• The KIPS Transactions:PartB
• /
• v.14B no.3 s.113
• /
• pp.191-198
• /
• 2007

In this paper, we present a marker detection and recognition method from camera image for a disabled person to interact with a server system which can control appliance of surrounding environment. It converts the camera image to a binary image by using multi-threshold and extracts contours of objects in the binary image. After that, it approximates the contours to a list of line segments. It finds rectangular markers by using geometrical features which are extracted from the approximated line segments. It normalizes the shape of extracted markers into exact squares by using the warping technique. It extracts feature vectors from marker image by using principal component analysis and then recognizes the marker. The proposed marker recognition system is robust for light change by using multi-threshold. Also, it is robust for angular variation of camera by using warping technique and principal component analysis. Experimental results show that the proposed method achieves 100% recognition rate at maximum for 21 markers and execution speed of 12 frames/sec.

• Food Science of Animal Resources
• /
• v.42 no.5
• /
• pp.833-848
• /
• 2022

The objective of this study was to characterize the meat quality traits that affect the texture and savory taste of Ufipa indigenous chickens by comparing the proximate composition, physical characteristics, collagen, and nucleic acid contents with those of commercial broilers. It was found that Ufipa chicken breast and thigh meat had a higher protein content (p<0.05) than broiler chicken meat, whereas the fat content was lower (p<0.01). The moisture content of thigh meat was lower in Ufipa chicken meat than in broiler chicken meat (p<0.05). Regarding meat color, broiler chickens had considerably higher L^* and b^* than Ufipa chickens in both the breast and the thigh meat, except for a^* (p<0.01). Regarding water holding capacity, Ufipa chicken breast exhibited higher drip loss but lower thawing and cooking losses than broiler chicken (p<0.01). In contrast, its thigh meat had a much lower drip and thawing losses but higher cooking losses (p<0.01). The shear force of Ufipa chickens' breasts and thighs was higher than that of broiler chickens (p<0.05), while the amount of total collagen in the thigh meat was higher than that of broiler chickens (p<0.05). Additionally, the inosine-5'-monophosphate (IMP) of Ufipa chicken breast and thigh meat was higher than that of broiler meat (p<0.05). The principal component analysis of meat quality traits provides a correlation between the proximate and physical-chemical prosperties of both breeds with some contrast. In conclusion, the present study provides information on healthy food with good-tasting Ufipa indigenous chickens, which offer a promising market due to consumers' preferences.

• Smart Structures and Systems
• /
• v.31 no.5
• /
• pp.485-500
• /
• 2023

Machine learning-based structural health monitoring (ML-based SHM) methods are researched extensively in the recent decade due to the availability of advanced information and sensing technology. ML methods are well-known for their pattern recognition capability for complex problems. However, the main obstacle of ML-based SHM is that it often requires pre-collected historical data for model training. In most actual scenarios, damage presence can be detected using the unsupervised learning method through anomaly detection, but to further identify the damage types would require prior knowledge or historical events as references. This creates the cold-start problem, especially for new and unobserved structures. Modal-based methods identify damages based on the changes in the structural global properties but often require dense measurements for accurate results. Therefore, a two-stage hybrid modal-machine learning damage detection scheme is proposed. The first stage detects damage presence using Principal Component Analysis-Frequency Response Function (PCA-FRF) in an unsupervised manner, whereas the second stage further identifies the damage. To solve the cold-start problem, mode shape assessment using the first mode is initiated when no trained model is available yet in the second stage. The damage identified by the modal-based method would be stored for future training. This work highlights the performance of the scheme in alleviating the cold-start issue as it transitions through different phases, starting from zero damage sample available. Results showed that single and multiple damages can be identified at an acceptable accuracy level even when training samples are limited.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.57 no.4
• /
• pp.489-496
• /
• 2024

This study has investigated ten different organic acids in 67 types of seasoning Jeotgal (seasoning salted fermented product, SSFP) and Sikhae (seasoning salted fermented product with rice, SSFPR) collected from the market. The levels of organic acids varied significantly depending on the raw materials used for the SSFP and SSFPR. The acetic acid and lactic acid content ranged from being not detected (ND) to 5,667.15 mg/kg for acetic acid in the SSFP, ND to 34,837.89 mg/kg for lactic acid in the SSFP, and 53.47 to 2,089.55 mg/kg for acetic acid in the SSFPR, and 10.69 to 1,733.27 mg/kg for lactic acid in the SSFPR. The range for succinic acid was 57.92 to 2,793.09 mg/kg in the SSFP and 21.21 to 453.32 mg/kg in the SSFPR. Propionic acid content ranged from 8.42 to 277.83 mg/kg in the SSFP and 13.21 to 133.07 mg/kg in SSFPR.The most abundant organic acids were succinic and lactic acid, comprising 48% of the seasoning Jeotgal (succinic acid) and more than 72% in Sikhae (lactic acid). Furthermore, this study suggests that the differences in the organic acid content of samples, based on their raw materials can be discriminated by principal component analysis (PCA).

• Conservation Science in Museum
• /
• v.28
• /
• pp.89-108
• /
• 2022

Among about 500,000 characters in metal types in National Museum of Korea, this study conducts a statistical analysis of 62 metal types from the early Joseon Dynasty, including 33 gabinja (甲寅字) types and 29 eulhaeja (乙亥字) Hangeul types by examining the shape, measuring the specific gravity, and identifying the components based on previously-studied Joseon metal types. Among them, 33 gabinja types and 24 eulhaeja types were made of two-component bronze (copper and tin) (Group A), and four eulhaeja types were produced with three-component bronze (copper, tin and lead). (Group B), and one eulhaeja type was created with two-component bronze (copper and tin) with a high tin content (Group C). By comparing with imjinja (壬辰字) types of the late Joseon Dynasty based on multiple statistical analyses of type components, this study confirms that late-Joseon types have low copper content and high zinc and lead content, and therefore it may be possible to distinguish between the types of early and late Joseon Dynasty.

• Journal of KIISE:Software and Applications
• /
• v.36 no.10
• /
• pp.836-850
• /
• 2009

Human posture recognition has attracted tremendous attention in ubiquitous environment, performing arts and robot control so that, recently, many researchers in pattern recognition and computer vision are working to make efficient posture recognition system. However the most of existing studies is very sensitive to human variations such as the rotation or the translation of body. This is why the feature, which is extracted from the feature extraction part as the first step of general posture recognition system, is influenced by these variations. To alleviate these human variations and improve the posture recognition result, this paper presents 3D Star Skeleton and Principle Component Analysis (PCA) based feature extraction methods in the multi-view environment. The proposed system use the 8 projection maps, a kind of depth map, as an input data. And the projection maps are extracted from the visual hull generation process. Though these data, the system constructs 3D Star Skeleton and extracts the rotation invariant feature using PCA. In experimental result, we extract the feature from the 3D Star Skeleton and recognize the human posture using the feature. Finally we prove that the proposed method is robust to human variations.

• Korean journal of applied entomology
• /
• v.32 no.3
• /
• pp.354-371
• /
• 1993

Behavior of adult females of the smoky brown cockroach was observed for 10-15 days continuously by using computer and the automatic sensoring system. Under the light condition of l2L~ 12D, individual variations were generally higher and the periodicity appeared less in terms of the presence time at the micro-habitats and the locomotory activity. The smoky brown cockroach appeared 20.42, 11.50, 6.31 and 2.66 % in a day in averages respectively at the shelter and the places for other individuals, feeding and drinking. It stayed 20.29 % in a day at the shelter when food, water, and other individuals were not supplied. Visiting rates were higher at the feeding and dnnking places than at. the ot.her micro-habitats. The degree of t.he locomotory activity was relatively lower when food, water, and other individuals were not supplied t.han when they were supplied. The Pnncipal Component Analysis (PCA) on the presence at the micro-habitats showed that. t.he presence pattern for each mdividual appeared differently (Q mode) while the differem time zones were grouped to the photophase and scotophase (R mode). When food, water, and other individuals were supplied the degree of grouping was higher at the shelter than at the places for feeding and drinking. When the act.ivityand the presence time at the different micro-habitats were jointly analyzed by PCA, the achvity of the female smoky brown cockroach appeared in relation with the presence time ( %) at the places for feeding and other individuals.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.11
• /
• pp.2105-2110
• /
• 2008

Embedding the compact wearable units to monitor the health status of a person has been analysed as a convenient solution for the home health care. This paper presents a method to detect fall from the other activities of daily living and also to classify those activities. This kind of ambulatory monitoring of the elderly and people with limited mobility can not only provide their general health status but also alarms whenever an emergency such as fall or gait has been occurred and a help is needed. A timely assistance in such a situation can reduce the loss of life. This work shows a detailed analysis of the data received from a chest worn sensor unit embedding a 3-axis accelerometer and depicts which features are important for the classification of human activities. How to arrange and reduce the features to a new feature set so that it can be classified using a simple classifier and also improving the classification resolution. Principal component analysis (PCA) has been used for modifying the feature set and afterwards for reducing the size of the same. Finally a Neural network classifier has been used to analyse the classification accuracies. The accuracy for detection of fall events was found to be 86%. The overall accuracy for the classification of Activities or daily living (ADL) and fall was around 94%.

• Korean Journal of Pharmacognosy
• /
• v.49 no.2
• /
• pp.145-154
• /
• 2018

Aconiti Kusnezoffii Tuber has been traditionally used to treat the symptoms of rheumatoid arthritis and joint pain. The main constituents are diterpenoid alkaloids such as benzoylmesaconine, benzoylaconine, mesaconitine, aconitine, and hypaconitine. In Korea, Aconiti Kusnezoffii Tuber is officially defined as the tubers of Aconitum kusnezoffii Reichb., A. ciliare Decasisne, and A. triphyllum Nakai. On the other hand, only the tuber of A. kusnezoffii is to be used in China. In order to identify the botanical origin of Aconiti Kusnezoffii Tuber circulated in Korea, we analyzed 24 samples of Aconiti Kusnezoffii Tuber obtained from local markets for comparative DNA analysis. The sequence analysis of nrRNA ITS 1 was useful to distinguish Aconitum species and revealed that the roots of A. karakolicum were circulated in Korean markets without discretion. HPLC quantitative analysis showed that aconitine was detected at the highest amount in A. karakolicum. Authentic diterpenoid alkaloids were coinjected for quantification of aconitine-type ingredients. All data were statistically grouped by Principal Component Analysis (PCA). This study suggests that both molecular and chemical analyses should be utilized for the standardization and the quality control for Aconiti Kusnezoffii Tuber.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.40 no.11
• /
• pp.1623-1630
• /
• 2011

The effects of addition of citrus peel powders (C 0, 0.5, 1 & 2%), dried Lentinus edodes powders (L 0, 0.5, 1 & 2%), and their combination (C-L) on the chemical, sensory and textural properties of pork sausages were studied. Addition of 0.5, 1 or 2% C, L, and C-L all significantly decreased moisture content, pH, and color a-values of sausage samples, whereas ash content and color b-value were increased (p<0.05). C, L, and C-L did not affect protein, fat, carbohydrates contents or texture characteristics. Sensory evaluation was performed by multivariate data analysis, namely principal component analysis (PCA). Eighty-two percent total variation was observed in the main structured information among the test groups: the first (PC1) and second (PC2) components of variation were 59 and 23%, respectively. Eight parameters (sweet flavor, pork aroma, bitterness, rancidity, salty flavor, color, sour flavor and citrus aroma) were utilized to describe the main sensory characteristic of the sausages. Addition of 0.5, 1 & 2% citrus peel was obviously correlated with PC1 (salty flavor, sour flavor and citrus aroma, pork aroma, and sweet flavor and rancidity), whereas addition of 0.5 & 1% Lentinus edodes was related with PC2 (aroma and rancidity).