• Journal of Broadcast Engineering
• /
• v.5 no.2
• /
• pp.281-289
• /
• 2000

While acquiring images, the relative motion between the imaging device and the object scene seriously damages the image quality. This phenomenon is called motion blur. The peak-trace approach, which is our recent previous work, identifies important parameters to characterize the point spread function (PSF) of the blur, given only the blurred image itself. With the peak-trace approach the direction of the motion blur can be extracted regardless of the noise corruption and does not need much Processing time. In this paper stochastic peak-trace approaches are introduced. The erroneous data can be selected through the ML classification, and can be made small through weighting. Therefore the distortion of the direction in the low frequency region can be prevented. Using the linear prediction method, the irregular data are prohibited from being selected as the peak point. The detection of the second peak using the proposed moving average least mean (MALM) method is used in the Identification of the motion extent. The MALM method itself includes a noise removal process, so it is possible to extract the parameters even an environment of heavy noise. In the experiment, we could efficiently restore the degraded image using the information obtained by the proposed algorithm.

• Smart Media Journal
• /
• v.11 no.11
• /
• pp.25-31
• /
• 2022

Deep learning-based image processing is essential for autonomous vehicles. To process road images in real-time in a System-on-Chip (SoC) environment, we need to execute deep learning models on a NPU (Neural Procesing Units) specialized for deep learning operations. In this study, we imported seven open-source image processing deep learning models, that were developed on GPU servers, to Texas Instrument Deep Learning (TIDL) NPU environment. We confirmed that the models imported in this study operate normally in the SoC virtual environment through performance evaluation and visualization. This paper introduces the problems that occurred during the migration process due to the limitations of NPU environment and how to solve them, and thereby, presents a reference case worth referring to for developers and researchers who want to port deep learning models to SoC environments.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.236-238
• /
• 2022

According to the current Road Traffic Act, the 2020 amendment bill is currently in effect as a system that designates vehicle types for each lane for the purpose of securing road use efficiency and traffic safety. When comparing the number of traffic accident fatalities per 10,000 vehicles in Germany and Korea, the number of traffic accident deaths in Germany is significantly lower than in Korea. The representative case of the German autobahn, which did not impose a speed limit, suggests that Korea's speeding laws are not the only answer to reducing the accident rate. The designated lane system, which is observed in accordance with the keep right principle of the Autobahn Expressway, plays a major role in reducing traffic accidents. Based on this fact, we propose a traffic enforcement system to crack down on vehicles violating the designated lane system and improve the compliance rate. We develop a designated lane enforcement system that recognizes vehicle types using Yolo5, a deep learning object recognition model, recognizes license plates and lanes using OpenCV, and stores the extracted data in the server to determine whether or not laws are violated.Accordingly, it is expected that there will be an effect of reducing the traffic accident rate through the improvement of driver's awareness and compliance rate.

• Journal of Internet Computing and Services
• /
• v.22 no.2
• /
• pp.59-68
• /
• 2021

As artificial intelligence(AI) technologies, which have made rapid growth recently, began to be applied to the marine environment such as ships, there have been active researches on the application of CNN-based models specialized for digital videos. In E-Navigation service, which is combined with various technologies to detect floating objects of clash risk to reduce human errors and prevent fires inside ships, real-time processing is of huge importance. More functions added, however, mean a need for high-performance processes, which raises prices and poses a cost burden on shipowners. This study thus set out to propose a method capable of processing information at a high rate while maintaining the accuracy by applying Quantization techniques of a deep learning model. First, videos were pre-processed fit for the detection of floating matters in the sea to ensure the efficient transmission of video data to the deep learning entry. Secondly, the quantization technique, one of lightweight techniques for a deep learning model, was applied to reduce the usage rate of memory and increase the processing speed. Finally, the proposed deep learning model to which video pre-processing and quantization were applied was applied to various embedded boards to measure its accuracy and processing speed and test its performance. The proposed method was able to reduce the usage of memory capacity four times and improve the processing speed about four to five times while maintaining the old accuracy of recognition.

• Journal of the Korea Convergence Society
• /
• v.11 no.11
• /
• pp.9-17
• /
• 2020

In this paper, we proposed an SNS review analysis method based on deep learning for user tendency. The existing SNS review analysis method has a problem that does not reflect a variety of opinions on various interests because most are processed based on the highest weight. To solve this problem, the proposed method is to extract the user's personal tendency from the SNS review for food. It performs classification using the YOLOv3 model, and after performing a sentiment analysis through the BiLSTM model, it extracts various personal tendencies through a set algorithm. Experiments showed that the performance of Top-1 accuracy 88.61% and Top-5 90.13% for the YOLOv3 model, and 90.99% accuracy for the BiLSTM model. Also, it was shown that diversity of the individual tendencies in the SNS review classification through the heat map. In the future, it is expected to extract personal tendencies from various fields and be used for customized service or marketing.

• The Journal of Bigdata
• /
• v.8 no.1
• /
• pp.49-59
• /
• 2023

There are various items in the safety and health standards of the manufacturing industry, but they can be divided into work-related diseases and musculoskeletal diseases according to the standards for sickness and accident victims. Musculoskeletal diseases occur frequently in manufacturing and can lead to a decrease in labor productivity and a weakening of competitiveness in manufacturing. In this paper, to detect the musculoskeletal harmful factors of manufacturing workers, we defined the musculoskeletal load work factor analysis, harmful load working postures, and key points matching, and constructed data for Artificial Intelligence(AI) learning. To check the effectiveness of the suggested dataset, AI algorithms such as YOLO, Lite-HRNet, and EfficientNet were used to train and verify. Our experimental results the human detection accuracy is 99%, the key points matching accuracy of the detected person is @AP0.5 88%, and the accuracy of working postures evaluation by integrating the inferred matching positions is LEGS 72.2%, NECT 85.7%, TRUNK 81.9%, UPPERARM 79.8%, and LOWERARM 92.7%, and considered the necessity for research that can prevent deep learning-based musculoskeletal diseases.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.30 no.1
• /
• pp.63-70
• /
• 2017

In this paper, a magnetic flux leakage(MFL) based steel bar damage detection was first researched to quantify the signals from damages on the wire rope. Though many researches inspecting damages using a MFL method was proceeded until the present, the researches are at the level that diagnose whether damages are or not. This has limitation to take measures in accordance with the damage level. Thus, a MFL inspection system was modeled using a finite element analysis(FEM) program dealing with electromagnetism problems, and a steel bar specimen was adopted as a ferromagnetic object. Then, an experimental study was also carried out to verify the simulation results with a steel bar which has same damage conditions as the simulation. The MFL signals was nearly not affected by the increase of the inspection velocity, and the magnitudes of the signals are not identical according to the change of the defect width even the defects have same depth. On the basis of the analysis, the signal properties from the damages were extracted to classify the type of damages, and it could be confirmed that classification of damages using extracted signal properties is feasible.

• 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.

• Journal of Advanced Navigation Technology
• /
• v.25 no.4
• /
• pp.305-312
• /
• 2021

Even though various types of sensors are being used for security applications, radar sensors are being suggested as an alternative due to the privacy issues. Among those radar sensors, PD radar has high-complexity receiver, but, FMCW radar requires fewer resources. However, FMCW has disadvantage from the use of 2D-FFT which increases the complexity, and it is difficult to distinguish people from objects those are stationary. In this paper, we present the design and the implementation results of the radar signal processor (RSP) that can distinguish between people and object by respiration measurement using phase estimation without 2D-FFT. The proposed RSP is designed with Verilog-HDL and is implemented on FPGA device. It was confirmed that the proposed RSP includes 6,425 LUT, 4,243 register, and 12,288 memory bits with 92.1% accuracy for target's breathing status.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.40 no.2
• /
• pp.91-108
• /
• 2022

In most cases, optical images have been used as training data of DL (Deep Learning) models for object detection, recognition, identification, classification, semantic segmentation, and instance segmentation. However, properties of 3D objects in the real-world could not be fully explored with 2D images. One of the major sources of the 3D geospatial information is DSM (Digital Surface Model). In this matter, characteristic information derived from DSM would be effective to analyze 3D terrain features. Especially, man-made objects such as buildings having geometrically unique shape could be described by geometric elements that are obtained from 3D geospatial data. The background and motivation of this paper were drawn from concept of the intrinsic image that is involved in high-level visual information processing. This paper aims to extract buildings after classifying terrain features by training DL model with DSM-derived information including slope, aspect, and SRI (Shaded Relief Image). The experiments were carried out using DSM and label dataset provided by ISPRS (International Society for Photogrammetry and Remote Sensing) for CNN-based SegNet model. In particular, experiments focus on combining multi-source information to improve training performance and synergistic effect of the DL model. The results demonstrate that buildings were effectively classified and extracted by the proposed approach.