• Journal of Biomedical Engineering Research
• /
• v.45 no.3
• /
• pp.109-117
• /
• 2024

We developed a device that integrates digital otoscope and stethoscope for telemedicine. The integrated device was utilized for the collection of tympanic membrane images and cardiac auscultation data. Data accumulated on the platform server can support real-time diagnosis of heart and eardrum diseases using artificial intelligence. Public data from Kaggle were used for deep learning. After comparing with various deep learning models, the MobileNetV2 model showed superior performance in analyzing tympanic membrane data, and the VGG16 model excelled in analyzing cardiac data. The classification algorithm achieved an accuracy of 89.9% for eardrums data and 100% for heart sound data. These results demonstrate the possibility of diagnosing diseases without the limitations of time and space by using this platform.

• Convergence Security Journal
• /
• v.23 no.4
• /
• pp.61-70
• /
• 2023

Clustering is an unsupervised learning method that involves grouping data based on features such as distance metrics, using data without known labels or ground truth values. This method has the advantage of being applicable to various types of data, including images, text, and audio, without the need for labeling. Traditional clustering techniques involve applying dimensionality reduction methods or extracting specific features to perform clustering. However, with the advancement of deep learning models, research on deep clustering techniques using techniques such as autoencoders and generative adversarial networks, which represent input data as latent vectors, has emerged. In this study, we propose a deep clustering technique based on deep learning. In this approach, we use an autoencoder to transform the input data into latent vectors, and then construct a vector space according to the cluster structure and perform k-means clustering. We conducted experiments using the MNIST and Fashion-MNIST datasets in the PyTorch machine learning library as the experimental environment. The model used is a convolutional neural network-based autoencoder model. The experimental results show an accuracy of 89.42% for MNIST and 56.64% for Fashion-MNIST when k is set to 10.

• Smart Media Journal
• /
• v.13 no.7
• /
• pp.36-44
• /
• 2024

Identifying plant species and diseases is crucial for maintaining biodiversity and achieving optimal crop yields, making it a topic of significant practical importance. Recent studies have extended plant disease recognition from traditional closed-set scenarios to open-set environments, where the goal is to reject samples that do not belong to known categories. However, in open-world tasks, it is essential not only to define unknown samples as "unknown" but also to classify them further. This task assumes that images and labels of known categories are available and that samples of unknown categories can be accessed. The model classifies unknown samples by learning the prior knowledge of known categories. To the best of our knowledge, there is no existing research on this topic in plant-related recognition tasks. To address this gap, this paper utilizes knowledge distillation to model the category space relationships between known and unknown categories. Specifically, we identify similarities between different species or diseases. By leveraging a fine-tuned model on known categories, we generate pseudo-labels for unknown categories. Additionally, we enhance the baseline method's performance by using a larger pre-trained model, dino-v2. We evaluate the effectiveness of our method on the large plant specimen dataset Herbarium 19 and the disease dataset Plant Village. Notably, our method outperforms the baseline by 1% to 20% in terms of accuracy for novel category classification. We believe this study will contribute to the community.

• Journal of Platform Technology
• /
• v.11 no.6
• /
• pp.34-46
• /
• 2023

The design of non-pneumatic tires, which are created by filling the space between the wheel and the tread with elastomeric compounds or polygonal spokes, has become an important research topic in the automotive and aerospace industries. In this study, a system was designed for the design of non-pneumatic tires through the implementation of a generative adversarial network. We specifically examined factors that could impact the design, including the type of non-pneumatic tire, its intended usage environment, manufacturing techniques, distinctions from pneumatic tires, and how spoke design affects load distribution. Using OpenCV, various shapes and spoke configurations were generated as images, and a GAN model was trained on the projected GANs to generate shapes and spokes for non-pneumatic tire designs. The designed non-pneumatic tires were labeled as available or not, and a Vision Transformer image classification AI model was trained on these labels for classification purposes. Evaluation of the classification model show convergence to a near-zero loss and a 99% accuracy rate confirming the generation of non-pneumatic tire designs.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.19 no.1
• /
• pp.95-107
• /
• 2017

In this study, a preliminary study was undertaken for development of a tunnel incident automatic detection system based on a machine learning algorithm which is to detect a number of incidents taking place in tunnel in real time and also to be able to identify the type of incident. Two road sites where CCTVs are operating have been selected and a part of CCTV images are treated to produce sets of training data. The data sets are composed of position and time information of moving objects on CCTV screen which are extracted by initially detecting and tracking of incoming objects into CCTV screen by using a conventional image processing technique available in this study. And the data sets are matched with 6 categories of events such as lane change, stoping, etc which are also involved in the training data sets. The training data are learnt by a resilience neural network where two hidden layers are applied and 9 architectural models are set up for parametric studies, from which the architectural model, 300(first hidden layer)-150(second hidden layer) is found to be optimum in highest accuracy with respect to training data as well as testing data not used for training. From this study, it was shown that the highly variable and complex traffic and incident features could be well identified without any definition of feature regulation by using a concept of machine learning. In addition, detection capability and accuracy of the machine learning based system will be automatically enhanced as much as big data of CCTV images in tunnel becomes rich.

• Journal of Astronomy and Space Sciences
• /
• v.20 no.4
• /
• pp.283-298
• /
• 2003

The 2001 Leonid meteor storm has been observed all over the world, and its most intense flux since the last few decades has caused great interest among both laymen and experts. Especially, its maximum hours occurred at dawn hours of Nov. 19 in the east Asia, during which moonless clear night at the Mt. Bohyun allowed us near perfect condition of observation. Observation was carried out in the period of 01:00∼05:40(KST), which include the predicted maximum hours, with all-sky camera installed for upper atmospheric airglow research. Tn this paper we analyze 68 all-sky images obtained in this period, which contain records of 172 meteors. Utilizing the zenith hourly rate(ZHR) of 3000 and magnitude distribution index of 2, which were reported to International Meteor Organization by visible observers in the east Asia, we estimate the limiting magnitude of about 3 for meteors detected in our all-sky images. We then derive magnitudes of 83 meteors with clear pixel brightness outlines among the initially detected 172 meteors by comparing with neighbor standard stars. Angular velocities of meteors needed for computing their passing times over an all-sky image are expressed with a simple formula of an angle between a meteor head and the Leonid radiant point. The derived magnitudes of 83 meteors are in the range of -6∼-1 magnitude, and its distribution shows a maximum new -3mag. The derived magnitudes are much smaller than the limiting magnitude inferred from the comparison with the result of naked-eye observations. The difference may be due to the characteristic difference between nearly instantaneuous naked-eye observations and CCD observations with a long exposure. We redetermine magnitudes of the meteors by adjusting a meteor lasting time to be consistent with the naked-eye observations. The relative distribution of the redetermined magnitudes, which has a maximum at 0 mag., resembles that of the magnitudes determined with the in-principle method. The relative distribution is quite different from ones that decrease monotonically with decreasing magnitudes for meteors(1∼6) sensitive to naked-eye observations. We conclude from the magnitude distribution of our all-sky observation that meteors brighter than about 0 mag., appeared more frequently during the 2001 Leonid maximum hours. The frequent appearance of bright meteors has significantly important implication for meteor research. We noted, however, considerably large uncertainties in magnitudes determined only by comparing standard stars due to the unknown lasting time of meteors and the non-linear sensitivity of all-sky camera.

• Tunnel and Underground Space
• /
• v.31 no.5
• /
• pp.374-384
• /
• 2021

In this study, we propose a new approach for automatic fracture detection in CT scan images of rock specimens. This approach is built on top of two-stage object detection deep learning algorithm called Faster R-CNN with a major modification of using rotated bounding box. The use of rotated bounding box plays a key role in the future work to overcome several inherent difficulties of fracture segmentation relating to the heterogeneity of uninterested background (i.e., minerals) and the variation in size and shape of fracture. Comparing to the commonly used bounding box (i.e., axis-align bounding box), rotated bounding box shows a greater adaptability to fit with the elongated shape of fracture, such that minimizing the ratio of background within the bounding box. Besides, an additional benefit of rotated bounding box is that it can provide relative information on the orientation and length of fracture without the further segmentation and measurement step. To validate the applicability of the proposed approach, we train and test our approach with a number of CT image sets of fractured granite specimens with highly heterogeneous background and other rocks such as sandstone and shale. The result demonstrates that our approach can lead to the encouraging results on fracture detection with the mean average precision (mAP) up to 0.89 and also outperform the conventional approach in terms of background-to-object ratio within the bounding box.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.24 no.3
• /
• pp.247-262
• /
• 2022

In domestic tunnels, it is mandatory to install CCTVs in tunnels longer than 200 m which are also recommended by installation of a CCTV-based automatic accident detection system. In general, the CCTVs in the tunnel are installed at a low height as well as near by the moving vehicles due to the spatial limitation of tunnel structure, so a severe perspective effect takes place in the distance of installed CCTV and moving vehicles. Because of this effect, conventional CCTV-based accident detection systems in tunnel are known in general to be very hard to achieve the performance in detection of unexpected accidents such as stop or reversely moving vehicles, person on the road and fires, especially far from 100 m. Therefore, in this study, the region of interest is set up and a new concept of inverse perspective transformation technique is introduced. Since moving vehicles in the transformed image is enlarged proportionally to the distance from CCTV, it is possible to achieve consistency in object detection and identification of actual speed of moving vehicles in distance. To show this aspect, two datasets in the same conditions are composed with the original and the transformed images of CCTV in tunnel, respectively. A comparison of variation of appearance speed and size of moving vehicles in distance are made. Then, the performances of the object detection in distance are compared with respect to the both trained deep-learning models. As a result, the model case with the transformed images are able to achieve consistent performance in object and accident detections in distance even by 200 m.

• Cartoon and Animation Studies
• /
• s.46
• /
• pp.63-84
• /
• 2017

Overwhelming images of vividly colored aircraft flying across the blue sky and steam gushing from massive machines are reminiscent of Japanese animation films, especially of works by master director Hayao Miyazaki. By presenting together steam engines, which are mechanical devices of the Industrial Age in the past, and aircraft of the future age, the director constructs ambiguous space and time. These special time and space constitute nostalgia for past time, with devices called steam engines as a medium, and a longing for science and the future as represented by aircraft. In addition, the anticipation and disappointment, ideals and regrets of humans who see these two from the perspective of the present are projected on the works. This shares the characteristic of the steam punk genre, which seeks to return to the past rather than to face current problems. A subgenre of science fiction (henceforth "sci-fi"), steam punk reflects fundamental skepticism of science and technology and mechanized civilization, which have developed beyond human control. In addition, as works that clearly display such characteristics, director Miyazaki's and < $Nausica{\ddot{a}}$ of the Valley of Wind> can be examined. With spectacles of steam engines and aircraft, these two works enticingly visualize narratives about nature and humans and about the environment and destruction. Such attractiveness on the part of the master director's works has led to support from fans worldwide. However, often in the backgrounds of director Miyazaki's works, which have depicted ideal worlds of nature, environment, and community as highly concentrated fantasies, lie presuppositions of war and the end of the world. As works that are especially prominent in such characteristics, there are and . These two works betray the expectations of the audience by establishing the actual wartime as the temporal background and proceeding toward narratives of reality. Trapped in the ontological identity of the director himself, the war depicted by him projects a subjective and romantic attitude. Such a problem stems also from the ambiguity of the hybrid space and time, which is basic to the steam punk genre. This is because the basic characteristic of steam punk is to transplant past time, which humans were able to control, in the future from a perspective of optimism and longing via steam engines rather than to face current problems. In this respect, steam punk animation films in themselves can be seen as having significance and limitations at the same time.

• Radiation Oncology Journal
• /
• v.21 no.3
• /
• pp.238-244
• /
• 2003

Purpose: The Planning of High-Dose-Rate (HDR) brachytherapy treatments are becoming individualized and more dependent on the treatment planning system. Therefore, computer software has been developed to perform independent point dose calculations with the integration of an isodose distribution curve display into the patient anatomy images. Meterials and Methods: As primary input data, the program takes patients'planning data including the source dwell positions, dwell times and the doses at reference points, computed by an HDR treatment planning system (TPS). Dosimetric calculations were peformed in a $10\times12\times10\;Cm^3$ grid space using the Interstitial Collaborative Working Group (ICWG) formalism and an anisotropy table for the HDR Iridium-192 source. The computed doses at the reference points were automatically compared with the relevant results of the TPS. The MR and simulation film images were then imported and the isodose distributions on the axial, sagittal and coronal planes intersecting the point selected by a user were superimposed on the imported images and then displayed. The accuracy of the software was tested in three benchmark plans peformed by Gamma-Med 12i TPS (MDS Nordion, Germany). Nine patients'plans generated by Plato (Nucletron Corporation, The Netherlands) were verified by the developed software. Results: The absolute doses computed by the developed software agreed with the commercial TPS results within an accuracy of $2.8\%$ in the benchmark plans. The isodose distribution plots showed excellent agreements with the exception of the tip legion of the source's longitudinal axis where a slight deviation was observed. In clinical plans, the secondary dose calculations had, on average, about a $3.4\%$ deviation from the TPS plans. Conclusion: The accurate validation of complicate treatment plans is possible with the developed software and the qualify of the HDR treatment plan can be improved with the isodose display integrated into the patient anatomy information.