• 대한전자공학회논문지SP
• /
• 제37권6호
• /
• pp.17-21
• /
• 2000

내용기반 영상 검색에서는 컬러, 형태, 질감의 세 가지 대표적인 영상 특징들이 주로 사용된다. 한 가지 특징만을 사용하는 검색 방법은 영상의 내용이 복잡하거나 비교대상이 되는 영상의 수가 많아질수록 좋은 성능을 보이지 못한다. 그래서 여러 가지 영상 특징들을 결합한 방법들이 많이 연구되고 있다. 그러나 여러 특징들을 결합해서 사용하는 검색 시스템이라 할지라도 각 특징들에 대한 가중치가 적합하게 부여되지 않으면 검색되는 결과 영상들의 순위가 크게 변하여 검색 성능이 떨어지게 된다. 이러한 문제점을 해결하기 위해 본 논문에서는 여러 영상 특징들이 결합해서 사용될 때 각 특징에 대한 가중치를 자동적으로 부여해서 검색 성능을 개선하고자 한다. 제안한 방법을 992개의 테스트 영상들로 구성된 데이터 베이스에서 실험을 하고 다양한 성능평가 방법을 통해 그 타당성을 확인하였으며 제안한 방법을 고정가중치 부여를 이용한 방법과 비교하여 검색 성능이 개선됨을 볼 수 있었다.

• 한국통신학회논문지
• /
• 제36권4C호
• /
• pp.226-231
• /
• 2011

본 논문은 기존 어류도감을 개선하기 위해 기존 어류도감에 IT 기술들이 적용된 전자어류도감의 개념을 도입하고 이를 위한 영상검색 알고리즘을 제안한다. 영상검색은 전자어류도감의 핵심이고 기존 어류도감을 압도하게 하는 기술이다. 어류는 동종일지라도 형태, 컬러, 질감에서 다른 특징들을 갖고, 심지어 동일 어류도 촬영 시의 자세나 주변 환경에 의해 다른 특징을 갖기 때문에 형태, 컬러, 질감의 단순한 특징을 이용하는 기존 영상검색은 전자어류도감에 적합하지 못하다. 제안된 영상검색은 어류의 머리, 몸통, 꼬리에서 추출된 상세 특징들을 채택하고, 특징들에는 그들의 불변성에 따라 가중치가 다르게 주어진다. 시뮬레이션 결과들은 제안된 알고리즘이 기존 알고리즘을 훨씬 능가하는 것을 보여준다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권10호
• /
• pp.3390-3405
• /
• 2022

Long Short-Term Memory (LSTM) combined with attention mechanism is extensively used to generate semantic sentences of images in image captioning models. However, features of salient regions and spatial information are not utilized sufficiently in most related works. Meanwhile, the LSTM also suffers from the problem of underutilized information in a single time step. In the paper, two innovative approaches are proposed to solve these problems. First, the Synergy-Gated Attention (SGA) method is proposed, which can process the spatial features and the salient region features of given images simultaneously. SGA establishes a gated mechanism through the global features to guide the interaction of information between these two features. Then, the Recurrent Fusion LSTM (RF-LSTM) mechanism is proposed, which can predict the next hidden vectors in one time step and improve linguistic coherence by fusing future information. Experimental results on the benchmark dataset of MSCOCO show that compared with the state-of-the-art methods, the proposed method can improve the performance of image captioning model, and achieve competitive performance on multiple evaluation indicators.

• 제어로봇시스템학회논문지
• /
• 제8권6호
• /
• pp.455-461
• /
• 2002

Object-based image retrieval method is addressed. A new image segmentation algorithm and image comparing method between segmented objects are proposed. For image segmentation, color and texture features are extracted from each pixel in the image. These features we used as inputs into VQ (Vector Quantization) clustering method, which yields homogeneous objects in terns of color and texture. In this procedure, colors are quantized into a few dominant colors for simple representation and efficient retrieval. In retrieval case, two comparing schemes are proposed. Comparing between one query object and multi objects of a database image and comparing between multi query objects and multi objects of a database image are proposed. For fast retrieval, dominant object colors are key-indexed into database.

• 대한임베디드공학회논문지
• /
• 제6권1호
• /
• pp.48-54
• /
• 2011

As personal or compact devices with image acquisition functionality are becoming easily available for common users, the voluminous images that need to be managed by image related services or systems demand efficient and effective methods in the perspective of image identification. The objective of image identification is to associate an image with a unique identifier. Moreover, whenever an image identifier needs to be regenerated, the newly generated identifier should be consistent. In this paper, we propose three image identifier generation methods utilizing image features: linear component, luminance area, and combination of both features. The linear component based method exploits the information of distribution of partial lines over an image, while the luminance area based method utilizes the partition of an image into a number of small areas according to the same luminance degree. The third method is proposed in order to take advantage of both former methods. In this paper, we also demonstrate the experimental evaluations for uniqueness and similarity analysis that have shown favorable results.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2003년도 학술회의 논문집 정보 및 제어부문 A
• /
• pp.191-194
• /
• 2003

In this paper, we propose a new accurate and robust image mosaic technique of the mural underdrawing taken from the infra-red camera, which is based on multiple image registration and adaptive blending technique. The image mosaicing methods which have been developed so far have the following deficits. It is hard to generate a high resolution image when there are regions that do not have features or intensity gradients, and there is a trade-off in overlapping region site in view of registration and blending. We consider these issues as follows. First, in order to mosaic Images with neither noticeable features nor intensity gradients, we use a Projected supplementary pattern and pseudo color image for features in the image Pieces which are registered. Second, we search the overlapping region size with minimum blending error between two adjacent images and then apply blending technique to minimum error overlapping region. Finally, we could find our proposed method is more effective and efficient for image mosaicing than conventional mosaic techniques and also is more adequate for the application of infra-red mural underdrawing mosaicing. Experimental results show the accuracy and robustness of the algorithm.

• Journal of Information Processing Systems
• /
• 제9권3호
• /
• pp.349-364
• /
• 2013

The big challenge in current content-based image retrieval systems is to reduce the semantic gap between the low level-features and high-level concepts. In this paper, we have proposed a novel framework for efficient image retrieval to improve the retrieval results significantly as a means to addressing this problem. In our proposed method, we first extracted a strong set of image features by using the dual-tree rotated complex wavelet filters (DT-RCWF) and dual tree-complex wavelet transform (DT-CWT) jointly, which obtains features in 12 different directions. Second, we presented a relevance feedback (RF) framework for efficient image retrieval by employing a support vector machine (SVM), which learns the semantic relationship among images using the knowledge, based on the user interaction. Extensive experiments show that there is a significant improvement in retrieval performance with the proposed method using SVMRF compared with the retrieval performance without RF. The proposed method improves retrieval performance from 78.5% to 92.29% on the texture database in terms of retrieval accuracy and from 57.20% to 94.2% on the Corel image database, in terms of precision in a much lower number of iterations.

• 패션비즈니스
• /
• 제9권5호
• /
• pp.1-14
• /
• 2005

The purpose of this study was to analysis the perception of men's wear brands (Intermezzo and Rogatis), for developing the possibility & strategy of the nichi-market in men's wear market for the apparel marketers and manufactures. For this study, the data obtained from 312 respondents were analyzed by descriptive statistics, ANOVA. The results from the study were as follow ; The perception of the 2 brand images revealed that Intermezzo accounted for 79.8% of the frequencies, and Rogatis accounted for 99%. Also, results revealed the total evaluation of Intermezzo accounted for 3.86 of the mean rated on 5 point Likert-type scales in the 9 features, and Rogatis accounted for 3.28. And then, results revealed that there were signifiant differences in 2 cluster of Rogatis that the purchasing cluster accounted for 3.46 of the mean, and the perceiving cluster accounted for 3.07. The brand images of Intermezzo and Rogatis were evaluated and rated on 5 point Likert-type scales of 17 pair adjectives. As a results, the image characteristic with Intermezzo was considered with more dynamic, trendy than the image characteristic with Rogatis. Also, results revealed that The Image with Intermezzo was considered with urban, lively, chic, modern, and sophsticated image-features, and the Image with Rogatis were evaluated mannish, urban, sophsticated, luxury, and static image-features.

• Journal of Multimedia Information System
• /
• 제9권2호
• /
• pp.75-86
• /
• 2022

The existing image quality assessment (IQA) datasets have a small number of samples. Some methods based on transfer learning or data augmentation cannot make good use of image quality-related features. A No Reference (NR)-IQA method based on multi-task training and quality awareness is proposed. First, single or multiple distortion types and levels are imposed on the original image, and different strategies are used to augment different types of distortion datasets. With the idea of weak supervision, we use the Full Reference (FR)-IQA methods to obtain the pseudo-score label of the generated image. Then, we combine the classification information of the distortion type, level, and the information of the image quality score. The ResNet50 network is trained in the pre-train stage on the augmented dataset to obtain more quality-aware pre-training weights. Finally, the fine-tuning stage training is performed on the target IQA dataset using the quality-aware weights to predicate the final prediction score. Various experiments designed on the synthetic distortions and authentic distortions datasets (LIVE, CSIQ, TID2013, LIVEC, KonIQ-10K) prove that the proposed method can utilize the image quality-related features better than the method using only single-task training. The extracted quality-aware features improve the accuracy of the model.

• 한국컴퓨터정보학회논문지
• /
• 제5권3호
• /
• pp.215-219
• /
• 2000

The purpose of this paper is to propose a method to maximize the efficiency of a content-based image retrieval for various kinds of images. This paper discuss the self-adaptivity for the change of image domain and the self-selection of optimal features for query image, and present the efficient method to maximize content-based retrieval for various kinds of images. In this method, a content-based retrieval system is adopted to select automatically distinctive feature patterns which have a maximum efficiency of image retrieval in various kinds of images. Experimental results show that the Proposed method is improved 3% than the method using individual features.