• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.382-384
• /
• 2021

In this study, we implement a system that automatically classifies types of plastics using YOLO (You Only Look Once), a real-time object recognition algorithm. The system consists of Nvidia jetson nano, a small computer for deep learning and computer vision, with model trained to recognize plastic separation emission marks using YOLO. Using a webcam, recycling marks of plastic waste were recognized as PET, HDPE, and PP, and motors were adjusted to be classified according to the type. By implementing this automatic classifier, it is convenient in that it can reduce the labor of separating and discharging plastic separation marks by humans and increase the efficiency of recycling through accurate recycling.

• Journal of the Institute of Convergence Signal Processing
• /
• v.23 no.3
• /
• pp.137-142
• /
• 2022

Every year, the number of wild animals appearing in human settlements increases, resulting in increased damage to property and human life. In particular, the damage is more severe when wild animals appear on highways or farmhouses. To solve this problem, ecological pathways and guide fences are being installed on highways. In addition, in order to solve the problem in farms, horn repelling using sensors, installing a net, and repelling by smell of excrement are being used. However, these methods are expensive and their effectiveness is not high. In this paper, we used YOLO (You Only Look Once), an AI-based image analysis method, to analyze harmful animals in real time to reduce malfunctions, and high-brightness LEDs and ultrasonic frequency speakers were used as extermination devices. The speaker outputs an audible frequency that only animals can hear, increasing the efficiency to only exterminate wild animals. The proposed system is designed using a general-purpose board so that it can be installed economically, and the detection performance is higher than that of the devices using the existing sensor.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.703-704
• /
• 2023

학교폭력은 교육 환경에서 심각한 문제이다. 피해자에게 심리적 고통과 육체적 상해를 입히고 학교 내 안전과 안정성을 위협한다. 이에 많은 교육기관과 정부 기관이 학교폭력 예방과 대처를 위한 다양한 방안을 제시하고 있지만, 여전히 어려운 문제이다. 최근에는 인공지능 기술을 활용하여 학교폭력 방지와 대처에 관한 연구가 이루어지고 있다. 본 연구에서는 YOLOv5(You Only Look Once version 5) 딥러닝 알고리즘을 활용하여 학교 내부에서 발생하는 폭력 행위를 실시간으로 탐지하는 모델을 제안한다. 이 모델은 CCTV와 같은 영상 데이터를 입력으로 받아들여 학교 내부에서 발생하는 폭력 행위를 실시간으로 식별하는 것을 목표로 한다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.697-698
• /
• 2021

본 논문에서는 시각 장애우의 식품 안전성 증진을 위해 광학 문자 인식 (optical character recognition, OCR) 및 실시간 객체 인식 (you only look once, YOLO) 알고리즘에 기반한 식품의 유통기한 자동 알림 시스템을 제안한다. 제안하는 시스템은 1) 스마트폰 카메라를 통해 실시간으로 입력되는 영상에서 YOLO 알고리즘을 활용하여 유통기한으로 예측되는 이미지 영역을 검출하고, 2) 검출된 영역에서 OCR 알고리즘을 활용하여 유통기한 데이터를 추출하며, 3) 최종 추출된 유통기한 데이터를 음성으로 변환하여 시각 장애우에게 전달한다. 개발된 시스템은 유통기한 정보를 추출해서 사용자에게 전달하기까지 평균 약 7초 이내의 빠른 응답 속도를 보였으며, 62.8%의 객체 인식 정확도와 93.6%의 문자 인식 정확도를 보였다. 이러한 결과들은 제안하는 시스템을 시각 장애우들이 실용적으로 활용할 수 있다는 가능성을 보여준다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.28-30
• /
• 2021

In this paper, an algorithm that detects the user's finger direction using the YOLO (You Only Look Once) library was proposed. The processing stage of the proposed finger direction detection algorithm consists of a learning data management stage, a data learning stage, and a finger direction detection stage. As a result of the experiment, it was found that the distance between the camera and the finger had a very large influence on the accuracy of detecting the direction of the finger. We plan to apply this function to Turtlebot3 after improving the accuracy and reliability of the proposed algorithm in the future.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.31-33
• /
• 2021

In this paper, a dataset was designed and built to improve the accuracy of finger direction detection using an object detection algorithm based on You Only Look Once (YOLO). In order to improve the object detection performance, about 200 finger image data sets were trained, and to confirm that the detection accuracy differs from each other according to the angle of the palm, 50 comparison groups of different angles were configured and tested. As a result of the experiment, it was confirmed that the detection accuracy of palm located in a direction close to 90° is higher than that of other angles.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.01a
• /
• pp.315-316
• /
• 2022

최근 전 세계적인 탄소 중립 정책으로 인해 전기차 보급 속도는 예상보다 훨씬 빠르게 증가하고 있다. 하지만 늘어나는 수요에 비해 전기차 충전기 수는 턱없이 부족하다. 그뿐만 아니라 일반 차들의 전기차 충전소 불법주차로 인해 전기차가 충전하지 못하는 불편함이 발생하고 있다. 본 논문에서는 에지 컴퓨터(edge computer)와 딥러닝 기반 객체 감지 시스템 YOLO(You only look once)를 이용한 전기차 충전소 불법주차 방지 시스템을 개발한다. 먼저, 이 시스템은 카메라를 통해 실시간으로 영상을 받아 YOLO를 이용하여 차량 번호판 인식이 되면 전기차 번호판의 특정 마크를 인식하여 전기차인지 일반 차인지를 판별하여 판별된 값에 따라 주차 차단기가 작동되는 시스템이다. 전기차이면 차단기가 내려가서 충전소를 이용할 수 있게 하고 일반차일 경우 주차 차단기가 내려가지 않고 막아 불법주차를 차단한다. 이와 같은 기술을 활용하여 전기차 충전소 불법주차 방지에 기여하고자 한다.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.26 no.4
• /
• pp.27-39
• /
• 2023

Detecting greenhouses from the remote sensing datasets is useful in identifying the illegal agricultural facilities and predicting the agricultural output of the greenhouses. This research proposed a methodology for automatically detecting greenhouses from a given Planetscope satellite imagery acquired in the areas of Gimje City using the deep learning technique through a series of steps. First, multiple training images with a fixed size that contain the greenhouse features were generated from the five training Planetscope satellite imagery. Next, the YOLO(You Only Look Once) model was trained using the generated training images. Finally, the greenhouse features were detected from the input Planetscope satellite image. Statistical results showed that the 76.4% of the greenhouse features were detected from the input Planetscope satellite imagery by using the trained YOLO model. In future research, the high-resolution satellite imagery with a spatial resolution less than 1m should be used to detect more greenhouse features.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.16 no.6
• /
• pp.1329-1338
• /
• 2021

There are rarely effective methods to help visually impaired people when they want to know the expiration date of products excepted to only Braille. In this study, we developed an expiration date notification system based on YOLO and OCR for visually impaired people. The handicapped people can automatically know the expiration date of a specific product by using our system without the help of a caregiver, fast and accurately. The proposed system is worked by four different steps: (1) identification of a target product by scanning its barcode; (2) segmentation of an image area with the expiration date using YOLO; (3) classification of the expiration date by OCR: (4) notification of the expiration date by TTS. Our system showed an average classification accuracy of about 86.00% when blindfolded subjects used the proposed system in real-time. This result validates that the proposed system can be potentially used for visually impaired people.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.6
• /
• pp.907-917
• /
• 2023

Recently, a self-driving car have applied deep learning technology to advanced driver assistance system can provide convenience to drivers, but it is shown deep that learning technology is vulnerable to adversarial evasion attacks. In this paper, we performed five adversarial evasion attacks, including MI-FGSM(Momentum Iterative-Fast Gradient Sign Method), targeting the object detection algorithm YOLOv5 (You Only Look Once), and measured the object detection performance in terms of mAP(mean Average Precision). In particular, we present a method applying morphology operations for YOLO to detect objects normally by removing noise and extracting boundary. As a result of analyzing its performance through experiments, when an adversarial attack was performed, YOLO's mAP dropped by at least 7.9%. The YOLO applied our proposed method can detect objects up to 87.3% of mAP performance.