• Title/Summary/Keyword: semantic distance

Search Result 83, Processing Time 0.022 seconds

On the subjective response caused by impulse sounds produced by leisure shooting (레저용 사격 소음에 대한 주관적 반응)

  • Kim, Deuk-Sung;Chang, Seo-Il
    • Proceedings of the Korean Society for Noise and Vibration Engineering Conference
    • /
    • 2008.04a
    • /
    • pp.714-720
    • /
    • 2008
  • This research presents a laboratory study about an subjective response of impulsive sound caused by leisure shooting. The sources are sampled from outdoor noise and their levels range from 40 to 75 dB at the interval of 5dB. The noise unit is based on A-weighted sound exposure level (ASEL; $L_{AE}$). To make equal ASEL of outdoor noise, finite impulse response (FIR) filter is applied to the originally sampled source to include the effect of distance attenuation. The evaluation method of the jury test adopted a Semantic Difference(SD) Method. In the result of the jury test for impulsive noise, the mean response rating expressed a linear relation and %HA(percent highly annoyed) displayed a exponential growth relation.

  • PDF

Assembly performance evaluation method for prefabricated steel structures using deep learning and k-nearest neighbors

  • Hyuntae Bang;Byeongjun Yu;Haemin Jeon
    • Smart Structures and Systems
    • /
    • v.32 no.2
    • /
    • pp.111-121
    • /
    • 2023
  • This study proposes an automated assembly performance evaluation method for prefabricated steel structures (PSSs) using machine learning methods. Assembly component images were segmented using a modified version of the receptive field pyramid. By factorizing channel modulation and the receptive field exploration layers of the convolution pyramid, highly accurate segmentation results were obtained. After completing segmentation, the positions of the bolt holes were calculated using various image processing techniques, such as fuzzy-based edge detection, Hough's line detection, and image perspective transformation. By calculating the distance ratio between bolt holes, the assembly performance of the PSS was estimated using the k-nearest neighbors (kNN) algorithm. The effectiveness of the proposed framework was validated using a 3D PSS printing model and a field test. The results indicated that this approach could recognize assembly components with an intersection over union (IoU) of 95% and evaluate assembly performance with an error of less than 5%.

Design and Implementation of Domain Ontology to Overcome Conceptual Heterogeneity in Annotation-based Image Retrieval (주석기반 이미지 검색에서 개념적 이질성 극복을 위한 도메인 온톨로지 설계 및 구현)

  • Kim Won-Pil;Kim Pan-Koo
    • Journal of Internet Computing and Services
    • /
    • v.4 no.4
    • /
    • pp.1-8
    • /
    • 2003
  • As the multimedia information retrieval system is advanced, the study of multimedia information retrieval is changing the method of low-level content based image retrieval to the semantical concept based retrieval. in this paper, we apply the theory of ontology to overcome the conceptual heterogeneity in the annotation based image retrieval. And we solve the some problems that happen when the ontology apply. As a result of our study, we try to apply the domain ontology to settle the conceptual heterogenity. In the experimental result, we knew that the semantic distance among the words is pretty dose when we apply the domain ontology than the wordnet. And in this paper, we show the possibility of the semantic image retrieval as we apply the domain ontology in the annotation based image retrieval.

  • PDF

Color Image Segmentation Using Adaptive Quantization and Sequential Region-Merging Method (적응적 양자화와 순차적 병합 기법을 사용한 컬러 영상 분할)

  • Kwak, Nae-Joung;Kim, Young-Gil;Kwon, Dong-Jin;Ahn, Jae-Hyeong
    • Journal of Korea Multimedia Society
    • /
    • v.8 no.4
    • /
    • pp.473-481
    • /
    • 2005
  • In this paper, we propose an image segmentation method preserving object's boundaries by using the number of quantized colors and merging regions using adaptive threshold values. First of all, the proposed method quantizes an original image by a vector quantization and the number of quantized colors is determined differently using PSNR each image. We obtain initial regions from the quantized image, merge initial regions in CIE Lab color space and RGB color space step by step and segment the image into semantic regions. In each merging step, we use color distance between adjacent regions as similarity-measure. Threshold values for region-merging are determined adaptively according to the global mean of the color difference between the original image and its split-regions and the mean of those variations. Also, if the segmented image of RGB color space doesn't split into semantic objects, we merge the image again in the CIE Lab color space as post-processing. Whether the post-processing is done is determined by using the color distance between initial regions of the image and the segmented image of RGB color space. Experiment results show that the proposed method splits an original image into main objects and boundaries of the segmented image are preserved. Also, the proposed method provides better results for objective measure than the conventional method.

  • PDF

Research on the Visual Characteristics of a Representative View of the Skyline; - Referring to Landscape Assessment of Mt. Mudeung from Various Viewpoints - (도시 배후 산 지형 스카이라인 경관의 조망 특성과 경관 대표성 평가 - 시점 위치에 따른 무등산 조망경관 분석을 중심으로 -)

  • Cho, Tong-Buhm
    • Journal of the Korean Institute of Landscape Architecture
    • /
    • v.35 no.6
    • /
    • pp.84-96
    • /
    • 2008
  • This research investigated the landscape characteristics of the skyline and the cognitive characteristics of Mt. Mudeung (1,186m) from various viewpoints. Mt. Mudeung, the representative landscape of Gwangju City, has been recognized as a natural landmark and theme of paintings. By analyzing the perspective from 32 points with a digital terrain model, some landscape indices of the skyline were derived and the relationships are discussed. Assessment of the semantic differential scale with 21 adjective variables and representativeness to 15 landscape photographs of the mountain were accomplished. 1. Through regression analysis of the skyline indices, significant relationships were found between them the angle from the visual axis and number of skyline jumps, the vertical angle fluctuation and number of jumps per degree, the visual depth fluctuation and vertical angle fluctuation of skyline, and between the vertical angle mean and number of jumps per degree. Meaningful relations were found between the number of jumps of skyline to number of jumps per degree and the angle from visual axis to visual distance. However, in the representative assessment no difference was found on the angle from visual axis of viewpoints. On the other hand, it seemed to relate representativeness with visual clarity based on visual distance. 2. We found 4 factors "familiarity", "fluctuation of skylines", "openness", and "feeling of texture" in the results of factor analysis of semantic differential assessment. When considering the results of assessment for representativeness, adjective words for familiarity and openness seemed to have a close assessment. Specifically, the research showed that the landscape representation was highly assessed in a view which could be seen from the higher parts to the lower part of hills. This result indicates that the management of viewpoints which could get a scene from intermediate to distant, and locating a high elevation is important. 3. In the picturesque expression of Mt. Mudeung, various impressions from the different points, a skyline based on the top of Mt. Mudeung and a mono structure by overlapping hills were common characteristics. These common characteristics were also partially found through the analysis of topographical landscape indices and landscape images. Therefore, the viewpoints for the representative landscape management should be selected in natural or open spaces.

What Quality Factors Affect to the e-Learning Performance (e-러닝 성과에 영향을 미치는 품질요인에 관한 연구)

  • Kim, Sung-Gyun;Sung, Hang-Nam;Jeong, Dae-Yul
    • The Journal of Information Systems
    • /
    • v.16 no.1
    • /
    • pp.201-230
    • /
    • 2007
  • Recently, the growth of e-Learning systems and its related information technology has presented a unique challenge for both schools and industry. It would make an extremely phenomenal paradigm shift in the educational method and practice. Methods of assessing the quality of e-teaming services and contents are critical issue in both practice and research. Moreover, many researchers are interested in what qualify factors more affect to the Performance of e-Learning service. Nevertheless, service quality is a construct that is difficult to define and measure. e-Learning services are composed of many factors, and they are more complicated than the traditional education services because they we performed on the distance basis and the many platforms of IT infrastructure. The purposes of our research are to classify the e-Learning service dimension and identify their factors, to develop the measurement of the factors, and finally to test empirically their relationship between the service factors and e-Learning service performance. For the development of the service factors we considered SERVQUAL model and SERVPERF model which were developed in the service marketing area. The SERVQUAL model was more fitted to the e-Learning services than the latter. From that we derived several factors that fit to our research domain, ie, tangibles, access, reliability, credibility, security, responsiveness, assurance, empathy. We combined three factors of them(reliability, credibility, security) into a factor, system stability for the semantic simplicity, and divided responsiveness factor into system operator responsiveness and teacher responsiveness as the entity based dimension classification. In the e-Learning services research, Most researcher are mentioned the quality factors of contents, so we added to two contents quality factors, ie, contents production method and richness of contents itself. We examined the relationship between the service quality factors and e-Learning performance(student satisfaction and service reuse intention). As result three quality factors(contents production method, teacher responsiveness, empathy) significantly affected student satisfaction. To the other performance variable, ie, service reuse intention, the teacher related quality factors(such as teacher responsiveness, assurance, empathy) affected only. In conclusion, even in the on-line distance teaming, the teacher's role md earnestness is as important as ever.

  • PDF

Terminology Recognition System based on Machine Learning for Scientific Document Analysis (과학 기술 문헌 분석을 위한 기계학습 기반 범용 전문용어 인식 시스템)

  • Choi, Yun-Soo;Song, Sa-Kwang;Chun, Hong-Woo;Jeong, Chang-Hoo;Choi, Sung-Pil
    • The KIPS Transactions:PartD
    • /
    • v.18D no.5
    • /
    • pp.329-338
    • /
    • 2011
  • Terminology recognition system which is a preceding research for text mining, information extraction, information retrieval, semantic web, and question-answering has been intensively studied in limited range of domains, especially in bio-medical domain. We propose a domain independent terminology recognition system based on machine learning method using dictionary, syntactic features, and Web search results, since the previous works revealed limitation on applying their approaches to general domain because their resources were domain specific. We achieved F-score 80.8 and 6.5% improvement after comparing the proposed approach with the related approach, C-value, which has been widely used and is based on local domain frequencies. In the second experiment with various combinations of unithood features, the method combined with NGD(Normalized Google Distance) showed the best performance of 81.8 on F-score. We applied three machine learning methods such as Logistic regression, C4.5, and SVMs, and got the best score from the decision tree method, C4.5.

Small Sample Face Recognition Algorithm Based on Novel Siamese Network

  • Zhang, Jianming;Jin, Xiaokang;Liu, Yukai;Sangaiah, Arun Kumar;Wang, Jin
    • Journal of Information Processing Systems
    • /
    • v.14 no.6
    • /
    • pp.1464-1479
    • /
    • 2018
  • In face recognition, sometimes the number of available training samples for single category is insufficient. Therefore, the performances of models trained by convolutional neural network are not ideal. The small sample face recognition algorithm based on novel Siamese network is proposed in this paper, which doesn't need rich samples for training. The algorithm designs and realizes a new Siamese network model, SiameseFacel, which uses pairs of face images as inputs and maps them to target space so that the $L_2$ norm distance in target space can represent the semantic distance in input space. The mapping is represented by the neural network in supervised learning. Moreover, a more lightweight Siamese network model, SiameseFace2, is designed to reduce the network parameters without losing accuracy. We also present a new method to generate training data and expand the number of training samples for single category in AR and labeled faces in the wild (LFW) datasets, which improves the recognition accuracy of the models. Four loss functions are adopted to carry out experiments on AR and LFW datasets. The results show that the contrastive loss function combined with new Siamese network model in this paper can effectively improve the accuracy of face recognition.

Reliability measure improvement of Phoneme character extract In Out-of-Vocabulary Rejection Algorithm (미등록어 거절 알고리즘에서 음소 특성 추출의 신뢰도 측정 개선)

  • Oh, Sang-Yeob
    • Journal of Digital Convergence
    • /
    • v.10 no.6
    • /
    • pp.219-224
    • /
    • 2012
  • In the communication mobile terminal, Vocabulary recognition system has low recognition rates, because this problems are due to phoneme feature extract from inaccurate vocabulary. Therefore they are not recognize the phoneme and similar phoneme misunderstanding error. To solve this problem, this paper propose the system model, which based on the two step process. First, input phoneme is represent by number which measure the distance of phonemes through phoneme likelihood process. next step is recognize the result through the reliability measure. By this process, we minimize the phoneme misunderstanding error caused by inaccurate vocabulary and perform error correction rate for error provrd vocabulary using phoneme likelihood and reliability. System performance comparison as a result of recognition improve represent 2.7% by method using error pattern learning and semantic pattern.

Assessing Landscape Impacts of Apartment Complex on Suburban Hilly Openspace; Multilateral Approach by Analysis of Physical Landscape Variables and Eye Fixation Movements (도시주변 능선녹지를 배경으로 하는 아파트 경관의 시각적 영향 - 물리적 경관변수 및 와시점분석에 의한 다각적 접근-)

  • Choi, Yun;Cho, Tong-Buhm
    • Journal of the Korean Institute of Landscape Architecture
    • /
    • v.22 no.2
    • /
    • pp.81-103
    • /
    • 1994
  • In recent years, the visual characteristics of natural open space and greenbelt surrounding the urban landscapes have been changed with sprawling of residential areas and highrised residential buildings. Since these natural areas being the background element of residential areas are topographically sloped mountains in many cities. It is easy to be seen in the distance and it is important to preserve these areas as a visual infrastructure of the urban landscape. The purposes of this study are to extract the factors of landscape impact evaluation for these areas and to clarify the physical landscape variables representing these factors, and to infer the visual-perceptional relationships between image and landscape variables. As results, conceptional three factors were extracted with semantic differential evaluation to classified 18 landscape slide, and three regression models were established with factor score of landscapes and physical variables measured in photographs. On the basis of these relationships, visual-perceptional characteristics were discussed by analyzing the data form eye-movement recording to each of landscapes. The factors of "spatial unfolding of backdropped hilly greenspace", "horizontal quence of residential buildings", and "landscape complexity" prove to be important. And it prove important variables of "skyline of mountainous ridge" and "visual edge of building structure" in regression models and eye fixation movements.

  • PDF