• Journal of Korea Multimedia Society
• /
• v.24 no.8
• /
• pp.1026-1034
• /
• 2021

Tire wear and defect are important factors for safe driving condition. These defects are generally inspected by some specialized experts or very expensive equipments such as stereo depth camera and depth gauge. In this paper, we propose tire safety vision inspector based on deep neural network (DNN). The status of tire wear is categorized into three: 'safety', 'warning', and 'danger' based on depth of tire tread. We propose an attention mechanism for emphasizing the feature of tread area. The attention-based feature is concatenated to output feature maps of the last convolution layer of ResNet-101 to extract more robust feature. Through experiments, the proposed tire wear classification model improves 1.8% of accuracy compared to the existing ResNet-101 model. For detecting the tire defections, the developed tire defect detection model shows up-to 91% of accuracy using the Mask R-CNN model. From these results, we can see that the suggested models are useful for checking on the safety condition of working tire in real environment.

• Journal of Information Technology Services
• /
• v.20 no.6
• /
• pp.97-105
• /
• 2021

Recently, with the development of IT technology, unmanned systems are being introduced in many industrial fields, and one of the most important factors for introducing unmanned systems in the automobile field is vehicle licence plate recognition(VLPR). The existing VLPR algorithms are configured to use image processing for a specific type of license plate to divide individual areas of a character within the plate to recognize each character. However, as the number of Korean vehicle license plates increases, the law is amended, there are old-fashioned license plates, new license plates, and different types of plates are used for each type of vehicle. Therefore, it is necessary to update the VLPR system every time, which incurs costs. In this paper, we use an object detection algorithm to detect character regardless of the format of the vehicle license plate, and apply a handwritten Hangul recognition(HHR) algorithm to enhance the recognition accuracy of a single Hangul character, which is called a Hangul unit. Since Hangul unit is recognized by combining initial consonant, medial vowel and final consonant, so it is possible to use other Hangul units in addition to the 40 Hangul units used for the Korean vehicle license plate.

• International journal of advanced smart convergence
• /
• v.11 no.2
• /
• pp.22-29
• /
• 2022

In this paper, for the purpose of designing an real-time unmanned monitoring system, the YOLOv5s (small) object detection model was applied on the NVIDIA TX2^TM AI (Artificial Intelligence) edge computing platform in order to design the fundamental function of an unmanned monitoring system that can detect objects in real time. YOLOv5s was applied to the our real-time unmanned monitoring system based on the performance evaluation of object detection algorithms (for example, R-CNN, SSD, RetinaNet, and YOLOv5). In addition, the performance of the four YOLOv5 models (small, medium, large, and xlarge) was compared and evaluated. Furthermore, based on these results, the YOLOv5s model suitable for the design purpose of this paper was ported to the NVIDIA TX2^TM AI edge computing system and it was confirmed that it operates normally. The real-time unmanned monitoring system designed as a result of the research can be applied to various application fields such as an security or monitoring system. Future research is to apply NMS (Non-Maximum Suppression) modification, model reconstruction, and parallel processing programming techniques using CUDA (Compute Unified Device Architecture) for the improvement of object detection speed and performance.

• The Journal of Korea Robotics Society
• /
• v.17 no.4
• /
• pp.417-424
• /
• 2022

In this paper, the developed trunk cargo unloading automation system is introduced, and the RGB-D sensor-based box loading situation recognition method and unloading plan applied to this system are suggested. First of all, it is necessary to recognize the position of the box in a truck. To do this, we first apply CNN-based YOLO, which can recognize objects in RGB images in real-time. Then, the normal vector of the center of the box is obtained using the depth image to reduce misrecognition in parts other than the box, and the inner wall of the truck in an image is removed. And a method of classifying the layers of the boxes according to the distance using the recognized depth information of the boxes is suggested. Given the coordinates of the boxes on the nearest layer, a method of generating the optimal path to take out the boxes the fastest using this information is introduced. In addition, kinematic analysis is performed to move the conveyor to the position of the box to be taken out of the truck, and kinematic analysis is also performed to control the robot arm that takes out the boxes. Finally, the effectiveness of the developed system and algorithm through a test bed is proved.

• Journal of the Semiconductor & Display Technology
• /
• v.22 no.1
• /
• pp.104-112
• /
• 2023

Agriculture plays an industrial and economic role, as well as an environmental and ecological conservation role, group harmony and the inheritance of traditional culture. However, no matter how advanced the industry is, the basic food necessary for human life can only be produced through the photosynthesis of plants with natural resources such as the sun, water, and air. The Food and Agriculture Organization of the United Nations (FAO) predicts that the world's population will increase by another 2 billion people by 2050, and it faces a myriad of complex and diverse factors to consider, including climate change, food security concerns, and global ecosystems and political factors. In particular, in order to solve problems such as increasing productivity and production of agricultural products, improving quality, and saving energy, it is difficult to solve them with traditional farming methods. Recently, with the wind of the 4th industrial revolution, ICT convergence technology and artificial intelligence have been rapidly developing in many fields, but it is also true that the application of new technologies is somewhat delayed due to the unique characteristics of agriculture. However, in recent years, as ICT and artificial intelligence utilization technologies have been developed and applied by many researchers, a revolution is also taking place in agriculture. This paper summarizes the current state of research so far in four categories of agriculture, namely crop cultivation environment management, soil management, pest management, and irrigation management, and smart farm research data that has recently been actively developed around the world.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.11a
• /
• pp.338-339
• /
• 2020

개인 방송의 보편화로 인해 인터넷 혹은 방송으로 유포되는 영상에서 일반인의 얼굴이 빈번히 노출되고 있으며, 동의 받지 않은 얼굴의 방송 노출은 개인 초상권 침해와 같은 사회적 문제를 일으킬 수 있다. 이러한 개인 초상권 침해 문제를 해결하고자 본 논문은 비디오에서 일반인의 얼굴을 검출하고 이에 마스킹을 가하는 방법을 제안한다. 제안 방법은 우선 딥러닝 기반의 Faster R-CNN을 이용하여 모자이킹을 하지 않을 특정인과 모자이킹을 가할 비특정인을 포함한 다수의 얼굴 영상을 학습한다. 학습된 네트워크를 이용하여 입력 비디오에 대해 사람의 얼굴을 검출하고 검출된 결과 중 특정인을 선별해 낸다. 최종적으로 입력 비디오에서 특정인을 제외한 나머지 검출된 얼굴에 대해 모자이킹 처리를 수행함으로써 비디오에서 지능적으로 비특정인의 얼굴을 가린다. 실험결과, 특정인과 비특정인을 포함한 얼굴 검출의 경우 99%의 정확도를 보였으며, 얼굴 검출 결과 중 특정인을 정확히 맞춘 경우는 86%의 정확도를 보였다. 제안 방법은 인터넷 동영상 서비스 및 방송 분야에서 개인 정보 보호를 위해 효과적으로 활용될 수 있을 것으로 기대된다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.273-275
• /
• 2021

본 논문에서는 기존 스마트주차관제 시스템의 한계로 인해 주차 관제의 사각지대에 있는 다세대 주택 주차 문제를 해결하는 솔루션을 제안한다. 기존 스마트 주차관제는 센서 기반의 고비용의 장비 및 시공비가 소요되며, 이러한 특성으로 인해 다세대 주택에 적용이 어렵다. 해당 문제를 해결하기 위해 본 논문은 기존 설비인 CCTV를 활용한 스마트 주차 관제 시스템을 제안하며, 해당 솔루션은 텐서플로 cnn중 알씨엔엔 RPN을 적용하여 차량 객체 인식 및 주차 공간 객체 인식을 구현하였으며, 다세대 주택 주변 CCTV 영상을 OpenCV를 활용하여 능동적이며 저비용의 스마트 주차 관제 방식을 구현하였으며 CCTV의 특성상 외곡되는 이미지를 OpenCV 이미지 변형을 통해 외곡 이미지를 복원하여 인식률을 높였다.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.26 no.3
• /
• pp.234-245
• /
• 2023

The Arena Fragmentation Test(AFT) is designed to analyze warhead performance by measuring fragmentation data. In order to evaluate the results of the AFT, a set of AFT images are captured by high-speed cameras. To detect objects in the AFT image set, ResNet-50 based Faster R-CNN is used as a detection model. However, because of the low resolution of the AFT image set, a detection model has shown low performance. To enhance the performance of the detection model, Super-resolution(SR) methods are used to increase the AFT image set resolution. To this end, The Bicubic method and three SR models: ZSSR, EDSR, and SwinIR are used. The use of SR images results in an increase in the performance of the detection model. While the increase in the number of pixels representing a fragment flame in the AFT images improves the Recall performance of the detection model, the number of pixels representing noise also increases, leading to a slight decreases in Precision performance. Consequently, the F1 score is increased by up to 9 %, demonstrating the effectiveness of SR in enhancing the performance of the detection model.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2023.05a
• /
• pp.40-41
• /
• 2023

This paper introduces the design of detection, recognition, and tracking algorithms for VTS radar image-based objects. The detection of objects in radar images utilizes artificial intelligence technology to determine the presence or absence of objects, and can classify the type of object using AI technology. Tracking involves the continuous tracking of detected objects over time, including technology to prevent confusion in the movement path. In particular, for land-based radar, there are unnecessary areas for detection depending on the terrain, so the function of detecting and recognizing vessels within the region of interest (ROI) set in the radar image is included. In addition, the extracted coordinate information is designed to enable various applications and interpretations by calculating speed, direction, etc.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.27 no.4
• /
• pp.311-323
• /
• 2023

This paper introduces the industrial problem, the solution, and the results of the research conducted with Define Inc. The client company wanted to improve the performance of an object detection model on the floor plan dataset. To solve the problem, we analyzed the operational principles, advantages, and disadvantages of the existing object detection model, identified the characteristics of the floor plan dataset, and proposed to use of YOLO v5 as an appropriate object detection model for training the dataset. We compared the performance of the existing model and the proposed model using mAP@60, and verified the object detection results with real test data, and found that the performance increase of mAP@60 was 0.08 higher with a 25% shorter inference time. We also found that the training time of the proposed YOLO v5 was 71% shorter than the existing model because it has a simpler structure. In this paper, we have shown that the object detection model for the floor plan dataset can achieve better performance while reducing the training time. We expect that it will be useful for solving other industrial problems related to object detection in the future. We also believe that this result can be extended to study object recognition in 3D floor plan dataset.