• Electrical & Electronic Materials
• /
• v.10 no.9
• /
• pp.916-921
• /
• 1997

In this study Characteristics of 3-D underwater object recognition independent of translation using the self-made ultrasonic sensor fabricated with porous piezoelectric resonator and presented. The sensor was satisfied with requirement of ultrasonic sensor. The recognition rates for the training data and the testing data are 97.45 and 91.25[%] respectively using the self-made ultrasonic sensor and SCL(Simple Competitive Learning) neural network. According to the experimental results It is believed that the self-made ultrasonic sensor can be applied as sensor of SONAR system.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 1996.11a
• /
• pp.316-319
• /
• 1996

In this study, characteristics of ultrasonic transducer fabricated with porous piezoelectric resonator are investigated, 3-D underwater object recognition using the self-made ultrasonic transducer and SOFM(Self-Organizing Feature Map) neural network are presented. The self-made transducer was satisfied the required condition of ultrasonic transducer in water, and the recognition rates for the training data and the testing data were 100 and 95.3% respectively. The experimental results have shown that the ultrasonic transducer fabricated with porous piezoelectric resonator could be applied for sonar system.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.67-70
• /
• 2022

NLP 분야에서 적극 활용되기 시작한 Transformer 모델을 Vision 분야에서 적용하기 시작하면서 object detection과 segmentation 등 각종 분야에서 기존 CNN 기반 모델의 정체된 성능을 극복하며 향상되고 있다. 또한, label 데이터 없이 이미지들로만 자기지도학습을 한 ViT(Vision Transformer) 모델을 통해 이미지에 포함된 여러 중요한 객체의 영역을 검출하는 saliency map을 추출할 수 있게 되었으며, 이로 인해 ViT의 자기지도학습을 통한 object detection과 semantic segmentation 연구가 활발히 진행되고 있다. 본 논문에서는 ViT 모델 뒤에 classifier를 붙인 모델에 일반 학습한 모델과 자기지도학습의 pretrained weight을 사용해서 전이학습한 모델의 시각화를 통해 각 saliency map들을 비교 분석하였다. 이를 통해, 클래스 분류 학습 기반 전이학습이 transformer의 saliency map에 미치는 영향을 확인할 수 있었다.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10b
• /
• pp.388-393
• /
• 1992

An infrared range finder using a self-mixing laser diode (SM-LD), which has been proposed and developed by the Authors, can measure not only a range of a moving target but its velocity simultaneously. In this paper, described is that the precise mode-hop pulse train can be obtained by employing a new signal processing circuit even when the backscattered light returning into the SM-LD is much more weaker. As a result, the distance to a tilted square sheet made from aluminium or white paper, which is placed 10 cm through 60 cm from the SM-LD, is measured with accuracy of a few percent even when the tilting angle is less than 75 degrees or 85 degrees, respectively. And in this paper, described is the range-image recognition of a plane object under the condition of standstill. The output laser beam is scanned by scanning two plane mirrors-equipped with each stepping motor. And we succeeded in the acquisition of the range-image of a plane object in a few tens of seconds. Furthermore, described is a feasibility study about the range-image recognition of a slowly moving plane object.

• The Journal of Korea Robotics Society
• /
• v.18 no.3
• /
• pp.346-351
• /
• 2023

With the release of numerous open driving datasets, the demand for domain adaptation in perception tasks has increased, particularly when transferring knowledge from rich datasets to novel domains. However, it is difficult to solve the change 1) in the sensor domain caused by heterogeneous LiDAR sensors and 2) in the environmental domain caused by different environmental factors. We overcome domain differences in the semi-supervised setting with 3-stage model parameter training. First, we pre-train the model with the source dataset with object scaling based on statistics of the object size. Then we fine-tine the partially frozen model weights with copy-and-paste augmentation. The 3D points in the box labels are copied from one scene and pasted to the other scenes. Finally, we use the knowledge distillation method to update the student network with a moving average from the teacher network along with a self-training method with pseudo labels. Test-Time Augmentation with varying z values is employed to predict the final results. Our method achieved 3rd place in ECCV 2022 workshop on the 3D Perception for Autonomous Driving challenge.

• Journal of Convergence for Information Technology
• /
• v.12 no.2
• /
• pp.1-16
• /
• 2022

Social network service (SNS) using mobile or web is growing rapidly, and the emergence of various platform services is causing innovative changes in social network service (SNS). This study is to identify the target relation factors of social network users and to empirically study the causal relationship of how much these factors affect user satisfaction through pleasure and self-efficacy. To present an effective and efficient development plan in. In order to empirically verify the research model of this study, a survey was conducted with the general public who had experience using social network services (SNS). Path analysis was performed. As a result, it was possible to verify the correlation of the object relational factors on user satisfaction through pleasure and self-efficacy.First, non-excluded had a significant effect on pleasure, but did not significantly affect self-efficacy. Second, stability attachment did not significantly affect both enjoyment and self-efficacy. Third, social ability did not significantly affect both enjoyment and self-efficacy. Fourth, self-centeredness did not have a significant effect on both enjoyment and self-efficacy. Fifth, pleasure had a significant effect on both self-efficacy and user satisfaction. Sixth, self-efficacy had a significant effect on user satisfaction.

• Journal of IKEEE
• /
• v.27 no.3
• /
• pp.251-257
• /
• 2023

In this paper, we propose a dual-structured self-attention method that improves the lack of regional features of the vision transformer's self-attention. Vision Transformers, which are more computationally efficient than convolutional neural networks in object classification, object segmentation, and video image recognition, lack the ability to extract regional features relatively. To solve this problem, many studies are conducted based on Windows or Shift Windows, but these methods weaken the advantages of self-attention-based transformers by increasing computational complexity using multiple levels of encoders. This paper proposes a dual-structure self-attention using self-attention and neighborhood network to improve locality inductive bias compared to the existing method. The neighborhood network for extracting local context information provides a much simpler computational complexity than the window structure. CIFAR-10 and CIFAR-100 were used to compare the performance of the proposed dual-structure self-attention transformer and the existing transformer, and the experiment showed improvements of 0.63% and 1.57% in Top-1 accuracy, respectively.

• Korean Journal of Mathematics
• /
• v.30 no.2
• /
• pp.297-304
• /
• 2022

Fixed point theory has many useful applications in applied sciences. The object of this paper is to obtain fixed point for continuous self mappings in Hilbert C^*-module with rational conditions.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.12
• /
• pp.3180-3199
• /
• 2013

While non-predefined object segmentation (NDOS) distinguishes an arbitrary self-assumed object from its background, predefined object segmentation (DOS) pre-specifies the target object. In this paper, a new and novel method to segment predefined objects is presented, by globally optimizing an orientation-based objective function that measures the fitness of the object boundary, in a discretized parameter space. A specific object is explicitly described by normalized discrete sets of boundary points and corresponding normal vectors with respect to its plane shape. The orientation factor provides robust distinctness for target objects. By considering the order of transformation elements, and their dependency on the derived over-segmentation outcome, the domain of translations and scales is efficiently discretized. A branch and bound algorithm is used to determine the transformation parameters of a shape model corresponding to a target object in an image. The results tested on the PASCAL dataset show a considerable achievement in solving complex backgrounds and unclear boundary images.

• Journal of computational fluids engineering
• /
• v.22 no.1
• /
• pp.88-94
• /
• 2017

The IR signal entering into a sensor is composed of the following components: the self-emitted component directly from the object surface, the reflected components of the solar and sky irradiance at the object surface, and the scattered component by the atmosphere without reference to any object surfaces. The self-emitted and reflected components from the object can be lowered by the atmospheric layer between the object and the IR sensor. The principle factors influencing the atmospheric transmittance are the air temperature, the relative humidity and the observation distance. Previous studies on IR signal transmission through the atmosphere are focused on uniform atmospheric conditions and the non-uniform nature of the atmosphere was not properly treated in modeling. In this study, we use the local atmospheric transmittance to simulate the non-uniform atmosphere in analyzing the IR signal from the object surface. The results show that the nonuniform analysis of the atmosphere becomes more important as the wavelength of IR signal increases.