• Journal of Aerospace System Engineering
• /
• v.16 no.5
• /
• pp.70-77
• /
• 2022

This paper proposes database generation and management system for small-pixelized airborne target recognition. The proposed system has five main features: 1) image extraction from in-flight test video frames, 2) automatic image archiving, 3) image data labeling and Meta data annotation, 4) virtual image data generation based on color channel convert conversion and seamless cloning and 5) HOG/LBP-based tiny-pixelized target augmented image data. The proposed framework is Python-based PyQt5 and has an interface that includes OpenCV. Using video files collected from flight tests, an image dataset for airborne target recognition on generates by using the proposed system and system input.

• The KIPS Transactions:PartB
• /
• v.11B no.5
• /
• pp.587-596
• /
• 2004

During recent years, a new framework, which aims to bring a unified and global approach in indexing, browsing and querying various digital multimedia data such as audio, video and image has been developed. This new system partitions each media stream into smaller units based on actual physical events. These physical events within oath media stream can then be effectively indexed for retrieval. In this paper, we present a new approach that exploits audio, image and video features to segment and analyze the audio-visual data. Integration of audio and visual analysis can overcome the weakness of previous approach that was based on the image or video analysis only. We Implement a web-based multi media data retrieval system called XCRAB and report on its experiment result.

• The KIPS Transactions:PartB
• /
• v.11B no.4
• /
• pp.509-516
• /
• 2004

Semantic interpretation of image is incomplete without some mechanism for understanding semantic content that is not directly visible. For this reason, human assisted content-annotation through natural language is an attachment of textual description to image. However, keyword-based retrieval is in the level of syntactic pattern matching. In other words, dissimilarity computation among terms is usually done by using string matching not concept matching. In this paper, we propose a method for computerized semantic similarity calculation In WordNet space. We consider the edge, depth, link type and density as well as existence of common ancestors. Also, we have introduced method that applied similarity measurement on semantic image retrieval. To combine wi#h the low level features, we use the spatial color distribution model. When tested on a image set of Microsoft's 'Design Gallery Line', proposed method outperforms other approach.

• The Transactions of the Korea Information Processing Society
• /
• v.2 no.6
• /
• pp.903-917
• /
• 1995

In most database systems images have been indexed indirectly using related texts such as captions, annotations and image attributes. But there has been an increasing requirement for the image database system supporting the storage and retrieval of images directly by content using the information contained in the images. There has been a few indexing methods based on contents. Among them, Pertains proposed an image indexing method considering spatial relationships and properties of objects forming the images. This is the expansion of the other studies based on '2-D string. But this method needs too much storage space and lacks flexibility. In this paper, we propose a more flexible index structure based on kd-tree using paging techniques. We show an example of extracting keys using normalization from the from the raw image. Simulation results show that our method improves in flexibility and needs much less storage space.

• Journal of Korea Multimedia Society
• /
• v.11 no.2
• /
• pp.143-152
• /
• 2008

Nowadays, many investigators are studying various methodologies concerning event expression for semantic retrieval of video data. However, most of the parts are still using annotation based retrieval that is defined into annotation of each data and content based retrieval using low-level features. So, we propose a method of creation of the motion unit and extracting event through the unit for the more semantic retrieval than existing methods. First, we classify motions by event unit. Second, we define semantic unit about classified motion of object. For using these to event extraction, we create rules that are able to match the low-level features, from which we are able to retrieve semantic event as a unit of video shot. For the evaluation of availability, we execute an experiment of extraction of semantic event in video image and get approximately 80% precision rate.

• Journal of KIISE:Databases
• /
• v.30 no.5
• /
• pp.494-506
• /
• 2003

Recently, explosive popularity of multimedia information has triggered the need for retrieving multimedia contents efficiently from the database including audio, video and images. In this paper, we propose an XML-based retrieval scheme and a data model that complement the weak aspects of annotation and conent based retrieval methods. The Property and hierarchy structure of image and video data are represented and manipulated based on the Multimedia Description Schema (MDS) that conforms to the MPEG-7 standard. For audio contents, pitch contours extracted from their acoustic features are converted into UDR string. Especially, to improve the retrieval performance, user's access pattern and frequency are utilized in the construction of an index. We have implemented a prototype system and evaluated its performance through various experiments.

• International journal of advanced smart convergence
• /
• v.8 no.4
• /
• pp.75-81
• /
• 2019

Though recent advancement of deep learning methods have provided satisfactory results from large data domain, somehow yield poor performance on few-shot classification tasks. In order to train a model with strong performance, i.e. deep convolutional neural network, it depends heavily on huge dataset and the labeled classes of the dataset can be extremely humongous. The cost of human annotation and scarcity of the data among the classes have drastically limited the capability of current image classification model. On the contrary, humans are excellent in terms of learning or recognizing new unseen classes with merely small set of labeled examples. Few-shot learning aims to train a classification model with limited labeled samples to recognize new classes that have neverseen during training process. In this paper, we increase the backbone depth of the embedding network in orderto learn the variation between the intra-class. By increasing the network depth of the embedding module, we are able to achieve competitive performance due to the minimized intra-class variation.

• International Journal of Contents
• /
• v.14 no.1
• /
• pp.1-6
• /
• 2018

Text, as one of the most influential inventions of humanity, has played an important role in human life since ancient times. The rich and precise information embodied in text is very useful in a wide range of vision-based applications such as the text data extracted from images that can provide information for automatic annotation, indexing, language translation, and the assistance systems for impaired persons. Therefore, natural-scene text detection with active research topics regarding computer vision and document analysis is very important. Previous methods have poor performances due to numerous false-positive and true-negative regions. In this paper, a fully-convolutional-network (FCN)-based method that uses supervised architecture is used to localize textual regions. The model was trained directly using images wherein pixel values were used as inputs and binary ground truth was used as label. The method was evaluated using ICDAR-2013 dataset and proved to be comparable to other feature-based methods. It could expedite research on text detection using deep-learning based approach in the future.

• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06c
• /
• pp.425-428
• /
• 2008

모바일기기에 탑재되어있는 디지털 카메라의 성능이 향상됨에 따라 이를 이용한 사진의 촬영 및 수집이 용이해졌으며, 따라서 사용자 로그정보를 이용하여 방대한 양의 사진을 분석하거나 브라우징해주는 방법들이 연구되고 있다. 본 논문에서는 모바일기기의 불확실한 로그정보와 사진 주석정보를 베이지안 네트워크로 모델링하여 사용자가 겪은 이벤트들을 추론하고 사용자의 일과를 요약해주는 방법을 제안한다. 우선 사진들을 시간과 위치정보에 따라 분할하여 사진그룹목록을 생성하고, 이를 모바일기기에 입력되어있는 사용자의 일정목록과 합하여 임시이벤트목록을 생성한다. 그 뒤 베이지안 네트워크를 이용하여 각 이벤트를 인식하고 이를 가장 잘 나타내는 사진을 선택한다. 제안하는 방법은 선택된 사진들을 나열하여 사진다이어리형식으로 사용자의 일과를 요약하여주며, 이때 특정 이벤트와 매치되는 사진이 없을 경우 미리 정의되어있는 만화 컷을 대신 사용하여 내용이 매끄럽게 이어지도록 하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06b
• /
• pp.378-380
• /
• 2012

디지털 카메라 및 카메라가 부착된 스마트폰의 보급으로 인해 개인 및 사업장에서 관리해야할 사진의 양이 증가 하였다. 본 논문에서는 이미지 어휘 트리를 이용하여 다량의 사진에 주석을 자동으로 추천 하는 알고리즘을 제안하여 대량의 사진 관리를 효과적으로 하도록 한다. 제안한 방법에서는 어휘트리 생성 시 학습 데이터 모음에 포함된 사진들의 주석에 대한 정보도 함께 갖도록 한다. 그래서 입력으로 들어온 사진과 가장 가까운 사진을 어휘 트리를 통해 찾고 찾은 사진이 가지고 있는 주석을 최종 출력으로 낸다. 제안한 알고리즘은 자동으로 다량의 사진의 주석을 빠르게 추천 할 수 있다.