• Journal of Advanced Navigation Technology
• /
• v.27 no.3
• /
• pp.306-313
• /
• 2023

Realistic and graphics-based virtual reality content is based on 360-degree videos, and viewport extraction through the viewer's intention or automatic recommendation function is essential. This paper designs a viewport extraction system based on multiple object tracking in 360-degree videos and proposes a parallel computing structure necessary for multiple viewport extraction. The viewport extraction process in 360-degree videos is parallelized by composing pixel-wise threads, through 3D spherical surface coordinate transformation from ERP coordinates and 2D coordinate transformation of 3D spherical surface coordinates within the viewport. The proposed structure evaluated the computation time for up to 30 viewport extraction processes in aerial 360-degree video sequences and confirmed up to 5240 times acceleration compared to the CPU-based computation time proportional to the number of viewports. When using high-speed I/O or memory buffers that can reduce ERP frame I/O time, viewport extraction time can be further accelerated by 7.82 times. The proposed parallelized viewport extraction structure can be applied to simultaneous multi-access services for 360-degree videos or virtual reality contents and video summarization services for individual users.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.3
• /
• pp.199-206
• /
• 2023

This study utilizes deep learning algorithms to automatically classify construction waste into three categories: wood waste, plastic waste, and concrete waste. Two models, VGG-16 and ViT (Vision Transformer), which are convolutional neural network image classification algorithms and NLP-based models that sequence images, respectively, were compared for their performance in classifying construction waste. Image data for construction waste was collected by crawling images from search engines worldwide, and 3,000 images, with 1,000 images for each category, were obtained by excluding images that were difficult to distinguish with the naked eye or that were duplicated and would interfere with the experiment. In addition, to improve the accuracy of the models, data augmentation was performed during training with a total of 30,000 images. Despite the unstructured nature of the collected image data, the experimental results showed that VGG-16 achieved an accuracy of 91.5%, and ViT achieved an accuracy of 92.7%. This seems to suggest the possibility of practical application in actual construction waste data management work. If object detection techniques or semantic segmentation techniques are utilized based on this study, more precise classification will be possible even within a single image, resulting in more accurate waste classification

• Journal of Aerospace System Engineering
• /
• v.16 no.1
• /
• pp.49-55
• /
• 2022

The role of drones has been expanded to various fields such as agriculture, construction, and logistics. In particular, agriculture drones are emerging as an effective alternative to solve the problem of labor shortage and reduce the input cost. In this study therefore, we proposed the fruit recognition algorithm and harvesting mechanism for fruit harvesting drone system that can safely harvest fruits at high positions. In the fruit recognition algorithm, we employ "You-Only-Look-Once" which is a deep learning-based object detection algorithm and verify its feasibility by establishing a virtual simulation environment. In addition, we propose the fruit harvesting mechanism which can be operated by a single driving motor. The rotational motion of the motor is converted into a linear motion by the scotch yoke, and the opened gripper moves forward, grips a fruit and rotates it for harvesting. The feasibility of the proposed mechanism is verified by performing Multi-body dynamics analysis.

• Analytical Science and Technology
• /
• v.22 no.3
• /
• pp.201-209
• /
• 2009

Four heavy metals (Pb, Cd, As, and Hg) in 38 species (total 325 samples) of oriental animality medicines were monitored by inductively coupled plasma-mass spectrometry (ICP-MS) and automatic mercury analyzer (AMA). The detected concentration range of Pb, Cd, As was presented as $0.02{\mu}gkg^{-1}$ $(D.L){\sim}11.29mgkg^{-1}$, $0.01{\mu}gkg^{-1}$ $(D.L){\sim}2.50 mgkg^{-1}$, $0.12{\mu}gkg^{-1}$ $(D.L){\sim}5.27mgkg^{-1}$, respectively. In case of Hg, it the concentration range was $0.01{\sim}77.11mgkg^{-1}$ except one sample which exceeded detection limit. In all samples of Amydae Carapax and Gallnut, it was not detected over the maximum residue limits of metals. Pb is in charge of the greatest portion of contamination in 22 species of animality medicines, and in case of Hg, 54.46% of total samples were over the maximum residue limits. Therefore, environmental levels of Pb and Hg are needed to continue the researches and the studies for tracking pollution source are required.

• Journal of Internet Computing and Services
• /
• v.24 no.6
• /
• pp.91-98
• /
• 2023

Recently, according to development of artificial intelligence, a wide range of industry being automatic and optimized. Also we can find out some research of using supervised learning for deteceting defect of railway in domestic rail industry. However, there are structures other than rails on the track, and the fastener is a device that binds the rail to other structures, and periodic inspections are required to prevent safety accidents. In this paper, we present a method of reducing cost for labeling using semi-supervised and transfer model trained on rail fastener data. We use Resnet50 as the backbone network pretrained on ImageNet. At first we randomly take training data from unlabeled data and then labeled that data to train model. After predict unlabeled data by trained model, we adopted a method of adding the data with the highest probability for each class to the training data by a predetermined size. Futhermore, we also conducted some experiments to investigate the influence of the number of initially labeled data. As a result of the experiment, model reaches 92% accuracy which has a performance difference of around 5% compared to supervised learning. This is expected to improve the performance of the classifier by using relatively few labels without additional labeling processes through the proposed method.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_4
• /
• pp.1085-1095
• /
• 2023

Time-series data generated from satellite data are crucial resources for change detection and monitoring across various fields. Existing research in time-series data generation primarily relies on single-image analysis to maintain data uniformity, with ongoing efforts to enhance spatial and temporal resolutions by utilizing diverse image sources. Despite the emphasized significance of time-series data, there is a notable absence of automated data collection and preprocessing for research purposes. In this paper, to address this limitation, we propose a system that automates the collection of satellite information in user-specified areas to generate time-series data. This research aims to collect data from various satellite sources in a specific region and convert them into time-series data, developing an automatic satellite image collection system for this purpose. By utilizing this system, users can collect and extract data for their specific regions of interest, making the data immediately usable. Experimental results have shown the feasibility of automatically acquiring freely available Landsat and Sentinel images from the web and incorporating manually inputted high-resolution satellite images. Comparisons between automatically collected and edited images based on high-resolution satellite data demonstrated minimal discrepancies, with no significant errors in the generated output.

• Journal of Navigation and Port Research
• /
• v.47 no.6
• /
• pp.410-418
• /
• 2023

Recently, advancements and commercialization in the field of maritime autonomous surface ships (MASS) has rapidly progressed. Concurrently, studies are also underway to develop methods for automatically surveying the condition of various on-board equipment remotely to ensure the navigational safety of MASS. One key issue that has gained prominence is the method to obtain values from analog gauges installed in various equipment through image processing. This approach has the advantage of enabling the non-contact detection of gauge values without modifying or changing already installed or planned equipment, eliminating the need for type approval changes from shipping classifications. The objective of this study was to identify a dynamically changing indicator needle within noisy images of analog gauges. The needle object must be identified because its position significantly affects the accurate reading of gauge values. An analog pressure gauge attached to an emergency fire pump model was used for image capture to identify the needle object. The acquired images were pre-processed through Gaussian filtering, thresholding, and morphological operations. The needle object was then identified through Hough Transform. The experimental results confirmed that the center and object of the indicator needle could be identified in images of noisy analog gauges. The findings suggest that the image processing method applied in this study can be utilized for shape identification in analog gauges installed on ships. This study is expected to be applicable as an image processing method for the automatic remote survey of MASS.

• Imaging Science in Dentistry
• /
• v.54 no.1
• /
• pp.81-91
• /
• 2024

Purpose: The objective of this study was to propose a deep-learning model for the detection of the mandibular canal on dental panoramic radiographs. Materials and Methods: A total of 2,100 panoramic radiographs (PANs) were collected from 3 different machines: RAYSCAN Alpha (n=700, PAN A), OP-100 (n=700, PAN B), and CS8100 (n=700, PAN C). Initially, an oral and maxillofacial radiologist coarsely annotated the mandibular canals. For deep learning analysis, convolutional neural networks (CNNs) utilizing U-Net architecture were employed for automated canal segmentation. Seven independent networks were trained using training sets representing all possible combinations of the 3 groups. These networks were then assessed using a hold-out test dataset. Results: Among the 7 networks evaluated, the network trained with all 3 available groups achieved an average precision of 90.6%, a recall of 87.4%, and a Dice similarity coefficient (DSC) of 88.9%. The 3 networks trained using each of the 3 possible 2-group combinations also demonstrated reliable performance for mandibular canal segmentation, as follows: 1) PAN A and B exhibited a mean DSC of 87.9%, 2) PAN A and C displayed a mean DSC of 87.8%, and 3) PAN B and C demonstrated a mean DSC of 88.4%. Conclusion: This multi-device study indicated that the examined CNN-based deep learning approach can achieve excellent canal segmentation performance, with a DSC exceeding 88%. Furthermore, the study highlighted the importance of considering the characteristics of panoramic radiographs when developing a robust deep-learning network, rather than depending solely on the size of the dataset.

• Progress in Medical Physics
• /
• v.24 no.2
• /
• pp.85-91
• /
• 2013

At present, megavoltage computed tomography (MVCT) is the only method used to correct the position of tomotherapy patients. MVCT produces extra radiation, in addition to the radiation used for treatment, and repositioning also takes up much of the total treatment time. To address these issues, we suggest the use of a video image-guided setup (VIGS) system for correcting the position of tomotherapy patients. We developed an in-house program to correct the exact position of patients using two orthogonal images obtained from two video cameras installed at $90^{\circ}$ and fastened inside the tomotherapy gantry. The system is programmed to make automatic registration possible with the use of edge detection of the user-defined region of interest (ROI). A head-and-neck patient is then simulated using a humanoid phantom. After taking the computed tomography (CT) image, tomotherapy planning is performed. To mimic a clinical treatment course, we used an immobilization device to position the phantom on the tomotherapy couch and, using MVCT, corrected its position to match the one captured when the treatment was planned. Video images of the corrected position were used as reference images for the VIGS system. First, the position was repeatedly corrected 10 times using MVCT, and based on the saved reference video image, the patient position was then corrected 10 times using the VIGS method. Thereafter, the results of the two correction methods were compared. The results demonstrated that patient positioning using a video-imaging method ($41.7{\pm}11.2$ seconds) significantly reduces the overall time of the MVCT method ($420{\pm}6$ seconds) (p<0.05). However, there was no meaningful difference in accuracy between the two methods (x=0.11 mm, y=0.27 mm, z=0.58 mm, p>0.05). Because VIGS provides a more accurate result and reduces the required time, compared with the MVCT method, it is expected to manage the overall tomotherapy treatment process more efficiently.

• Economic and Environmental Geology
• /
• v.55 no.5
• /
• pp.551-561
• /
• 2022

Geophysical exploration methods are very useful for generating high-resolution images of underground structures, and such methods can be applied to investigation of buried cultural properties and for determining their exact locations. In this study, image feature extraction and image segmentation methods were applied to automatically distinguish the structures of buried relics from the high-resolution ground-penetrating radar (GPR) images obtained at the center of Silla Kingdom, Gyeongju, South Korea. The major purpose for image feature extraction analyses is identifying the circular features from building remains and the linear features from ancient roads and fences. Feature extraction is implemented by applying the Canny edge detection and Hough transform algorithms. We applied the Hough transforms to the edge image resulted from the Canny algorithm in order to determine the locations the target features. However, the Hough transform requires different parameter settings for each survey sector. As for image segmentation, we applied the connected element labeling algorithm and object-based image analysis using Orfeo Toolbox (OTB) in QGIS. The connected components labeled image shows the signals associated with the target buried relics are effectively connected and labeled. However, we often find multiple labels are assigned to a single structure on the given GPR data. Object-based image analysis was conducted by using a Large-Scale Mean-Shift (LSMS) image segmentation. In this analysis, a vector layer containing pixel values for each segmented polygon was estimated first and then used to build a train-validation dataset by assigning the polygons to one class associated with the buried relics and another class for the background field. With the Random Forest Classifier, we find that the polygons on the LSMS image segmentation layer can be successfully classified into the polygons of the buried relics and those of the background. Thus, we propose that these automatic classification methods applied to the GPR images of buried cultural heritage in this study can be useful to obtain consistent analyses results for planning excavation processes.