• 예술인문사회 융합 멀티미디어 논문지
• /
• 제9권11호
• /
• pp.1133-1144
• /
• 2019

최근 스마트 기기의 발전으로 인터넷상에 존재하는 이미지 데이터의 양이 급속하게 증가하는 상황에서 효과적인 이미지 검색을 위한 다양한 방법들이 연구되고 있다. 기존의 이미지 검색 방법들은 이미지에 존재하는 물체들을 단순하게 검출하여 각 물체들의 라벨 정보에 근거한 검색을 수행하기 때문에 사용자가 원하는 이미지와 검색 결과로 얻은 이미지 간에 의미적 차이인 시맨틱 갭(Semantic Gap)이 발생된다. 이미지 검색에서 발생하는 시맨틱 갭을 줄이기 위해, 본 논문에서는 딥러닝 기반의 다중 객체 분류 모듈과 사람의 행위를 분류하는 모듈을 연결하고, 이 모듈들에 행위 온톨로지를 결합하였다. 즉, 딥러닝과 행위 온톨로지의 결합을 기반으로 객체들 간의 연관성을 고려한 이미지 검색 시스템을 제안한다. 이미지에 포함된 동적인 행위를 고려하기 위해 Walking과 Running 데이터를 이용하여 실험한 결과를 분석하였다. 제안한 방법은 향후 이미지 검색 결과의 정확도를 높일 수 있는 영상의 자동 주석 생성 연구에 확장하여 적용할 수 있다.

• 대한방사선기술학회지:방사선기술과학
• /
• 제44권6호
• /
• pp.655-661
• /
• 2021

Positron emission tomography (PET) images is affected by acquisition time, short acquisition times results in low gamma counts leading to degradation of image quality by statistical noise. Noise2Void(N2V) is self supervised denoising model that is convolutional neural network (CNN) based deep learning. The purpose of this study is to evaluate denoising performance of N2V for PET image with a short acquisition time. The phantom was scanned as a list mode for 10 min using Biograph mCT40 of PET/CT (Siemens Healthcare, Erlangen, Germany). We compared PET images using NEMA image-quality phantom for standard acquisition time (10 min), short acquisition time (2min) and simulated PET image (S2 min). To evaluate performance of N2V, the peak signal to noise ratio (PSNR), normalized root mean square error (NRMSE), structural similarity index (SSIM) and radio-activity recovery coefficient (RC) were used. The PSNR, NRMSE and SSIM for 2 min and S2 min PET images compared to 10min PET image were 30.983, 33.936, 9.954, 7.609 and 0.916, 0.934 respectively. The RC for spheres with S2 min PET image also met European Association of Nuclear Medicine Research Ltd. (EARL) FDG PET accreditation program. We confirmed generated S2 min PET image from N2V deep learning showed improvement results compared to 2 min PET image and The PET images on visual analysis were also comparable between 10 min and S2 min PET images. In conclusion, noisy PET image by means of short acquisition time using N2V denoising network model can be improved image quality without underestimation of radioactivity.

• 로봇학회논문지
• /
• 제15권2호
• /
• pp.197-204
• /
• 2020

In this paper, we present a learning platform for robotic grasping in real world, in which actor-critic deep reinforcement learning is employed to directly learn the grasping skill from raw image pixels and rarely observed rewards. This is a challenging task because existing algorithms based on deep reinforcement learning require an extensive number of training data or massive computational cost so that they cannot be affordable in real world settings. To address this problems, the proposed learning platform basically consists of two training phases; a learning phase in simulator and subsequent learning in real world. Here, main processing blocks in the platform are extraction of latent vector based on state representation learning and disentanglement of a raw image, generation of adapted synthetic image using generative adversarial networks, and object detection and arm segmentation for the disentanglement. We demonstrate the effectiveness of this approach in a real environment.

• 천문학회보
• /
• 제44권2호
• /
• pp.70.4-70.4
• /
• 2019

Image super-resolution (SR) is a technique that enhances the resolution of a low resolution image. In this study, we use three SR models (RCAN, ProSRGAN and Bicubic) for enhancing solar SDO/HMI magnetograms using deep learning. Each model generates a high resolution HMI image from a low resolution HMI image (4 by 4 binning). The pixel resolution of HMI is about 0.504 arcsec. Deep learning networks try to find the hidden equation between low resolution image and high resolution image from given input and the corresponding output image. In this study, we trained three models with HMI images in 2014 and test them with HMI images in 2015. We find that the RCAN model achieves higher quality results than the other two methods in view of both visual aspects and metrics: 31.40 peak signal-to-noise ratio(PSNR), Correlation Coefficient (0.96), Root mean square error (RMSE) is 0.004. This result is also much better than the conventional bi-cubic interpolation. We apply this model to a full-resolution SDO/HMI image and compare the generated image with the corresponding Hinode NFI magnetogram. As a result, we get a very high correlation (0.92) between the generated SR magnetogram and the Hinode one.

• 한국컴퓨터정보학회논문지
• /
• 제28권2호
• /
• pp.19-25
• /
• 2023

이 논문에서는 image signal processing 을 고려하여 저조도에서 촬영된 저품질의 raw 이미지를 딥러닝에 기반하여 개선하는 방법을 제안한다. 스마트폰 카메라의 경우 DSLR 카메라에 비해 렌즈나 센서의 확장에 제약이 있어 저조도 상황에서 이미지에 노이즈가 증가되고 품질이 저하되는 문제점을 보인다. 기존 딥러닝 기반 저조도 이미지 처리 방식은 image signal processing의 주요 요소인 렌즈 쉐이딩 효과와 화이트 밸런스를 고려하지 못하여 부자연스러운 이미지를 생성하기도 한다. 본 논문에서는 렌즈 쉐이딩 효과와 화이트 밸런스를 딥러닝 모델에 적용하기 위해 중심거리와 채널 평균을 활용한다. 스마트폰으로 촬영된 저조도 이미지를 통한 실험에서 제안하는 방법이 기존 방법에 비해 더 높은 peak signal to noise ratio 와 structural similarity index measure를 달성함과 동시에 높은 품질의 저조도 이미지를 생성함을 확인한다.

• 한국멀티미디어학회논문지
• /
• 제25권5호
• /
• pp.757-768
• /
• 2022

In this paper, we proposed a system that visualizes a hologram device in 3D by utilizing the CT image segmentation function based on artificial intelligence deep learning. The input axial CT medical image is converted into Sagittal and Coronal, and the input image and the converted image are divided into 3D volumes using ResUNet, a deep learning model. In addition, the volume is created by segmenting the tumor region in the segmented liver image. Each result is integrated into one 3D volume, displayed in a medical image viewer, and converted into a video. When the converted video is transmitted to the hologram device and output from the device, a 3D image with a sense of space can be checked. As for the performance of the deep learning model, in Axial, the basic input image, DSC showed 95.0% performance in liver region segmentation and 67.5% in liver tumor region segmentation. If the system is applied to a real-world care environment, additional physical contact is not required, making it safer for patients to explain changes before and after surgery more easily. In addition, it will provide medical staff with information on liver and liver tumors necessary for treatment or surgery in a three-dimensional manner, and help patients manage them after surgery by comparing and observing the liver before and after liver resection.

• 대한임베디드공학회논문지
• /
• 제19권1호
• /
• pp.19-25
• /
• 2024

This paper proposes an artificial intelligence-based personal information area object detection optimization method in an embedded system to de-identify personal information in video data. As an object detection optimization method, first, in order to increase the detection rate for personal information areas when detecting objects, a gyro sensor is used to collect the shooting angle of the image data when acquiring the image, and the image data is converted into a horizontal image through the collected shooting angle. Based on this, each learning model was created according to changes in the size of the image resolution of the learning data and changes in the learning method of the learning engine, and the effectiveness of the optimal learning model was selected and evaluated through an experimental method. As a de-identification method, a shuffling-based masking method was used, and double-key-based encryption of the masking information was used to prevent restoration by others. In order to reuse the original image, the original image could be restored through a security key. Through this, we were able to secure security for high personal information areas and improve usability through original image restoration. The research results of this paper are expected to contribute to industrial use of data without personal information leakage and to reducing the cost of personal information protection in industrial fields using video through de-identification of personal information areas included in video data.

• International Journal of Contents
• /
• 제16권2호
• /
• pp.8-18
• /
• 2020

Data is critical in deep learning but the scarcity of data often occurs in research, especially in the preparation of the paired training data. In this paper, document image binarization with unpaired data is studied by introducing adversarial learning, excluding the need for supervised or labeled datasets. However, the simple extension of the previous unpaired training to binarization inevitably leads to poor performance compared to paired data training. Thus, a new deep learning approach is proposed by introducing a multi-diversity of higher quality generated images. In this paper, a two-stage model is proposed that comprises the generative adversarial network (GAN) followed by the U-net network. In the first stage, the GAN uses the unpaired image data to create paired image data. With the second stage, the generated paired image data are passed through the U-net network for binarization. Thus, the trained U-net becomes the binarization model during the testing. The proposed model has been evaluated over the publicly available DIBCO dataset and it outperforms other techniques on unpaired training data. The paper shows the potential of using unpaired data for binarization, for the first time in the literature, which can be further improved to replace paired data training for binarization in the future.

• 한국초등과학교육학회지:초등과학교육
• /
• 제28권3호
• /
• pp.245-262
• /
• 2009

This research was to examine the understandings of elementary school teachers on the phenomena related to light and image, and to survey their perception change related to learning contents of optics. The subjects were selected from the elementary teachers who were enrolled in a graduate course, 'Science education seminar' at an education college located in Chungchungbuk-Do, South Korea. Among the five students who exposed their perceptions clearly in the class, the three of them were selected who agreed to the proposal of the case study. To achieve the purpose of this study, semi-structured interviews following the conception test with the 3 elementary teachers were conducted. During the analysis of the data, additional interviews by phone, e-mail, and internet messenger were conducted if necessary. According to the results, all of the elementary school teachers lacked the scientific conceptions of the phenomena related to light and image. Unfortunately, their learning experiences did not help them to understand the scientific concepts. During the interviews, the teachers recognized the importance of the viewpoints of seeing, image, cognition of light, point light source to understand the phenomena related to light and image.

• 전기전자학회논문지
• /
• 제25권1호
• /
• pp.139-147
• /
• 2021

본 논문에서는 저조도 영상을 개선하기 위한 영상 분해 기반 심층 학습 방법 및 분해 채널 특성에 따른 손실함수를 제안한다. 기존 기법들의 문제점인 색신호 왜곡 및 할로 현상을 제거하기 위해, 입력 영상의 휘도 채널을 반사 성분과 조도 성분으로 분해하고, 반사 성분, 조도 성분 및 색차 신호를 신호 특성에 적합한 심층학습 과정을 적용하는 분해 기반 다중 구조 심층 학습 방법을 제안한다. 더불어, 분해 채널들의 특성에 따른 혼합 놈 기반의 손실함수를 정의하여 복원 영상의 안정성을 증대하고 열화 현상을 제거하기 위한 기법에 대해 기술한다. 실험 결과를 통해 제안한 방법이 다양한 저조도 영상을 효과적으로 개선하였음을 확인할 수 있었다.