• Journal of the Korea Society of Computer and Information
• /
• v.27 no.2
• /
• pp.25-32
• /
• 2022

In this paper, we propose a pet monitoring system based on deep learning using an activity recognition device. The system consists of a pet's activity recognition device, a pet owner's smart device, and a server. Accelerometer and gyroscope data were collected from an Arduino-based activity recognition device, and the number of steps was calculated. The collected data is pre-processed and the amount of activity is measured by recognizing the activity in five types (sitting, standing, lying, walking, running) through a deep learning model that hybridizes CNN and LSTM. Finally, monitoring of changes in the activity, such as daily and weekly briefing charts, is provided on the pet owner's smart device. As a result of the performance evaluation, it was confirmed that specific activity recognition and activity measurement of pets were possible. Abnormal behavior detection of pets and expansion of health care services can be expected through data accumulation in the future.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.5-5
• /
• 2023

제3차 우주개발 진흥 기본계획의 일환으로써 개발되는 차세대 중형위성 5호인 수자원위성은 수자원/수재해 감시 전용 위성으로 2025년 발사 예정이다. 수자원위성의 메인 센서인 C-band 영상레이더(Synthetic Aperture Radar, SAR)는 기상조건 및 주야 상관없이 지표면 관측이 가능한 센서로 급변하는 수재해 양상에 효과적으로 대응하기 위해 탑재된 센서이다. 본 연구사업은 차세대 수자원위성의 효과적 활용 방안 및 SAR 자료기반의 활용산출물 및 주제도 서비스를 위한 알고리즘 구조설계 및 표출시스템 시범개발을 목표로 하고 있으며, 홍수/가뭄/안전/환경모니터링을 주제로 수자원 및 원격탐사 분야의 다학제적 전문가들로 구성된 컨소시엄을 구성하여 추진하고 있다. 본 연구의 내용은 가뭄 모니터링을 위해 개발 중인 SAR 기반 토양수분과 농업적 가뭄지수 산정 알고리즘 개발 및 공간적 표출을 포함한다. 토양수분은 SAR 영상에서 지표피복별로 추출된 후방산란계수와 수문학적 개념의 융합을 통해 논/밭/산림에 대해 산정한다. 물리적 특성에 기반한 변화탐지모델을 활용해 토양수분량을 추출 후, 기계학습기법과 S C S - C N 방법에서 파생된 수문학적 개념 5일 선행강우량과 결합한 토양수분 산정 알고리즘을 개발하였다. 산정된 토양수분을 기반으로, 논 지역은 벼 재배에 따른 담수 시기를 고려한 토양의 포화/불포화상태, 밭 지역은 토양 종류에 따른 토양의 물리적 특성, 산림 지역은 수문학적 개념 및 식생지수를 활용하여 가뭄 판단 기준을 구축하고, 가뭄의 해갈 여부와 해갈되는 시점의 강우량을 산정 가능한 알고리즘을 개발하였다. 개발된 가뭄 모니터링 기법은 향후 고도화, 최적화 및 안정화를 통해 수자원위성의 핵심 활용기술로써 구현할 계획이다.

• Journal of the Institute of Convergence Signal Processing
• /
• v.23 no.3
• /
• pp.150-159
• /
• 2022

This paper presents an implementation of detecting whether a helmet is worn and there is a fall accident through individual image analysis in real-time from extracting the image objects of several workers active in the industrial field. In order to detect image objects of workers, YOLO, a deep learning-based computer vision model, was used, and for whether a helmet is worn or not, the extracted images with 5,000 different helmet learning data images were applied. For whether a fall accident occurred, the position of the head was checked using the Pose real-time body tracking algorithm of Mediapipe, and the movement speed was calculated to determine whether the person fell. In addition, to give reliability to the result of a falling accident, a method to infer the posture of an object by obtaining the size of YOLO's bounding box was proposed and implemented. Finally, Telegram API Bot and Firebase DB server were implemented for notification service to administrators.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.4
• /
• pp.19-30
• /
• 2009

In this paper, We propose a method with which we can detect facial components and face in input image. We use eye map and mouth map to detect facial components using eyes and mouth. First, We find out eye zone, and second, We find out color value distribution of skin region using the color around the eye zone. Skin region have characteristic distribution in YCbCr color space. By using it, we separate the skin region and background area. We find out the color value distribution of the extracted skin region and extract around the region. Then, detect mouth using mouthmap from extracted skin region. Proposed method is better than traditional method the reason for it comes good result with accurate mouth region.

• Journal of the Semiconductor & Display Technology
• /
• v.22 no.1
• /
• pp.11-17
• /
• 2023

This paper proposes a lightweight deep learning network for detecting an image splicing forgery. The research on image forgery detection using CNN, a deep learning network, and research on detecting and localizing forgery in pixel units are in progress. Among them, CAT-Net, which learns the discrete cosine transform coefficients of images together with images, was released in 2022. The DCT coefficients presented by CAT-Net are combined with the JPEG artifact learning module and the backbone model as pre-learning, and the weights are fixed. The dataset used for pre-training is not included in the public dataset, and the backbone model has a relatively large number of network parameters, which causes overfitting in a small dataset, hindering generalization performance. In this paper, this learning module is designed to learn the characterization depending on the DCT domain in real-time during network training without pre-training. The DCT RRU-Net proposed in this paper is a network that combines RRU-Net which detects forgery by learning only images and JPEG artifact learning module. It is confirmed that the network parameters are less than those of CAT-Net, the detection performance of forgery is better than that of RRU-Net, and the generalization performance for various datasets improves through the network architecture and training method of DCT RRU-Net.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.2
• /
• pp.235-241
• /
• 2022

Several studies using Convolutional Neural Network (CNN) have been conducted to detect and classify the sounds of marine mammals in underwater acoustic data collected through passive acoustic monitoring. In this study, the possibility of automatic classification of bearded seal sounds was confirmed using a CNN model based on the underwater acoustic spectrogram images collected from August 2017 to August 2018 in East Siberian Sea. When only the clear seal sound was used as training dataset, overfitting due to memorization was occurred. By evaluating the entire training data by replacing some training data with data containing noise, it was confirmed that overfitting was prevented as the model was generalized more than before with accuracy (0.9743), precision (0.9783), recall (0.9520). As a result, the performance of the classification model for bearded seals signal has improved when the noise was included in the training data.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.1D
• /
• pp.211-217
• /
• 2006

Growing urban areas modify patterns of local land use and land cover. Land use changes associated with urban expansion. One way to understand and document land use change and urbanization is to establish benchmark maps compiled from satellite imagery. Old satellite Imagery is useful data to extract urban information. CORONA is a photo satellite reconnaissance program used from 1960 to 1972 and its imagery was declassified and has been available to the public since 1995. Since CORONA images are collected with panoramic cameras, several types of geometric distortions are involved. In this study we proposed mathematical modeling method which use modified collinearity equations. After the geometric modeling, we mosaicked images. We can successfully extract urban boundaries using GLCM method and visual interpretation in CORONA (1972) and SPOT (1995) imagery and detect urban changes in Seoul quantitatively.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.406-409
• /
• 2022

In recent years, the developments in technologies, such as AI systems, Blockchain, VR/AR, 3D printing, robotics, and nanotechnology, are reshaping the future of healthcare right before our eyes. And also, healthcare has seen a paradigm shift towards prevention-oriented medicine, with a focus on consumers requirements. The spread of infectious diseases such as Covid-19 have altered the definition of healthcare and treatment facilities, necessitating immediate action to redesign hospitals' physical environments, adapt communication models to address social distancing requirements, implement virtual health solutions, and establish new clinical protocols. Hospitals, which have traditionally served as the hub of healthcare systems, are pursuing or being forced to reestablish themselves against this landscape. Rather than only treating ailments, future healthcare is predicted to focus on wellness and prevention. In personalized care, long-term prevention strategies, remote monitoring, early diagnosis, and detection are critical. Given the growing interest in smart healthcare defined by these modern technologies, this study looked into the definitions and service kinds of smart healthcare. The background and technical aspects of smart hospitals were also explored through a literature review.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.556-558
• /
• 2022

This paper proposes a CanSat system with a vehicle tracking function based on Jetson Nano, a high-performance small computer capable of operating artificial intelligence algorithms. The CanSat system consists of a CanSat and a ground station. The CanSat falls in the atmosphere and transmits the data obtained through the installed sensors to the ground station using wireless communication. The existing CanSat is limited to the mission of simply transmitting the collected information to the ground station, and there is a limit to efficiently performing the mission due to the limited fall time and bandwidth limitation of wireless communication. The Jetson Nano based CanSat proposed in this paper uses a pre-trained neural network model to detect the location of a vehicle in each image taken from the air in real time, and then uses a 2-axis motor to move the camera to track the vehicle.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.5
• /
• pp.843-847
• /
• 2023

In this paper, we propose a model for detecting efficient fire identification to prevent fires that can lead to various industrial accidents, farmland and large forest fires, with the widespread use of various chemicals and flammable substances as modern technology advances. This paper presents an algorithm that can detect fire smoke in a high-efficiency and short time using images, and an algorithm based on SVM(Support Vector Machine) and K fold cross-verification technologies. By analyzing images, fire and smoke detection algorithms have relatively superior detection performance compared to existing algorithms, and the analysis of fire and smoke characteristics detected in this paper is analyzed stably and efficiently and is expected to be used in various fields that may be exposed to fire risks in the future.