• Journal of the Korean Society of Food Culture
• /
• v.10 no.4
• /
• pp.227-233
• /
• 1995

This nutritional survey was conducted from February 8 to March 10, 1994, in order to investigate the nutritional knowledge, food habits, nutritional attitude, food preferences and nutrient intakes of urban middle-aged women living in Daegu and Pohang district. The subjects of this survey were 164 urban middle-aged women living in Daegu and Pohang Area. The completely answered questionnaires were analyzed for nutritional knowledge, food habits, nutritional attitude, food preference and nutrient intakes of urban middle-aged women. The results obtained are summarized as follows: The subjects had a high level of perceived knowledge (82.9%), that is the knowledge that each subject believed she had, but the accuracy of the knowledge was only 66.1%. The average nutrition knowledge score was 8.26 out of possible 15 points, and food habit score was 5.50 out of 10 points. Most of the subjects belonged to 'Fair' or 'Good' food habit group, which is considered to be relatively good. With increasing age, the percentage of perceived knowledge, accuracy, and nutrition knowledge score were getting lower. But food habit score and nutritional attitude score were getting higher at 40's women than 30's women. The correlation between nutritional knowledge score and food habit score was low (r=0.0748). The correlation between nutritional attitude score and food habit score was low, too (r=-0.0653). Food preferences for kimchi, potato, cooked rice, beef, noodle, cabbage and milk were high. Average calorie and protein intake of the subjects were $1967.4{\pm}27.8\;Kcal$, $75.8{\pm}1.4\;g$ respectively. Carbohydrate, protein and fat ratio on energy composition was 63.3%:15.5%:21.2%.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.4
• /
• pp.793-799
• /
• 2007

Generally, a semiconductor chip measured with a few micro units is captured by line scan camera for higher inspection accuracy. However, the faulty inspection requires an exact boundary detection algorithm, because it is very sensitive to scan speed and lighting conditions. In this paper we propose boundary detection with subpixel edge detection in order to increase the accuracy of bump faulty detection on chips. The bump edge is detected by first derivative to four directions from bump center point and the exact edge positions are searched by the subpixel method. Also, the exact bump boundary to calculate the actual bump size is computed by LSM(Least Squares Method) to minimize errors since the bump size is varied such as bump protrusion, bump bridge, and bump discoloration. Experimental results exhibit that the proposed algorithm shows large improvement comparable to the other conventional boundary detection algorithms.

• Korean Journal of Remote Sensing
• /
• v.19 no.2
• /
• pp.170-179
• /
• 2003

In order to classify Land Use/Land Cover using multispectral images, we have to give consequence to defining proper classes and selecting training sample with higher class separability. The process of satellite hyperspectral image which has a lot of bands is difficult and time-consuming. Furthermore, classification result of hyperspectral image with noise is often worse than that of a multispectral image. When selecting training fields according to the signatures in the study area, it is difficult to calculate covariance matrix in some clusters with pixels less than the number of bands. Therefore in this paper we presented an overview of feature extraction methods for classification of Hyperion data and examined effectiveness of feature extraction through the accuracy assesment of classified image. Also we evaluated the classification accuracy of optimal meaningful features by class separation distance, which is also a method for band reduction. As a result, the classification accuracies of feature-extracted image and original image are similar regardless of classifiers. But the number of bands used and computing time were reduced. The classifiers such as MLC, SAM and ECHO were used.

• Korean Journal of Remote Sensing
• /
• v.22 no.5
• /
• pp.373-377
• /
• 2006

Seasonal threshold values for fog detection over the ten airport areas in the Korean Peninsula have been derived, using the satellite-observed data of polar-orbit (Aqua/Terra MODIS) and geostationary (GOES-9) during two years. The values are obtained from reflectance at $0.65{\mu}m\;(R_{0.65})$ and the difference in brightness temperature between $3.7{\mu}m\;and\;11{\mu}m\;(T_{3.7-11})$. In order to examine the discrepancy between the threshold values of two kinds of satellites, the following parameters have been analyzed under the condition of daytime/nighttime and fog/clear-sky, utilizing their simultaneous observations over the Seoul Metropolitan Area. The parameters are the brightness temperature at $3.7{\mu}m\;(T_{3.7})$, the temperature at $11{\mu}m\;(T_{11}$, and $T_{3.7-11}$ for day and night. The $R_{0.65}$ data are additionally included in the daytime. The GOES-9 thresholds over the seven airport areas except the Cheongju airport have revealed the accuracy of 50% in the daytime and 70% in the nighttime, based on statistical verification for the independent samples as follows; FAR, POD and CSI. However, the accuracy decreases in the foggy cases with twilight, precipitation, short persistence, or the higher cloud above fog.

• Journal of Digital Convergence
• /
• v.19 no.12
• /
• pp.377-382
• /
• 2021

In order to advance forest resource management, it is necessary to solve problems such as the aging of forest-related industry workers and the field investigation system centered on manpower. Therefore, in this study, the trend of establishment and utilization of overseas forest geospatial information applied with the latest technology for scientific forest resource management was investigated to identify the domestic application plan. Overseas, photogrammetry and LiDAR technologies were being used to construct and utilize forest geospatial information. In the case of photogrammetry, it was used to measure the volume of vegetation, diameter, and tree height. And LiDAR technology has been applied to the measurement of diameter, and tree height. Through the analysis of overseas cases, it was identified how to construct forest geospatial information using photogrammetry and LiDAR, and it was found that LiDAR showed higher accuracy than photogrammetry. In the future, if the construction of forest geospatial information using various LiDAR sensors are performed and the accuracy and work efficiency are analyzed, it will be possible to present the possibility of using new technologies in the construction of forest geospatial information in Korea.

• Journal of Biomedical Engineering Research
• /
• v.40 no.2
• /
• pp.62-67
• /
• 2019

The purpose of this study was to evaluate the performance of deep neural network model in order to determine whether there is a risk factor for coronary artery disease based on the cardiac variation parameter. The study used unidentifiable 297 data to evaluate the performance of the model. Input data consists of heart rate parameters, which are SDNN (standard deviation of the N-N intervals), PSI (physical stress index), TP (total power), VLF (very low frequency), LF (low frequency), HF (high frequency), RMSSD (root mean square of successive difference) APEN (approximate entropy) and SRD (successive R-R interval difference), the age group and sex. Output data are divided into normal and patient groups, and the patient group consists of those diagnosed with diabetes, high blood pressure, and hyperlipidemia among the various risk factors that can cause coronary artery disease. Based on this, a binary classification model was applied using Deep Neural Network of deep learning techniques to classify normal and patient groups efficiently. To evaluate the effectiveness of the model used in this study, Kernel SVM (support vector machine), one of the classification models in machine learning, was compared and evaluated using same data. The results showed that the accuracy of the proposed deep neural network was train set 91.79% and test set 85.56% and the specificity was 87.04% and the sensitivity was 83.33% from the point of diagnosis. These results suggest that deep learning is more efficient when classifying these medical data because the train set accuracy in the deep neural network was 7.73% higher than the comparative model Kernel SVM.

• Smart Media Journal
• /
• v.9 no.4
• /
• pp.9-16
• /
• 2020

Recently, various studies are being conducted to integrate Image Segmentation into smart factory industries and autonomous driving fields. In particular, Image Segmentation systems using deep learning algorithms have been researched and developed enough to learn from large volumes of data with higher accuracy. In order to use image segmentation in the autonomous driving sector, sufficient amount of learning is needed with large amounts of data and the streaming environment that processes drivers' data in real time is important for the accuracy of safe operation through highways and child protection zones. Therefore, we proposed a novel DFCN algorithm that enhanced existing FCN algorithms that could be applied to various road environments, demonstrated that the performance of the DFCN algorithm improved 1.3% in terms of "loss" value compared to the previous FCN algorithms. Moreover, the proposed DFCN algorithm was applied to the existing U-Net algorithm to maintain the information of frequencies in the image to produce better results, resulting in a better performance than the classical FCN algorithm in the autonomous environment.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.7
• /
• pp.972-980
• /
• 2022

The butterfly species recognition technology based on machine learning using images has the effect of reducing a lot of time and cost of those involved in the related field to understand the diversity, number, and habitat distribution of butterfly species. In order to improve the accuracy and time efficiency of butterfly species classification, various features used as the inputs of machine learning models have been studied. Among them, branch length similarity(BLS) entropy or color intensity entropy methods using the concept of entropy showed higher accuracy and shorter learning time than other features such as Fourier transform or wavelet. This paper proposes a feature extraction algorithm using RGB color intensity entropy for butterfly color images. In addition, we develop butterfly recognition systems that combines the proposed feature extraction method with representative ensemble models and evaluate their performance.

• International Journal of Computer Science & Network Security
• /
• v.23 no.9
• /
• pp.150-156
• /
• 2023

For a doctor, diagnosing a patient's heart disease is not easy. It takes the ability and experience with high flying hours to be able to accurately diagnose the type of patient's heart disease based on the existing factors in the patient. Several studies have been carried out to develop tools to identify types of heart disease in patients. However, most only focus on the results of patient answers and lab results, the rest use only echocardiography data or electrocardiogram results. This research was conducted to test how accurate the results of the classification of heart disease by using two medical data, namely echocardiography and electrocardiogram. Three treatments were applied to the two medical data and analyzed using the decision tree approach. The first treatment was to build a classification model for types of heart disease based on echocardiography and electrocardiogram data, the second treatment only used echocardiography data and the third treatment only used electrocardiogram data. The results showed that the classification of types of heart disease in the first treatment had a higher level of accuracy than the second and third treatments. The accuracy level for the first, second and third treatment were 78.95%, 73.69% and 50%, respectively. This shows that in order to diagnose the type of patient's heart disease, it is advisable to look at the records of both the patient's medical data (echocardiography and electrocardiogram) to get an accurate level of diagnosis results that can be accounted for.

• Journal of Platform Technology
• /
• v.11 no.4
• /
• pp.3-12
• /
• 2023

In the post-COVID era, the importance of quarantine measures is greatly emphasized, and accordingly, research related to the detection of mask wearing conditions and prevention of other infectious diseases using deep learning is being conducted. However, research on the detection and tracking of visitors to cultural facilities to prevent the spread of diseases is equally important, so research on this should be conducted. In this paper, a convolutional neural network-based object detection model is trained through transfer learning using a pre-collected dataset. The weights of the trained detection model are then applied to a multi-object tracking model to monitor visitors. The visitor detection model demonstrates results with a precision of 96.3%, recall of 85.2%, and an F1-score of 90.4%. Quantitative results of the tracking model include a MOTA (Multiple Object Tracking Accuracy) of 65.6%, IDF1 (ID F1 Score) of 68.3%, and HOTA (Higher Order Tracking Accuracy) of 57.2%. Furthermore, a qualitative comparison with other multi-object tracking models showcased superior results for the model proposed in this paper. The research of this paper can be applied to the hygiene systems within cultural facilities in the post-COVID era.