• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.2
• /
• pp.855-870
• /
• 2020

Windows are key components of building facades. Detecting windows, crucial to 3D semantic reconstruction and scene parsing, is a challenging task in computer vision. Early methods try to solve window detection by using hand-crafted features and traditional classifiers. However, these methods are unable to handle the diversity of window instances in real scenes and suffer from heavy computational costs. Recently, convolutional neural networks based object detection algorithms attract much attention due to their good performances. Unfortunately, directly training them for challenging window detection cannot achieve satisfying results. In this paper, we propose an approach for window detection. It involves an improved Faster R-CNN architecture for window detection, featuring in a window region proposal network, an RoI feature fusion and a context enhancement module. Besides, a post optimization process is designed by the regular distribution of windows to refine detection results obtained by the improved deep architecture. Furthermore, we present a newly collected dataset which is the largest one for window detection in real street scenes to date. Experimental results on both existing datasets and the new dataset show that the proposed method has outstanding performance.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.7
• /
• pp.19-26
• /
• 2018

This paper presents a structuring process of unstructured social network service (SNS) messages on rail services. We crawl messages about rail services posted on SNS and extract keywords indicating date and time, rail operating company, station name, direction, and rail service types from each message. Among them, the rail service types are classified by machine learning according to predefined rail service types, and the rest are extracted by regular expressions. Words are converted into vector representations using Word2Vec and a conventional Convolutional Neural Network (CNN) is used for training and classification. For performance measurement, our experimental results show a comparison with a TF-IDF and Support Vector Machine (SVM) approach. This structured information in the database and can be easily used for services for railway users.

• Journal of the Korean Society of Safety
• /
• v.34 no.6
• /
• pp.29-37
• /
• 2019

Since construction sites are exposed to outdoor environments, working conditions are significantly dangerous. Thus, wearing of the personal protective equipments such as safety helmet is very important for worker safety. However, construction workers are often wearing-off the helmet as inconvenient and uncomportable. As a result, a small mistake may lead to serious accident. For this, checking of wearing safety helmet is important task to safety managers in field. However, due to the limited time and manpower, the checking can not be executed for every individual worker spread over a large construction site. Therefore, if an automatic checking system is provided, field safety management should be performed more effectively and efficiently. In this study, applicability of deep learning based computer vision technology is investigated for automatic checking of wearing safety helmet in construction sites. Faster R-CNN deep learning algorithm for object detection and classification is employed to develop the automatic checking model. Digital camera images captured in real construction site are used to validate the proposed model. Based on the results, it is concluded that the proposed model may effectively be used for automatic checking of wearing safety helmet in construction site.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.05a
• /
• pp.447-448
• /
• 2018

라벨인식과 같은 광학 문자 인식은 영상처리를 활용한 컴퓨터 비전의 대표적인 연구분야이다. 본 연구에서는 딥러닝 기반의 라벨인식 시스템을 고안하였다, 생산 라인에 적용되는 라벨인식 시스템은 인식 속도가 중요하기 때문에 기존의 R-CNN기반의 딥러닝 신경망보다 월등히 빠른 오브젝트 검출 시스템 YOLO를 활용하여 문자를 학습 및 인식 시스템을 개발하였다. 본 시스템은 기존 시스템에 근접하는 문자인식 정확도를 제공하고 자동으로 문자영역을 검출 가능하며, 라벨의 인쇄불량을 판독하도록 하였다. 또한 개발, 배포, 적용이 한번에 가능한 프레임워크를 통하여 생산현장에서 발생하는 다양한 이미지 처리에 활용될 전망이다.

• Journal of the Korean Society of Industry Convergence
• /
• v.24 no.5
• /
• pp.547-553
• /
• 2021

Dynamic object recognition is an important task for autonomous vehicles. Since dynamic objects exhibit a higher collision risk than static objects, our own trajectories should be planned to match the future state of moving elements in the scene. Time information such as optical flow can be used to recognize movement. Existing optical flow calculations are based only on camera sensors and are prone to misunderstanding in low light conditions. In this regard, to improve recognition performance in low-light environments, we applied a normalization filter and a correction function for Gamma Value to the input images. The low light quality improvement algorithm can be applied to confirm the more accurate detection of Object's Bounding Box for the vehicle. It was confirmed that there is an important in object recognition through image prepocessing and deep learning using YOLO.

• Korean Journal of Remote Sensing
• /
• v.37 no.1
• /
• pp.111-122
• /
• 2021

Building footprint extraction is an active topic in the domain of remote sensing, since buildings are a fundamental unit of urban areas. Deep convolutional neural networks successfully perform footprint extraction from optical satellite images. However, semantic segmentation produces coarse results in the output, such as blurred and rounded boundaries, which are caused by the use of convolutional layers with large receptive fields and pooling layers. The objective of this study is to generate visually enhanced building objects by directly extracting the vertices of individual buildings by combining instance segmentation and keypoint detection. The target keypoints in building extraction are defined as points of interest based on the local image gradient direction, that is, the vertices of a building polygon. The proposed framework follows a two-stage, top-down approach that is divided into object detection and keypoint estimation. Keypoints between instances are distinguished by merging the rough segmentation masks and the local features of regions of interest. A building polygon is created by grouping the predicted keypoints through a simple geometric method. Our model achieved an F1-score of 0.650 with an mIoU of 62.6 for building footprint extraction using the OpenCitesAI dataset. The results demonstrated that the proposed framework using keypoint estimation exhibited better segmentation performance when compared with Mask R-CNN in terms of both qualitative and quantitative results.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.12
• /
• pp.4816-4834
• /
• 2020

This paper proposes transfer learning and fine-tuning techniques for a deep learning model to detect three distinct brain tumors from Magnetic Resonance Imaging (MRI) scans. In this work, the recent YOLOv4 model trained using a collection of 3064 T1-weighted Contrast-Enhanced (CE)-MRI scans that were pre-processed and labeled for the task. This work trained with the partial 29-layer YOLOv4-Tiny and fine-tuned to work optimally and run efficiently in most platforms with reliable performance. With the help of transfer learning, the model had initial leverage to train faster with pre-trained weights from the COCO dataset, generating a robust set of features required for brain tumor detection. The results yielded the highest mean average precision of 93.14%, a 90.34% precision, 88.58% recall, and 89.45% F1-Score outperforming other previous versions of the YOLO detection models and other studies that used bounding box detections for the same task like Faster R-CNN. As concluded, the YOLOv4-Tiny can work efficiently to detect brain tumors automatically at a rapid phase with the help of proper fine-tuning and transfer learning. This work contributes mainly to assist medical experts in the diagnostic process of brain tumors.

• 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.