• Proceedings of the KIEE Conference
• /
• 1995.11a
• /
• pp.225-227
• /
• 1995

An object detection algorithm using a modified IDM(Image Differential Method) is proposed for detecting an object in a level crossing area. The conventional object detection method using LASER light has the deadzone that it cannot detect small objects, while the object detection method using image data in a level crossing area can detect such small objects. But the image data in a level crossing area can be changeable easily because the data is outdoor and sensitive to such surrounding environments as the change of the sun beam, the shadow of cars, and so on. So we resolve these problems by adding the normalization and the process for shadow of the image data in a level crossing area to the basic IDM(Image Differential Method).

• Proceedings of the KIEE Conference
• /
• 2001.11c
• /
• pp.169-171
• /
• 2001

Unlike rigid objects, the edge intensity of a gaseous object is various along the object boundary (edge intensities of some pixels on a gaseous object boundary are weaker than those of small rigid objects or noise itself). Therefore, the conventional edge detectors may not adequately detect boundary-like edge pixels for gaseous objects. In this paper A new methodology for segmenting gaseous object images is introduced. Proposed method consists of fuzzy-based boundary detector applicable to gaseous as well as rigid objects and concave region filling to recover object regions.

• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.3
• /
• pp.1-7
• /
• 2022

This paper aims to compare the mean object size of M/G/1/PS model with that of M/BP/1 model used in the web service. The mean object size is one of important measure to control and manage web service economically. M/G/1/PS model utilizes the processor sharing in which CPU rotates in round-robin order giving time quantum to multiple tasks. M/BP/1 model uses the Bounded Pareto distribution to describe the web service according to file size. We may infer that the mean waiting latencies of M/G/1/PS and M/BP/1 model are equal to the mean waiting latency of the deterministic model using the round robin scheduling with the time quantum. Based on the inference, we can find the mean object size of M/G/1/PS model and M/BP/1 model, respectively. Numerical experiments show that when the system load is smaller than the medium, the mean object sizes of the M/G/1/PS model and the M/BP/1 model become the same. In particular, when the shaping parameter is 1.5 and the lower and upper bound of the file size is small in the M/BP/1 model, the mean object sizes of M/G/1/PS model and M/BP/1 model are the same. These results confirm that it is beneficial to use a small file size in a web service.

• Journal of the Korean Institute of Navigation
• /
• v.23 no.1
• /
• pp.29-43
• /
• 1999

The production process of the compound-curved hull plates includes hull design, definition, fairing, modeling, lofting, cutting, and forming in sequence. Traditional fabrication methods and shop environment caused low level to productivity in medium and small sized shipyards. The most effective solution to solve those problems is to rationalize the layout of facilities. For the well-balanced development of domestic shipbuilding industry, it is urgently required to reduce the gap between modernized large sized shipyards and traditional small and medium sized shipyards in production technologies and efficiencies. For the efficient and accurate hull piece forming, all information from design to forming should be clarified and organized in a systematic manner. Thus, management of the information plays an important role in the computerized and automated of hull piece forming. The object of this paper is to survey the status of the field, to find out the feasibility and to introduce a draft of hull piece fabrication line for small and medium sized shipyards. The development of required system follows the object oriented technology to extend to simulation based system for carrying out physical product flow and facilities layout analysis. It is feasible to operate such a modernized facility for a group of small and medium shipyards who are unable to have each of their own facility because of its large amount of initial investment and insufficient work load.

• The Journal of Korea Robotics Society
• /
• v.17 no.4
• /
• pp.425-430
• /
• 2022

As robots are considered one of the mainstream digital transformations, robots with machine vision becomes a main area of study providing the ability to check what robots watch and make decisions based on it. However, it is difficult to find a small object in the image mainly due to the flaw of the most of visual recognition networks. Because visual recognition networks are mostly convolution neural network which usually consider local features. So, we make a model considering not only local feature, but also global feature. In this paper, we propose a detection method of a small marker on the object using deep learning and an algorithm that considers global features by combining Transformer's self-attention technique with a convolutional neural network. We suggest a self-attention model with new definition of Query, Key and Value for model to learn global feature and simplified equation by getting rid of position vector and classification token which cause the model to be heavy and slow. Finally, we show that our model achieves higher mAP than state of the art model YOLOr.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.8
• /
• pp.1976-1995
• /
• 2023

The anchor-free object detector CenterNet regards the object as a center point and predicts it based on the Gaussian circle region. For each object's center point, CenterNet directly regresses the width and height of the objects and finally gets the boundary range of the objects. However, the critical range of the object's center point can not be accurately limited by using the Gaussian circle region to constrain the prediction region, resulting in many low-quality centers' predicted values. In addition, because of the large difference between the width and height of different objects, directly regressing the width and height will make the model difficult to converge and lose the intrinsic relationship between them, thereby reducing the stability and consistency of accuracy. For these problems, we proposed a center point prediction method based on the Gaussian elliptic region and a size component regression method based on the small solution space. First, we constructed a Gaussian ellipse region that can accurately predict the object's center point. Second, we recode the width and height of the objects, which significantly reduces the regression solution space and improves the convergence speed of the model. Finally, we jointly decode the predicted components, enhancing the internal relationship between the size components and improving the accuracy consistency. Experiments show that when using CenterNet as the improved baseline and Hourglass-104 as the backbone, on the MS COCO dataset, our improved model achieved 44.7%, which is 2.6% higher than the baseline.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_4
• /
• pp.1925-1934
• /
• 2022

Monitoring nuclear activity for inaccessible areas using remote sensing technology is essential for nuclear non-proliferation. In recent years, deep learning has been actively used to detect nuclear-activity-related small objects. However, high-resolution satellite imagery containing small objects can result in class imbalance. As a result, there is a performance degradation problem in detecting small objects. Therefore, this study aims to improve detection accuracy by analyzing the effect of the ratio of small objects related to nuclear activity in the input data for the performance of the deep learning model. To this end, six case datasets with different ratios of small object pixels were generated and a U-Net model was trained for each case. Following that, each trained model was evaluated quantitatively and qualitatively using a test dataset containing various types of small object classes. The results of this study confirm that when the ratio of object pixels in the input image is adjusted, small objects related to nuclear activity can be detected efficiently. This study suggests that the performance of deep learning can be improved by adjusting the object pixel ratio of input data in the training dataset.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2016.10a
• /
• pp.506-509
• /
• 2016

We present real-time pedestrian detection that exploit accuracy of Faster R-CNN network. Faster R-CNN has shown to success at PASCAL VOC multi-object detection tasks, and their ability to operate on raw pixel input without the need to design special features is very engaging. Therefore, in this work we apply and adjust Faster R-CNN to single object detection, which is pedestrian detection. The drawback of Faster R-CNN is its failure when object size is small. Previously, small sized object problem was solved by Scale-aware Network. We incorporate Scale-aware Network to Faster R-CNN. This made our method Scale-aware Faster R-CNN (DF R-CNN) that is both fast and very accurate. We separated Faster R-CNN networks into two sub-network, that is one for large-size objects and another one for small-size objects. The resulting approach achieves a 28.3% average miss rate on the Caltech Pedestrian detection benchmark, which is competitive with the other best reported results.

• Journal of Internet of Things and Convergence
• /
• v.9 no.1
• /
• pp.97-104
• /
• 2023

This paper is related to object segmentation using ESRGAN(Enhanced Super Resolution GAN) and SSS(Semantic Soft Segmentation). The segmentation performance of the object segmentation method using Mask R-CNN and SSS proposed by the research team in this paper is generally good, but the segmentation performance is poor when the size of the objects is relatively small. This paper is to solve these problems. The proposed method aims to improve segmentation performance of small objects by performing super-resolution through ESRGAN and then performing SSS when the size of an object detected through Mask R-CNN is below a certain threshold. According to the proposed method, it was confirmed that the segmentation characteristics of small-sized objects can be improved more effectively than the previous method.

• The Journal of Korea Robotics Society
• /
• v.17 no.2
• /
• pp.124-132
• /
• 2022

With growth in intelligence of mobile robots, interaction with humans is emerging as a very important issue for mobile robots and the pedestrian tracking technique following the designated person is adopted in many cases in a way that interacts with humans. Among the existing multi-object tracking techniques for pedestrian tracking, Simple Online and Realtime Tracking (SORT) is suitable for small mobile robots that require real-time processing while having limited computational performance. However, SORT fails to reflect changes in object detection values caused by the movement of the mobile robot, resulting in poor tracking performance. In order to solve this performance degradation, this paper proposes a more stable pedestrian tracking algorithm by correcting object tracking errors caused by robot movement in real time using wheel odometry information of a mobile robot and dynamically managing the survival period of the tracker that tracks the object. In addition, the experimental results show that the proposed methodology using data collected from actual mobile robots maintains real-time and has improved tracking accuracy with resistance to the movement of the mobile robot.