• 대한전기학회논문지:시스템및제어부문D
• /
• 제52권4호
• /
• pp.212-222
• /
• 2003

The ability to efficiently and robustly recover accurate 3D terrain models from sets of stereoscopic images is important to many civilian and military applications. A stereo matching has been an important tool for reconstructing three dimensional terrain. However, there exist many factors causing stereo matching error, such as occlusion, no feature or repetitive pattern in the correlation window, intensity variation, etc. Among them, occlusion can be only resolved by true multi-image stereo. In this paper, we present multi-image stereo method using DEM fusion as one of efficient and reliable true multi-image methods. Elevations generated by all pairs of images are combined by the fusion process which accepts an accurate elevation and rejects an outlier. We propose three fusion schemes: THD(Thresholding), BPS(Best Pair Selection) and MS(Median Selection). THD averages elevations after rejecting outliers by thresholding, while BPS selects the most reliable elevation. To determine the reliability of a elevation or detect the outlier, we employ the measure of self-consistency. The last scheme, MS, selects the median value of elevations. We test the effectiveness of the proposed methods with a quantitative analysis using simulated images. Experimental results indicate that all three fusion schemes showed much better improvement over the conventional binocular stereo in natural terrain of 29 Palms and urban site of Avenches.

• Journal of Information Processing Systems
• /
• 제13권3호
• /
• pp.464-475
• /
• 2017

In this paper, we present a new texture image retrieval method which combines color and texture features extracted from images by a set of multi-resolution multi-direction (MRMD) filters. The MRMD filter set chosen is simple and can be separable to low and high frequency information, and provides efficient multi-resolution and multi-direction analysis. The color space used is HSV color space separable to hue, saturation, and value components, which are easily analyzed as showing characteristics similar to the human visual system. This experiment is conducted by comparing precision vs. recall of retrieval and feature vector dimensions. Images for experiments include Corel DB and VisTex DB; Corel_MR DB and VisTex_MR DB, which are transformed from the aforementioned two DBs to have multi-resolution images; and Corel_MD DB and VisTex_MD DB, transformed from the two DBs to have multi-direction images. According to the experimental results, the proposed method improves upon the existing methods in aspects of precision and recall of retrieval, and also reduces feature vector dimensions.

• 한국멀티미디어학회논문지
• /
• 제8권6호
• /
• pp.776-782
• /
• 2005

This paper presents a SPIHT image compression method using biorthogonal multi wavelets on [-1,1]. A family of biorthogonal scaling vectors is constructed using fractal interpolation function, and the associated biorthogonal multi wavelets are constructed. This paper uses biorthogonal multi wavelets to be supported in [-1,1] associated with biorthogonal scaling vectors to be supported in [-1,1]. The scaling vectors and wavelets remain biorthogonal when restricted to integer intervals, making them well suited for bounded domains. The experiment results of simulation of the proposed image compression using biorthogonal multiwavelets on [-1,1] based on SPIHT were found to be excellent PSNR for LENA and PEPPERS images except for BABOON image than already existing single wavelets and DGHM multi wavelets.

• 산업기술연구
• /
• 제30권A호
• /
• pp.25-29
• /
• 2010

In this paper, we implement a multi-touch system using FIDR. The implementation consists of hardware manufacture and development of image processing system. In the hardware system, touch screen, infrared LED placements and infrared camera are made. The image processing procedure is to extract each pointer's coordinates from image data and includes binary-coding, noise-elimination, labeling and calculation of mass center. From the implementation, we are able to make a multi-touch system with considerably lower cost than the existing ones.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2010년도 춘계학술대회
• /
• pp.589-590
• /
• 2010

All these images consist a considerable database. What's more, the semantic meanings of images are well presented by the surrounding text and links. But only a small minority of these images have precise assigned keyphrases, and manually assigning keyphrases to existing images is very laborious. Therefore it is highly desirable to automate the keyphrases extraction process. In this paper, we first introduce WWW image annotation methods, based on low level features, page tags, overall word frequency and local word frequency. Then we put forward our method of multi-cues integration image annotation. Also, show multi-cue image annotation method is more superior than other method through an experiment.

• 스마트미디어저널
• /
• 제12권9호
• /
• pp.45-59
• /
• 2023

Weakly supervised object localization (WSOL) is a task of localizing an object in an image using only image-level labels. Previous studies have followed the conventional class activation mapping (CAM) pipeline. However, we reveal the current CAM approach suffers from problems which cause original CAM could not capture the complete defects features. This work utilizes a convolutional neural network (CNN) pretrained on image-level labels to generate class activation maps in a multi-scale manner to highlight discriminative regions. Additionally, a vision transformer (ViT) pretrained was treated to produce multi-head attention maps as an auxiliary detector. By integrating the CNN-based CAMs and attention maps, our approach localizes defective regions without requiring bounding box or pixel-level supervision during training. We evaluate our approach on a dataset of apple images with only image-level labels of defect categories. Experiments demonstrate our proposed method aligns with several Object Detection models performance, hold a promise for improving localization.

• 반도체디스플레이기술학회지
• /
• 제22권3호
• /
• pp.149-154
• /
• 2023

This paper presents a novel approach to multi-focus image fusion using light field cameras. The proposed neural network, LFFCNN (Light Field Focus Convolutional Neural Network), is composed of three main modules: feature extraction, feature fusion, and feature reconstruction. Specifically, the feature extraction module incorporates SPP (Spatial Pyramid Pooling) to effectively handle images of various scales. Experimental results demonstrate that the proposed model not only effectively fuses a single All-in-Focus image from images with multi focus images but also offers more efficient and robust focus fusion compared to existing methods.

• 대한가정학회지
• /
• 제45권1호
• /
• pp.51-61
• /
• 2007

The purpose of this study was to examine if multi-shop's store image affects store loyalty and brand switching. Two hundred fifty females and males who have purchased fashion products in multi-shop participated in this survey. For data analysis, descriptive statistics, factor analysis, Pearson's correlation and regression analysis were used for this study. The results were as followed. First, respondents' the most favorite multi-shop was MUE, followed by Boon the shop and ABC mart. Second, store image was classified into four factors such as store atmosphere, service of store, store recognition and product variety. Store loyalty was classified into five factors such as emotional relationship, pursue of novelty, trust about salesperson, satisfaction about service, and active loyalty. Third, result revealed that 'product variety' and 'store atmosphere', 'store recognition', 'service of store' accounted for 39.6% of the explained varience in store loyalty, and 'store recognition' accounted for 4% of the explained varience in brand switching behavior, while 'trust about salesperson', 'pursue of novelty' accounted for 5% of the explained varience in brand switching behavior. Based on these results, multi-shop's fashion marketing strategy would be suggested.

• 산업융합연구
• /
• 제18권6호
• /
• pp.33-39
• /
• 2020

멀티스크린 영상시스템은 하나의 스크린에 상영되는 영상을 용도에 맞게 여러 화면으로 구성하여 상영할 수 있다. 관객들은 특별한 장치 없이 폭 넓은 실감 영상을 감상할 수 있어 강한 몰입감을 얻을 수 있다. 최근 멀티스크린을 이용한 실감 영상은 Screen X나 Escape와 같은 멀티 프로젝션 기술을 바탕으로 스토리와 정보 전달이 가능한 영화로 제작되고 있다. 또한, 디스플레이의 크기는 점점 대형화 추세이고 화질도 고해상도로 향상되고 있어 HD나 UHD급 디스플레이 여러대를 입체적으로 구성한 디지털 사이니지 형태로 발전을 가속하고 있으나 촬영된 영상을 사용하는 데는 화질열화로 인해 한계가 있다. 이번 연구는 1950년부터 현재까지 멀티스크린 영상의 발전사와 기술적 분석, 제작 방법에 대해 분석하고 멀티스크린을 이용한 영상 콘텐츠 상영 시 발생되는 화질 열화 최소화 방법에 대해 연구한다. 연구 결과를 바탕으로 촬영된 실사 영상은 디지털 사이니지와 같은 고해상도 영상 상영이 가능한 플랫폼에서 실감영상 구현이 가능할 것이라 기대한다.

• 한국의류학회지
• /
• 제40권1호
• /
• pp.12-25
• /
• 2016

Based on the self-congruity theory, this study investigated how congruity between multi-brand store image and consumers' self-image affect store attributes and consumer responses. A total of 331 questionnaires were used to analyze data. The results of research were: 1) 'Sophistication' as the congruity factor between store image and consumers' self-image affected 'utility', 'atmosphere', and 'design' among store attribute factors. Also, 'sincerity' influenced 'utility' as the store attribute factor. 2) 'Atmosphere' as the store attribute factor positively influenced consumers' emotional responses, and 'utility' and 'design' factors positively influenced consumers' cognitive responses. 3) Consumers' emotional responses had a positive impact on consumers' cognitive responses; in addition, consumers' emotional and cognitive responses had positive impacts on consumers' behavioral responses. 4) A-Land indicated higher scores on 'sophistication', 'atmosphere', and 'design' factors than ABC Mart. ABC Mart had shown higher scores on 'ruggedness' and 'utility' factors than A-Land. This study provides practical implications to develop effective marketing strategies to manage multi-brand stores.