• International journal of advanced smart convergence
• /
• v.8 no.3
• /
• pp.131-137
• /
• 2019

This study proposes a method of learning and recognizing the characteristics that are the classification criteria of Hangul using Mask R-CNN, one of the deep learning techniques, to recognize and classify atypical Hangul characters. The atypical characters on the Hangul signboard have a lot of deformed and colorful shapes beyond the general characters. Therefore, in order to recognize the Hangul signboard character, it is necessary to learn a separate atypical Hangul character rather than the existing formulaic one. We selected the Hangul character '닭' as sample data and constructed 5,383 Hangul image data sets and used them for learning and verifying the deep learning model. The accuracy of the results of analyzing the performance of the learning model using the test set constructed to verify the reliability of the learning model was about 92.65% (the area detection rate). Therefore we confirmed that the proposed method is very useful for Hangul signboard character recognition, and we plan to extend it to various Hangul data.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.1301-1304
• /
• 2022

최근 인공지능 신경망에 대한 활발한 연구를 바탕으로 다양한 분야에서의 적용에 대해 많은 시도들이 이루어지고 있다. 이러한 흐름에 맞추어 화학 문서에서 화학 구조를 인식하는 문제 또한 딥러닝을 이용하여 해결하려는 시도들이 생겨나고 있다. 본 논문에서는 화학 문서에서 화학 구조를 인식하는 모델을 학습시키기 위한 합성 데이터셋을 제안하였다. 문서의 구조를 이용하여 정교하게 화학 구조들을 문서에 합성하여 데이터셋을 생성하였고, 이를 최신 딥러닝 모델 중 하나인 Mask R-CNN[7]에 학습시켜 제안한 데이터셋을 이용하여 문서에서 화학 구조를 인식할 수 있음을 보였다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• fall
• /
• pp.88-90
• /
• 2021

본 논문은 화학 문서 내에 존재하는 표를 검출하는 문제를 다룬다. 우선 문서에서 표가 있을 만한 영역만을 남긴 후, 객체 검출 분야에서 좋은 성능을 보이는 Cascade Mask R-CNN을 이용하여 화학 문서 내 표 검출을 수행하였다. 더 나아가 감마 보정과 스캔 잡음을 이용하여 학습 데이터를 증강함으로써 다양한 스타일의 표들을 강인하게 검출할 수 있도록 하였다. 합성 화학문서와 실제 화학 문서에 대해 제안한 방법을 적용하여 표 검출 성능을 측정하였다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.47-50
• /
• 2020

최근 Panorama와 360도 영상이 대표되는 몰입형(Immersive) 미디어 콘텐츠의 사용이 증가하고 있다. 몰입형 영상 콘텐츠는 사용자에게 현장감을 제공해야 하지만, 촬영 카메라 간의 시차(Parallax)로 인해 영상 콘텐츠에서 시차 왜곡이 발생할 수 있고, 이는 사용자의 콘텐츠 몰입을 제한하기 때문에 해당 영상 콘텐츠의 제작 기술인 영상 스티칭의 높은 정확도가 요구되고 있다. 지금까지 스티칭 영상의 시차 왜곡을 줄이기 위하여 다중 호모그래피 추정 방법과 Seam Optimization 방법이 제안되었지만, 영상 내 사물 배치에 따라 기술 적용이 제한될 수 있다. 이에 본 논문에서는 Mask R-CNN을 활용하여 사물을 세그먼트화하고, 사물의 종류에 따라 각각 다른 가중치 적용을 통해 시차 왜곡을 방지하며, 영상 내 사물의 배치에 따라 시차 왜곡이 발생할 상황에서는 사용자의 인지 중요도가 낮은 사물로 시차 왜곡을 유도하는 영상 스티칭 방법을 제안한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.562-565
• /
• 2021

스타일 변환 모델은 이미지 전체나 이미지 내에서 사용자가 지정한 영역을 대상으로 스타일을 변환시킨다. 이런 방식은 이미지 내의 다수의 객체에 대해 스타일 변환을 시행할 때 일일이 영역을 지정해 줘야 한다는 불편함과 결과물의 전체 해상도가 떨어진다는 한계를 가지고 있다. 본 논문에서는 이런 한계들을 극복하기 위해 객체탐지 모델과 스타일변환 모델을 연동한 객체스타일변환모델을 제안하고 모델 간 연동방법에 대해 자세히 서술한다. 객체탐지모델인 Mask R-CNN 을 통해 필요한 객체를 탐지하고 탐지한 객체의 특징맵들을 스타일변환 모델인 zi2zi 의 입력 값으로 전달하여 이미지 내의 필요한 객체들만 스타일변환이 이루어지도록 모델이 동작한다. 이러한 모델은 기존에 있는 두 모델을 재사용함으로써 모델을 처음부터 새로 설계할 필요가 없다는 장점이 있으며, 공개된 다양한 모델들을 서로 융합하여 사용할 수 있는 방법을 제시하는데 도움을 줄 것이다.

• The Journal of Korea Robotics Society
• /
• v.17 no.3
• /
• pp.255-263
• /
• 2022

3D object detection pipelines often incorporate RGB-based object detection methods such as YOLO, which detects the object classes and bounding boxes from the RGB image. However, in complex environments where objects are heavily cluttered, bounding box approaches may show degraded performance due to the overlapping bounding boxes. Mask based methods such as Mask R-CNN can handle such situation better thanks to their detailed object masks, but they require much longer time for data preparation compared to bounding box-based approaches. In this paper, we present a 3D object recognition pipeline which uses either the YOLO or Mask R-CNN real-time object detection algorithm, K-nearest clustering algorithm, mask reduction algorithm and finally Principal Component Analysis (PCA) alg orithm to efficiently detect 3D poses of objects in a complex environment. Furthermore, we also present an improved YOLO based 3D object detection algorithm that uses a prioritized heightmap clustering algorithm to handle overlapping bounding boxes. The suggested algorithms have successfully been used at the Artificial-Intelligence Robot Challenge (ARC) 2021 competition with excellent results.

• Journal of Korea Society of Industrial Information Systems
• /
• v.27 no.6
• /
• pp.13-23
• /
• 2022

In this paper, we propose a detailed component image classification algorithm by fashion item for unstructured data retrieval in the fashion field. Due to the COVID-19 environment, AI-based online shopping malls are increasing recently. However, there is a limit to accurate unstructured data search with existing keyword search and personalized style recommendations based on user surfing behavior. In this study, pre-processing using Mask R-CNN was conducted using images crawled from online shopping sites and then classified components for each fashion item through CNN. We obtain the accuaracy for collar of the shirt's as 93.28%, the pattern of the shirt as 98.10%, the 3 classese fit of the jeans as 91.73%, And, we further obtained one for the 4 classes fit of jeans as 81.59% and the color of the jeans as 93.91%. At the results for the decorated items, we also obtained the accuract of the washing of the jeans as 91.20% and the demage of jeans accuaracy as 92.96%.

• Journal of Broadcast Engineering
• /
• v.25 no.6
• /
• pp.944-953
• /
• 2020

Object detection plays a crucial role in a self-driving system. With the advances of image recognition based on deep convolutional neural networks, researches on object detection have been actively explored. In this paper, we proposed a lightweight model of the mask R-CNN, which has been most widely used for object detection, to efficiently predict location and shape of various objects on the road environment. Furthermore, feature maps are adaptively re-calibrated to improve the detection performance by applying an attention module to the neural network layer that plays different roles within the mask R-CNN. Various experimental results for real driving scenes demonstrate that the proposed method is able to maintain the high detection performance with significantly reduced network parameters.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.36 no.6
• /
• pp.469-481
• /
• 2018

DL (Deep Learning) is getting popular in various fields to implement artificial intelligence that resembles human learning and cognition. DL based on complicate structure of the ANN (Artificial Neural Network) requires computing power and computation cost. Variety of DL models with improved performance have been developed with powerful computer specification. The main purpose of this paper is to detect buildings from aerial images and evaluate performance of Mask R-CNN (Region-based Convolutional Neural Network) developed by FAIR (Facebook AI Research) team recently. Mask R-CNN is a R-CNN that is evaluated to be one of the best ANN models in terms of performance for semantic segmentation with pixel-level accuracy. The performance of the DL models is determined by training ability as well as architecture of the ANN. In this paper, we characteristics of the Mask R-CNN with various types of the images and evaluate possibility of the generalization which is the ultimate goal of the DL. As for future study, it is expected that reliability and generalization of DL will be improved by using a variety of spatial information data for training of the DL models.

• Journal of the Semiconductor & Display Technology
• /
• v.20 no.2
• /
• pp.126-130
• /
• 2021

The ball grid array is one of the packaging methods that used in high density printed circuit board. Solder void defects caused by voids in the solder ball during the BGA process do not directly affect the reliability of the product, but it may accelerate the aging of the device on the PCB layer or interface surface depending on its size or location. Void inspection is important because it is related in yields with products. The most important process in the optical inspection of solder void is the segmentation process of solder and void. However, there are several segmentation algorithms for the vision inspection, it is impossible to inspect all of images ideally. When X-Ray images with poor contrast and high level of noise become difficult to perform image processing for vision inspection in terms of software programming. This paper suggests the solution to deal with the suggested problem by means of using Mask R-CNN instead of digital image processing algorithm. Mask R-CNN model can be trained with images pre-processed to increase contrast or alleviate noises. With this process, it provides more efficient system about complex object segmentation than conventional system.