• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.9
• /
• pp.1330-1339
• /
• 2022

3D object detection generally aims to detect relatively large data such as automobiles, buses, persons, furniture, etc, so it is vulnerable to small object detection. In addition, in an environment with limited resources such as embedded devices, it is difficult to apply the model because of the huge amount of computation. In this paper, the accuracy of small object detection was improved by focusing on local features using only one layer, and the inference speed was improved through the proposed knowledge distillation method from large pre-trained network to small network and adaptive quantization method according to the parameter size. The proposed model was evaluated using SUN RGB-D Val and self-made apple tree data set. Finally, it achieved the accuracy performance of 62.04% at mAP@0.25 and 47.1% at mAP@0.5, and the inference speed was 120.5 scenes per sec, showing a fast real-time processing speed.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.15 no.4
• /
• pp.55-61
• /
• 2015

Today, IT technology provide convenience to many people. Smartphone era is opened, and market environment is changing rapidly. Pornography market is active by using smartphone use free internet. Many people access mobile harmful site of USA and Japan. App store of the apple has been cut off the porn service, but access block to mobile Web page is an impossible situation. In this paper, we proposed the harmful image detection method of using skin and non skin features to detect harmful image. Our proposed method can provide enough performance than previous method.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.1
• /
• pp.15-21
• /
• 2023

This paper is a study on the development of an artificial intelligence model that recognizes a hole cup on a golf green. A CNN-based object detection algorithm was used to recognize the hole cup on the green. Also, Apple's CreateML was used to create a model of the object detection algorithm. This paper created a JSON file with 120 training images and annotations to meet the needs of CreateML. In addition, for more accurate learning, data amplification algorithm was used for learning data and 288 learning data were used for learning. By changing the Iterations, Batch size, and Grid size required by CreateML, we found parameter values that improve the performance of the model. A prototype app was developed by applying the developed model, and performance was measured on an actual golf course green using the prototype app. As a result of the measurement, it was found that the hole cup was accurately recognized within 10m, which is the typical golfer's putting distance.

• Mass Spectrometry Letters
• /
• v.8 no.4
• /
• pp.105-108
• /
• 2017

Recently, deciduous teeth have been proposed as a promising biomatrix for estimating internal and external chemical exposures of an individual from prenatal periods to early childhood. Therefore, detection of organic chemicals in teeth has received increasing attention. Organic materials in tooth matrix are mostly collagen type proteins, but lipids and other small organic chemicals are also present in the tooth matrix. In this study, matrix-assisted laser desorption/ionization (MALDI) mass spectrometry (MS) was employed to obtain lipid fingerprints from deciduous teeth. Phospholipids and triacylglcerols (TAGs) from deciduous teeth were successfully detected by MALDI MS with 2,5-dihydroxybenzoic acid (DHB) or gold nanoparticle (AuNP) as a matrix.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2002.11b
• /
• pp.869-872
• /
• 2002

네트워크 기반 침입탐지시스템은 연속적으로 발생하는 패킷의 무손실 축소와, 패킷으로 정상 또는 비정상 행위패턴을 정확히 모델링한 모델 생성이 전체성능을 판단하는 중요한 요소가 된다. 네트워크 기반 비정상행위 판정 침입탐지시스템에서는 이러한 탐지모델 구축을 위해 주로 감독학습 알고리즘을 사용한다. 본 논문은 탐지모델 구축에 사용하는 감독 학습 방식이 가지는 문제점을 지적하고, 그에 대한 대안으로 비감독 학습방식의 학습알고리즘을 제안한다. 감독 학습을 사용하여 탐지모델을 구축하기 위해서는 정상행위의 패킷을 취합해야 하는 사전 부담이 있는 반면에 비감독 학습을 사용하게 되면 이러한 사전작업 없이 탐지모델을 구축할 수 있다. 본 논문에서는 비감독학습 알고리즘을 비교 분석하기 위해서 COBWEB, k-means, Autoclass 알고리즘을 사용했으며, 성능을 평가하기 위해서 비정상행위도(Abnormal Behavior Level)를 계산하여 에러율을 구하였다.

• Mass Spectrometry Letters
• /
• v.9 no.4
• /
• pp.110-114
• /
• 2018

Since accumulation of chemicals in deciduous teeth can occur from the second trimester of fetal development to shedding, a deciduous tooth has been considered as an attractive biomatrix for estimating individual chemical exposures recently. Therefore, detection of organic chemicals from teeth has received an increasing attention in exposomics research. Most previous studies on organic chemical analysis of teeth not only focused on a few targeted chemicals but also ignored potential contaminants from an enamel surface or a dental pulp. Recently, our group started developing a multi-class organic analysis method for deciduous teeth and tried to find a proper incubation condition of tooth materials. Our results showed that incubation with methanolic HCl provided the best performance among tested.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2024.01a
• /
• pp.297-298
• /
• 2024

펜데믹, 전쟁 등을 포함한 국제 정세 변화에 따른 물류대란, 원자재가격 상승 및 환율 급등으로 인해 2023년 기준 대한민국의 물가는 크게 오르고 있는 추세이다. 물가 상승은 사업장의 인건비 부담 증가로 이어지고 있고 특히 노동 집약 산업인 농업 분야에서의 인건비 부담 문제는 더욱 심각한 실정이다. 외국인 근로자 고용이 대안이 될 수 있지만 인건비 절감 효과는 미미하기에 농업계 관계자들은 자동화 시스템 도입에 관심이 집중되고 있다. 따라서, 본 논문에서는 사과 분류 작업 자동화 체계의 핵심 요소에 해당하는 사과 품질 실시간 모니터링 시스템을 제안한다. 제안한 방식에서는 딥러닝 기반의 영상 분석 기법 및 무게 센서 데이터 분석을 통해 사과의 품질에 따른 등급 책정을 자동화 한다.

• The Bulletin of The Korean Astronomical Society
• /
• v.43 no.1
• /
• pp.64.4-65
• /
• 2018

We present a study of citizen science performed at Korean Minjok Leadership Academy (KMLA). The importance of citizen involvements in scientific studies has been increasing, with remarkable results and performances. For instance, the discovery of an impact scar near Jupiter's south polar region (A. Wesley, 2009) led to an international campaign of professional observations to understand the asteroidal collision responsible for the scar. Citizen science at KMLA has been and will be mainly conducted by members of the astronomical observation club 'Apple-Pie' through amateur telescopes. Members of 'Apple-Pie' are specialized in various fields related to astronomy, from planetary science to cosmology. The spectrum not only includes fields that are directly related to astronomy but also fields such as computer science and astrophotography. The scheduled construction of a new observatory will further enable students to participate in higher level projects such as planetary monitoring over long timescales and the observation and detection of solar system bodies and exoplanets. In addition, a new supervisor with expertise and research experience in galactic astronomy, planetary science, and meteorology has joined the school faculty. He will supplement students with fundamental theoretical backgrounds and essential research techniques to enhance astronomical research at KMLA. KMLA's ultimate goal is to deploy a remote-controlled observatory available to aspiring scientists around the world to create a network of citizen science system. The prime observational conditions of KMLA and the willingness of the students and faculty members will provide a competitive edge for KMLA over other similar institutes in Korea.

• Applied Biological Chemistry
• /
• v.35 no.2
• /
• pp.115-121
• /
• 1992

Avermectin B1(abamectin) is an very effective acaricide/insecticide. Because it is applied at low rates, the resultant residue level would be quite low and this requires highly sensitive analytical method. In this study, three analytical methods for avermectin $B1_a$, were compared in view of detectability and sensitivity. The first analytical method was an HPLC method employing the fluorescence detection of avermectin. The second analytical technique to quantitate avermectin $B1_a$, and its photodegradative delta 8,9-isomer employed trifluoroacetic anhydride and 1-methylimidazole in DMF. The new method was the modification of trifluorescence method. The average recoveries of avermectin $B1_a$ for the concentration range from 1 and 10 ng/g in whole apple fruit by fluorescence method were 90.3% and 88.2% respectively. In trifluorescence method, the recoveries of the avermectin $B1_a$ and delta 8,9-isomer were 100.7% and 94.7% in concentration from 5 ng/g and 25 ng/g. The average recoveries of 5 ng and 25 ng/g in the newly modified method were 95.0, 99.0. 96.0, 92.8% in whole apple and soil respectively.

• Korean Journal of Food Science and Technology
• /
• v.32 no.3
• /
• pp.531-537
• /
• 2000

The simple microgel electrophoresis of single cells, a 'comet assay', on fruit seeds enabled the rapid identification of irradiated fruits by comparing the intact non-irradiated cells and the damaged cells of irradiated fruits. Grapes and plums were irradiated with 0.1, 0.5, 0.7, 1.0 kGy and strawberries, peaches, apples, and nectarines were irradiated with only 1.0 kGy. Seeds were isolated, crushed, and the suspended cells were embedded in an agarose layer. After lysis of the cells, they were subjected to microgel electrophoresis for 2 minutes, and then stained. The DNA radiation-induced fragmentation of all the fruits stretched and migrated out of the cells forming a tail toward the anode giving the appearance of a comet, while the undamaged cells appeared as intact nuclei without tails. Grape and plum seeds irradiated at 0.5 kGy and higher showed significant increases in tail length. With increasing the irradiation doses, longer extention of the DNA from the nucleus toward the anode was observed. Strawberry, peach, apple, and nectarine seeds irradiated with 1.0 kGy also showed the longer tails than non-irradiated ones. DNA comet assay as a rapid and inexpensive screening technique could be an officially validated method for the detection of irradiated fruits.