http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
불확실성을 이용한 딥러닝 기반의 항공 이미지 객체 탐지
박주찬(Joo-Chan Park),이선훈(Seon-Hoon Lee),정준욱(Jun-Uk Jung),손성빈(Sung-Bin Son),오흥선(Heung-Seon Oh),정유철(Yuchul Jung) 제어로봇시스템학회 2020 제어·로봇·시스템학회 논문지 Vol.26 No.11
Object detection in aerial images is an important task because it is used in various applications such as land management, disaster monitoring, national security, and map production, However, owing to the characteristics of aerial images, such as high resolution, data imbalance between classes, lack of data, and densely appearing objects, it is difficult to improve the performance even with the recent deep learning-based object detection models. To overcome these challenges, this paper proposes an uncertainty-based max-margin learning method and a data augmentation method based on attribute transformation specialized for aerial images. The superiority of the proposed methods based on a deep learning-based object detection model is revealed by it winning the aerial image object detection contest 2020.
딥 앙상블을 이용한 딥러닝 기반의 항공 이미지 객체 탐지
박주찬(Joo-Chan Park),손성빈(Sung-Bin Son),이선훈(Seon-Hoon Lee),정준욱(Jun-Uk Jung),박용준(Yong-Jun Park),오흥선(Heung-Seon Oh) 제어로봇시스템학회 2021 제어·로봇·시스템학회 논문지 Vol.27 No.12
Object detection in aerial images is continuously studied for various purposes such as national security, disaster monitoring, and meteorological observation. It is difficult to improve recent object detection methods based on a single model using deep learning due to severe class imbalance. This paper proposes a deep ensemble method combining two models with different strengths and a class-dependent thresholding method by considering the object distribution. We demonstrate the superiority of our methods in a series of experiments. In addition, we take 1st place in both public and private scores in the Arirang satellite image AI object detection contest.
반도체 불량원인 분석을 위한 딥뉴럴네트워크 기반의 패치 이미지 병합 시스템
손성빈(Sung-Bin Son),이선훈(Seon-Hoon Lee),박주찬(Joo-Chan Park),정준욱(Jun-Uk Jung),박용준(Yong-Joon Park),오흥선(Heung-Seon Oh) 제어로봇시스템학회 2021 제어·로봇·시스템학회 논문지 Vol.27 No.8
In the integrated circuit/chip manufacturing process, failure analysis performed to find defects utilizes high-resolution chip images obtained through auto-shot scope equipment, which combines microscopy and automatic photography. However, due to the incorrect focus and the unexpected overlap size depending on the distance between the microscope and the chip, these systems are noisy. Thus, failure analysis cannot be performed effectively because the individual conduction the examination is exposed to noisy images, thereby taking a long time. We proposed a system called DeepMerge that utilizes deep learning-based learning-based features such as pint extraction and feature matching to overcome the aforementioned challenges. We will be indicating the effectiveness and efficiency of our system by obtaining practical image data from the industry.