• Transactions of the Korean Society of Automotive Engineers
• /
• v.21 no.4
• /
• pp.33-42
• /
• 2013

This paper describes a multiple vehicle tracking algorithm using an integrated probabilistic data association filter (IPDAF) in urban environments. The algorithm consists of two parts; a pre-processing stage and an IPDA tracker. In the pre-processing stage, measurements are generated by a feature extraction method that manipulates raw data into predefined geometric features of vehicles as lines and boxes. After that, the measurements are divided into two different objects, dynamic and static objects, by using information of ego-vehicle motion. The IPDA tracker estimates not only states of tracks but also existence probability recursively. The existence probability greatly assists reliable initiation and termination of track in cluttered environment. The algorithm was validated by using experimental data which is collected in urban environment by using single laser scanner.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 1994.04a
• /
• pp.25-32
• /
• 1994

The pre-processor for finite to element structural analysis considering the user-friendly device is developed by using GUI. This can be used on WINDOWS' environment which is realized the multi-tasking and the concurrency by object-oriented paradigm. Data input can be done easily through menu, dialog box, automatic stepwise input and concurrent representation with the structural geometry on multiple windows. It in designed to control integratedly the pre-processing, execution and the post-processing of the finite element structural analysis program on multiple windows, and input data can be seen with result outputs at the same time. In addition, the object-oriented programming environment makes convenient revision and addition of the program components for expanding the scope of analysis and making better user environment.

• Journal of the Korean Society of Industry Convergence
• /
• v.2 no.2
• /
• pp.41-49
• /
• 1999

In general, FORTRAN is used for numerical analysis and OPS5 or LISP is used for expert systems, This causes problems at the interface because the various applications require different computing languages or environments. This paper describes the approach used to take AutoCAD as a user-interface for an existing finite element structural analysis package. Some principles concerning database management related to data-exchange interface of pre- and post-processing system for FORTRAN structural analysis program are discussed, and numerical examples demonstrate the power of the combination of these programs.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.11-16
• /
• 1997

This paper describes the grenade counting system developed for DPICM submunition analysis using the infrared video streams, and its some video stream processing technique. The video stream data processing procedure consists of four sequences; Analog infrared video stream recording, video stream capture, video stream pre-processing, and video stream analysis including the grenade counting. Some applications of this algorithms to real bursting test has shown the possibility of automation for submunition counting.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.2
• /
• pp.453-458
• /
• 2024

Optical Character Recognition is the technology that recognizes text in images and converts them into digital format. Deep learning-based OCR is being used in many industries with large quantities of recorded data due to its high recognition performance. To improve medical services, deep learning-based OCR was actively introduced by the medical industry. In this paper, we discussed trends in OCR engines and medical OCR and provided a roadmap for development of medical OCR. By using natural language processing on detected text data, current medical OCR has improved its recognition performance. However, there are limits to the recognition performance, especially for non-standard handwriting and modified text. To develop advanced medical OCR, databaseization of medical data, image pre-processing, and natural language processing are necessary.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2000.02a
• /
• pp.451-456
• /
• 2000

• Journal of Information Processing Systems
• /
• v.19 no.5
• /
• pp.673-687
• /
• 2023

Convolutional neural network-based deep learning technology is the most commonly used in image identification, but it requires large-scale data for training. Therefore, application in specific fields in which data acquisition is limited, such as in the military, may be challenging. In particular, the identification of ground weapon systems is a very important mission, and high identification accuracy is required. Accordingly, various studies have been conducted to achieve high performance using small-scale data. Among them, the ensemble method, which achieves excellent performance through the prediction average of the pre-trained models, is the most representative method; however, it requires considerable time and effort to find the optimal combination of ensemble models. In addition, there is a performance limitation in the prediction results obtained by using an ensemble method. Furthermore, it is difficult to obtain the ensemble effect using models with imbalanced classification accuracies. In this paper, we propose a transfer learning-based feature fusion technique for heterogeneous models that extracts and fuses features of pre-trained heterogeneous models and finally, fine-tunes hyperparameters of the fully connected layer to improve the classification accuracy. The experimental results of this study indicate that it is possible to overcome the limitations of the existing ensemble methods by improving the classification accuracy through feature fusion between heterogeneous models based on transfer learning.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2003.04a
• /
• pp.271-276
• /
• 2003

The Moderate Resolution Imaging Spectroradiometer (MODIS) on the Earth Observing System (EOS) of Terra and Aqua satellites, launched in December 1999 and May 2002, has been directly received by Korea Aerospace Research Institute (KARI) ground station facility from July 2002. MODIS scans a swath width of 2330 km that is sufficiently wide to cover Korean peninsular, Yellow and East Sea at once. The MODIS has 36 spectral bands between 0.415 $\mu\textrm{m}$ and 14.235 $\mu\textrm{m}$, i.e., through the visible into the thermal infrared. MODIS has been observed active fires, floods, smoke transport, dust storms, severe storms since February of 2000. The satellite imagery obtained through the MODIS will be utilized for many application such as national territorial management, agriculture, natural environment, atmosphere and ocean, etc. In this study is to introduce various application field of MODIS imagery and data processing system.

• Journal of Information Processing Systems
• /
• v.16 no.4
• /
• pp.795-808
• /
• 2020

Due to the poor evaluation indexes such as detection accuracy and recall rate when Yolov3 network detects aircraft in remote sensing images, in this paper, we propose a remote sensing image aircraft detection method based on machine vision. In order to improve the target detection effect, the Inception module was introduced into the Yolov3 network structure, and then the data set was cluster analyzed using the k-means algorithm. In order to obtain the best aircraft detection model, on the basis of our proposed method, we adjusted the network parameters in the pre-training model and improved the resolution of the input image. Finally, our method adopted multi-scale training model. In this paper, we used remote sensing aircraft dataset of RSOD-Dataset to do experiments, and finally proved that our method improved some evaluation indicators. The experiment of this paper proves that our method also has good detection and recognition ability in other ground objects.

• Journal of Korea Society of Industrial Information Systems
• /
• v.29 no.3
• /
• pp.79-87
• /
• 2024

Recently, the importance of pre-trained language models (PLM) has been emphasized for natural language processing (NLP) such as text classification, sentiment analysis, and question answering. Korean PLM shows high performance in NLP in general-purpose domains, but is weak in domains such as finance, medicine, and law. The main goal of this study is to propose a language model learning process and method to build a financial-specific language model that shows good performance not only in the financial domain but also in general-purpose domains. The five steps of the financial-specific language model are (1) financial data collection and preprocessing, (2) selection of model architecture such as PLM or foundation model, (3) domain data learning and instruction tuning, (4) model verification and evaluation, and (5) model deployment and utilization. Through this, a method for constructing pre-learning data that takes advantage of the characteristics of the financial domain and an efficient LLM training method, adaptive learning and instruction tuning techniques, were presented.