machine learning are being applied to various fields, and the demand is also increasing. In particular, with the development of AR, VR, and MR technologies related to image processing, the utilization of computer vision based on deep learning has increased. The algorithms for object recognition and detection based on deep learning required for image processing are diversified and advanced. Accordingly, problems that were difficult to solve with the existing methodology were solved more simply and easily by using deep learning. This paper introduces various deep learning-based object recognition and extraction algorithms used to detect and recognize various objects in an image and analyzes the technologies that attract attention.

In this study, SiNx and Al₂O₃ thin film was manufactured using PECVD deposition process and applied to crystalline silicon solar cells, resulting in 16.7% conversion efficiency. The structural improvement experiment of the rear electrode resulted in a 1.7% improvement in conversion efficiency compared to the reference cell by reducing the recombination rate of minority carriers and increasing the carrier lifetime by forming a passivation layer consisting of SiNx and Al₂O₃ thin films through the PECVD process.

The author demonstrated organic ferroelectric thin-film transistors with ferroelectric materials of P(VDF-TrFE) and an amorphous oxide semiconducting In-Ga-Zn-O channel on the silicon substrates. The organic ferroelectric layers were deposited on an oxide semiconductor layer by Langmuir-Blodgett method and then annealed at 128℃ for 30min. The carrier mobility and current on/off ratio of the memory transistors showed 9 ㎠V^-1s^-1 and 6 orders of magnitude, respectively. We can conclude from the obtained results that proposed memory transistors were quite suitable to realize flexible and werable electronic applications.

Silicon wafer etching in micro electro mechanical systems (MEMS) fabrication is challenging to form 3-D structures. Well known Si-wet etch of silicon employs potassium hydroxide (KOH), tetramethylammonium hydroxide (TMAH) and sodium hydroxide (NaOH). However, the existing silicon wet etching process has a fatal disadvantage that etching of the back side of the wafer is hard to avoid. In this study, a wet etching bath for 150 mm wafers was designed to prevent back-side etching of silicon wafer, and we demonstrated the optimized process recipe to have anisotropic wet etching of silicon wafer without any damage on the backside. We also presented the design of wet bath for 300 mm wafer processing as a promising process development.

In this paper, we introduce the novel idea to improve the B/W usage efficiency of on-chip networks used for TAM to test multiple DRAMs. In order to avoid the local bottleneck of test packets caused by an ATE, we make test patterns using microcode-based instructions within ATE and adopt a test bus to transmit test responses from DRAM DFT (Design for Testability) called Test Generator (TG) to ATE. The proposed test platform will contribute to increasing the test economics of memory IC industry.

Image segmentation is the task of partitioning an image into multiple sets of pixels based on some characteristics. The objective is to simplify the image into a representation that is more meaningful and easier to analyze. In this paper, we apply deep-learning to pre-train the learning model, and implement an algorithm that performs image segmentation in real time by extracting frames for the stream input from the Android device. Based on the open source of DeepLab-v3+ implemented in Tensorflow, some convolution filters are modified to improve real-time operation on the Android platform.

In this paper, the new architecture of seizure prediction using CNN and LSTM and DWT was presented. In the proposed architecture, EEG data was labeled into a preictal and interictal section, and DWT was adopted to the preprocessing process to apply the characteristics of the time and frequency domain of the processed EEG signal. Also, CNN was applied to extract the spatial characteristics of each electrode used for EEG measurement, and LSTM neural network was applied to verify the logical order of the preictal section. The learning of the proposed architecture utilizes the CHB-MIT Scalp EEG dataset, and the sliding window technique is applied to balance the dataset between the number of interictal sections and the number of preictal sections. As a result of the simulation of the proposed architecture, a sensitivity of 81.22% and an FPR of 0.174 were obtained.

According to brilliant development of smart devices, many related services are being devised. And, almost every service is designed to provide user-centric services based on personal information. In this situation, to prevent unintentional leakage of personal information is essential. Conventionally, ID and Password system is used for the user authentication. This is a convenient method, but it has a vulnerability that can cause problems due to information leakage. To overcome these problem, many methods related to face recognition is being researched. Through this paper, we investigated the trend of user authentication through biometrics and a representative model for face recognition techniques. One is DeepFace of FaceBook and another is FaceNet of Google. Each model is based on the concept of Deep Learning and Distance Metric Learning, respectively. And also, they are based on Convolutional Neural Network (CNN) model. In the future, further research is needed on the equipment configuration requirements for practical applications and ways to provide actual personalized services.

Over the past several years, deep learning has been widely used for feature extraction in image and video for various applications such as object classification and facial recognition. This paper introduces an implantation of embedded Linux system for embossed digits recognition using CNN based deep learning methods. For this purpose, we implemented a coin recognition system based on deep learning with the Keras open source library on Raspberry PI. The performance evaluation has been made with the success rate of coin classification using the images captured with ultra-wide angle camera on Raspberry PI. The simulation result shows 98% of the success rate on average.

Today's placement machines can pick and place thousands of components per hour with a very high degree of accuracy. The packaged semiconductor chips are inserted into a carrier at regular intervals, covered with a tape to protect the chips from external impact, and supplied in a roll form. These packaging processes also progress rapidly in a consistent direction, affecting the peelback strength between the cover tape and carrier depending on the main process conditions. In this paper, we analyzed the main process variables that affect peelback strength in the reel tape packaging process for packaging semiconductor chips. The main effects and interactions were analyzed. The peelback strength range required in the packaging process was set as the nominal the best characteristics, and the optimum process condition satisfying this was derived.

In this paper, a heat pump using a Peltier device was developed for heat dissipation in a sealed electric box. The heat pump was designed with a cooling fin attached to both sides of the Peltier device, and a fan was mounted on the cooling fin on the hot side to increase the efficiency. The heat dissipation efficiency could be improved by directly connecting the electronic component having high heat to the cooling fin using a heat conducting wire. The fabricated heat pump was designed to operate only in the temperature range set by the temperature control system to improve the problem of high power consumption of the Peltier element.