• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.4
• /
• pp.58-63
• /
• 2022

Recently, the improvement of computational processing ability due to the rapid development of computing technology has greatly advanced the field of artificial intelligence, and research to apply it in various domains is active. In particular, in the national defense field, attention is paid to intelligent recognition among machine learning techniques, and efforts are being made to develop object identification and monitoring systems using artificial intelligence. To this end, various image processing technologies and object identification algorithms are applied to create a model that can identify friendly and enemy weapon systems and personnel in real-time. In this paper, we conducted image processing and object identification focused on tanks among various weapon systems. We initially conducted processing the tanks' image using a convolutional neural network, a deep learning technique. The feature map was examined and the important characteristics of the tanks crucial for learning were derived. Then, using YOLOv5 Network, a CNN-based object detection network, a model trained by labeling the entire tank and a model trained by labeling only the turret of the tank were created and the results were compared. The model and labeling technique we proposed in this paper can more accurately identify the type of tank and contribute to the intelligent recognition system to be developed in the future.

• Journal of the Korean Society of Industry Convergence
• /
• v.26 no.2_1
• /
• pp.217-223
• /
• 2023

With the rapid advancement of technologies, need for different research fields where this technology can be used is also increasing. One of the most researched topic in computer vision is object detection, which has widely been implemented in various fields which include healthcare, video surveillance and education. The main goal of object detection is to identify and categorize all the objects in a target environment. Specifically, methods of object detection consist of a variety of significant techniq ues, such as image processing and patterns recognition. Anomaly detection is a part of object detection, anomalies can be found various scenarios for example crowded places such as subway stations. An abnormal event can be assumed as a variation from the conventional scene. Since the abnormal event does not occur frequently, the distribution of normal and abnormal events is thoroughly imbalanced. In terms of public safety, abnormal events should be avoided and therefore immediate action need to be taken. When abnormal events occur in certain places, real time detection is required to prevent and protect the safety of the people. To solve the above problems, we propose a modified YOLOv5 object detection algorithm by implementing dilated convolutional layers which achieved 97% mAP50 compared to other five different models of YOLOv5. In addition to this, we also created a simple mobile application to avail the abnormal event detection on mobile phones.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.4
• /
• pp.959-979
• /
• 2024

Rice pest identification is essential in modern agriculture for the health of rice crops. As global rice consumption rises, yields and quality must be maintained. Various methodologies were employed to identify pests, encompassing sensor-based technologies, deep learning, and remote sensing models. Visual inspection by professionals and farmers remains essential, but integrating technology such as satellites, IoT-based sensors, and drones enhances efficiency and accuracy. A computer vision system processes images to detect pests automatically. It gives real-time data for proactive and targeted pest management. With this motive in mind, this research provides a novel farmland fertility algorithm with a deep learning-based automated rice pest detection and classification (FFADL-ARPDC) technique. The FFADL-ARPDC approach classifies rice pests from rice plant images. Before processing, FFADL-ARPDC removes noise and enhances contrast using bilateral filtering (BF). Additionally, rice crop images are processed using the NASNetLarge deep learning architecture to extract image features. The FFA is used for hyperparameter tweaking to optimise the model performance of the NASNetLarge, which aids in enhancing classification performance. Using an Elman recurrent neural network (ERNN), the model accurately categorises 14 types of pests. The FFADL-ARPDC approach is thoroughly evaluated using a benchmark dataset available in the public repository. With an accuracy of 97.58, the FFADL-ARPDC model exceeds existing pest detection methods.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.15 no.7
• /
• pp.4508-4515
• /
• 2014

Real-time body detection has been researched actively. On the other hand, the detection rate of color distorted images is low because most existing detection methods use static skin color model. Therefore, this paper proposes a new method for detecting the skin color region using a gradient map and saturation features. The basic procedure of the proposed method sequentially consists of creating a gradient map, extracting a gradient feature of skin regions, noise removal using the saturation features of skin, creating a cluster for extraction regions, detecting skin regions using cluster information, and verifying the results. This method uses features other than the color to strengthen skin detection not affected by light, race, age, individual features, etc. The results of the detection rate showed that the proposed method is 10% or more higher than the traditional methods.

• Journal of Korea Multimedia Society
• /
• v.23 no.8
• /
• pp.891-902
• /
• 2020

The Re-Identification(Re-ID) is one of the most popular researches in the field of computer vision due to a variety of applications. To achieve a high-level re-identification performance, recently other methods have developed the deep learning based networks that are specialized for only person or vehicle. However, most of the current methods are difficult to be used in real-world applications that require re-identification of both person and vehicle at the same time. To overcome this limitation, this paper proposes a deep neural network learning method that combines triplet and softmax loss to improve performance and re-identify people and vehicles simultaneously. It's possible to learn the detailed difference between the identities(IDs) by combining the softmax loss with the triplet loss. In addition, weights are devised to avoid bias in one-side loss when combining. We used Market-1501 and DukeMTMC-reID datasets, which are frequently used to evaluate person re-identification experiments. Moreover, the vehicle re-identification experiment was evaluated by using VeRi-776 and VehicleID datasets. Since the proposed method does not designed for a neural network specialized for a specific object, it can re-identify simultaneously both person and vehicle. To demonstrate this, an experiment was performed by using a person and vehicle re-identification dataset together.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2015.07a
• /
• pp.187-190
• /
• 2015

주행 중 차선 이탈 경고 시스템은 사고 발생 예방 차원에서 매우 높은 효과가 인정되어서 차선이탈 경고 장치(LDWS) 제품들이 출시되고 있다. 본 논문은 블랙박스의 영상을 이용하여 차선 검출에 정확도를 향상하기 위한 알고리즘을 연구한 것으로 특히 차량에 장착되어 있는 블랙박스 영상을 영상 변환 없이, 실시간 소프트웨어 만 으로 처리할 수 있는 알고리즘을 연구한다. 차선인식을 위한 최적의 영상 ROI를 결정하고, 차선 인식 정확도를 향상하기 위한 전 처리 과정을 적용하고, 동영상의 연속성을 잘못된 차선인식에 대한 보정, 인식이 되지 않는 차선에 대한 후보 차선 추천 알고리즘과 시점 변환에 의한 야간, 곡선 도로에 대한 오인식율을 최소화 하는 방법을 제안한다. 도로주행의 다양한 환경에 대한 실험을 진행했으며, 각각의 방법 적용에 의한 오인식율의 감소와 많은 인식 알고리즘 적용에 의한 처리 속도 저하를 개선하기 위한 연구를 진행했으며, 본 논문은 블랙박스 영상을 이용하여 주행 차선 인식을 위한 최적 알고리즘을 제안한다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.3
• /
• pp.143-149
• /
• 2018

Recently, due to the large-scale damage of natural disasters caused by global climate change, a monitoring system applying remote sensing technology is being constructed in disaster areas. Among remote sensing platforms, the drone has been actively used in the private sector due to recent technological developments, and has been applied in the disaster areas owing to advantages such as timeliness and economical efficiency. This paper deals with the development of a preprocessing system that can map the drone image data in a near-real time manner as a basis for constructing the disaster monitoring system using the drones. For the research purpose, our system is based on the SURF algorithm which is one of the computer vision technologies. This system aims to performs the desired correction through the feature point matching technique between reference images and shot images. The study area is selected as the lower part of the Gahwa River and the Daecheong dam basin. The former area has many characteristic points for matching whereas the latter area has a relatively low number of difference, so it is possible to effectively test whether the system can be applied in various environments. The results show that the accuracy of the geometric correction is 0.6m and 1.7m respectively, in both areas, and the processing time is about 30 seconds per 1 scene. This indicates that the applicability of this study may be high in disaster areas requiring timeliness. However, in case of no reference image or low-level accuracy, the results entail the limit of the decreased calibration.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.2
• /
• pp.692-712
• /
• 2022

Buoys used for Aid to Navigation systems are widely used to guide the sea paths and are powered by batteries, requiring continuous battery replacement. However, since human labor is required to replace the batteries, humans can be exposed to dangerous situation, including even collision with shipping vessels. In addition, Maritime sensors are installed on the route signs, so that these are often damaged by collisions with small and medium-sized ships, resulting in significant financial loss. In order to prevent these accidents, maritime object detection technology is essential to alert ships approaching buoys. Existing studies apply a number of filters to eliminate noise and to detect objects within the sea image. For this process, most studies directly access the pixels and process the images. However, this approach typically takes a long time to process because of its complexity and the requirements of significant amounts of computational power. In an emergent situation, it is important to alarm the vessel's rapid approach to buoys in real time to avoid collisions between vessels and route signs, therefore minimizing computation and speeding up processes are critical operations. Therefore, we propose Fast Connected Component Labeling (FCCL) which can reduce computation to minimize the processing time of filter applications, while maintaining the detection performance of existing methods. The results show that the detection performance of the FCCL is close to 30 FPS - approximately 2-5 times faster, when compared to the existing methods - while the average throughput is the same as existing methods.

• Journal of Korea Water Resources Association
• /
• v.49 no.6
• /
• pp.469-480
• /
• 2016

The present study aims to develop a real-time surface image velocimeter (SIV) using an Android smartphone. It can measure river surface velocity by using its built-in sensors and processors. At first the SIV system figures out the location of the site using the GPS of the phone. It also measures the angles (pitch and roll) of the device by using its orientation sensors to determine the coordinate transform from the real world coordinates to image coordinates. The only parameter to be entered is the height of the phone from the water surface. After setting, the camera of the phone takes a series of images. With the help of OpenCV, and open source computer vision library, we split the frames of the video and analyzed the image frames to get the water surface velocity field. The image processing algorithm, similar to the traditional STIV (Spatio-Temporal Image Velocimeter), was based on a correlation analysis of spatio-temporal images. The SIV system can measure instantaneous velocity field (1 second averaged velocity field) once every 11 seconds. Averaging this instantaneous velocity measurement for sufficient amount of time, we can get an average velocity field. A series of tests performed in an experimental flume showed that the measurement system developed was greatly effective and convenient. The measured results by the system showed a maximum error of 13.9 % and average error less than 10 %, when we compared with the measurements by a traditional propeller velocimeter.

• Korean Journal of Applied Biomechanics
• /
• v.24 no.2
• /
• pp.181-187
• /
• 2014

In this study, a real-time evaluation system for quantitative dynamic fitting during pedaling was developed. The system is consisted of LED markers, a digital camera connected to a computer and a marker detecting program. LED markers are attached to hip, knee, ankle joint and fifth metatarsal in the sagittal plane. Playstation3 eye which is selected as a main digital camera in this paper has many merits for using motion capture, such as high FPS (Frame per second) about 180FPS, $320{\times}240$ resolution, and low-cost with easy to use. The maker detecting program was made by using Labview2010 with Vision builder. The program was made up of three parts, image acquisition & processing, marker detection & joint angle calculation, and output section. The digital camera's image was acquired in 95FPS, and the program was set-up to measure the lower-joint angle in real-time, providing the user as a graph, and allowing to save it as a test file. The system was verified by pedalling at three saddle heights (knee angle: 25, 35, $45^{\circ}$) and three cadences (30, 60, 90 rpm) at each saddle heights by using Holmes method, a method of measuring lower limbs angle, to determine the saddle height. The result has shown low average error and strong correlation of the system, respectively, $1.18{\pm}0.44^{\circ}$, $0.99{\pm}0.01^{\circ}$. There was little error due to the changes in the saddle height but absolute error occurred by cadence. Considering the average error is approximately $1^{\circ}$, it is a suitable system for quantitative dynamic fitting evaluation. It is necessary to decrease error by using two digital camera with frontal and sagittal plane in future study.