• Title/Summary/Keyword: Five features

Search Result 1,113, Processing Time 0.033 seconds

EDMFEN: Edge detection-based multi-scale feature enhancement Network for low-light image enhancement

  • Canlin Li;Shun Song;Pengcheng Gao;Wei Huang;Lihua Bi
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • v.18 no.4
    • /
    • pp.980-997
    • /
    • 2024
  • To improve the brightness of images and reveal hidden information in dark areas is the main objective of low-light image enhancement (LLIE). LLIE methods based on deep learning show good performance. However, there are some limitations to these methods, such as the complex network model requires highly configurable environments, and deficient enhancement of edge details leads to blurring of the target content. Single-scale feature extraction results in the insufficient recovery of the hidden content of the enhanced images. This paper proposed an edge detection-based multi-scale feature enhancement network for LLIE (EDMFEN). To reduce the loss of edge details in the enhanced images, an edge extraction module consisting of a Sobel operator is introduced to obtain edge information by computing gradients of images. In addition, a multi-scale feature enhancement module (MSFEM) consisting of multi-scale feature extraction block (MSFEB) and a spatial attention mechanism is proposed to thoroughly recover the hidden content of the enhanced images and obtain richer features. Since the fused features may contain some useless information, the MSFEB is introduced so as to obtain the image features with different perceptual fields. To use the multi-scale features more effectively, a spatial attention mechanism module is used to retain the key features and improve the model performance after fusing multi-scale features. Experimental results on two datasets and five baseline datasets show that EDMFEN has good performance when compared with the stateof-the-art LLIE methods.

Pharmacophore Hypothesis for Atypical Antipsychotics

  • Sekhar, Kondapalli Venkata Gowri Chandra;Vyas, Devambhatla Ravi Kumar;Nagesh, Hunsur Nagendra;Rao, Vajja Sambasiva
    • Bulletin of the Korean Chemical Society
    • /
    • v.33 no.9
    • /
    • pp.2930-2936
    • /
    • 2012
  • A three-dimensional pharmacophore hypothesis was developed for atypical antipsychotics in order to map common structural features of highly active compounds by using HipHop in CATALYST program. The pharmacophore hypotheses were generated using 12 compounds as training set and validated using 11 compounds as test set. The most predictive hypothesis (Hypo1) comprises five features viz. two hydrophobic regions, two hydrogen bond acceptor lipid and one aromatic ring. In the absence of information like crystallized structure of 5-$HT_{2A}$ receptor and binding mode of antipsychotics with 5-$HT_{2A}$ receptor, this hypothesis will serve as a potentially valuable tool in the design of novel atypical antipsychotics acting primarily at 5-$HT_{2A}$ and $D_2$ receptors.

Automated Classification of Audio Genre using Sequential Forward Selection Method

  • Lee Jong Hak;Yoon Won lung;Lee Kang Kyu;Park Kyu Sik
    • Proceedings of the IEEK Conference
    • /
    • 2004.08c
    • /
    • pp.768-771
    • /
    • 2004
  • In this paper, we propose a content-based audio genre classification algorithm that automatically classifies the query audio into five genres such as Classic, Hiphop, Jazz, Rock, Speech using digital signal processing approach. From the 20 second query audio file, 54 dimensional feature vectors, including Spectral Centroid, Rolloff, Flux, LPC, MFCC, is extracted from each query audio. For the classification algorithm, k-NN, Gaussian, GMM classifier is used. In order to choose optimum features from the 54 dimension feature vectors, SFS (Sequential Forward Selection) method is applied to draw 10 dimension optimum features and these are used for the genre classification algorithm. From the experimental result, we verify the superior performance of the SFS method that provides near $90{\%}$ success rate for the genre classification which means $10{\%}$-$20{\%}$ improvements over the previous methods

  • PDF

Foreign Body Granulomas after the Use of Dermal Fillers: Pathophysiology, Clinical Appearance, Histologic Features, and Treatment

  • Lee, Jeong Min;Kim, Yu Jin
    • Archives of Plastic Surgery
    • /
    • v.42 no.2
    • /
    • pp.232-239
    • /
    • 2015
  • A foreign body granuloma is a non-allergic chronic inflammatory reaction that is mainly composed of multinucleated giant cells. Foreign body granulomas may occur after the administration of any dermal filler. Factors such as the volume of the injection, impurities present in the fillers, and the physical properties of fillers affect granuloma formation. The formation of granulomas involves five phases: protein adsorption, macrophage adhesion, macrophage fusion, and crosstalk. The clinical and pathologic features of granulomas vary depending on the type of filler that causes them. Foreign body granulomas can be treated effectively with intralesional corticosteroid injections. Surgical excisions of granulomas tend to be incomplete because granulomas have ill-defined borders and moreover, surgical excisions may leave scars and deformities.

Text Classification for Patents: Experiments with Unigrams, Bigrams and Different Weighting Methods

  • Im, ChanJong;Kim, DoWan;Mandl, Thomas
    • International Journal of Contents
    • /
    • v.13 no.2
    • /
    • pp.66-74
    • /
    • 2017
  • Patent classification is becoming more critical as patent filings have been increasing over the years. Despite comprehensive studies in the area, there remain several issues in classifying patents on IPC hierarchical levels. Not only structural complexity but also shortage of patents in the lower level of the hierarchy causes the decline in classification performance. Therefore, we propose a new method of classification based on different criteria that are categories defined by the domain's experts mentioned in trend analysis reports, i.e. Patent Landscape Report (PLR). Several experiments were conducted with the purpose of identifying type of features and weighting methods that lead to the best classification performance using Support Vector Machine (SVM). Two types of features (noun and noun phrases) and five different weighting schemes (TF-idf, TF-rf, TF-icf, TF-icf-based, and TF-idcef-based) were experimented on.

An Analysis of Mobile Virtual Manipulatives Apps for the Teaching of Elementary School Mathematics (초등학교 수학의 교수를 위한 모바일 가상조작물 앱 분석)

  • Shin, Mikyung
    • Journal of Korea Multimedia Society
    • /
    • v.20 no.6
    • /
    • pp.935-949
    • /
    • 2017
  • The purpose of this study was to analyze the characteristics of virtual manipulatives apps that can be used to teach students struggling to learn mathematics. To achieve this goal, ten general characteristics of 23 virtual manipulatives apps were evaluated. The instructional, interface, and interactive design features of apps were also evaluated on five-point scale ratings of 18 items. In addition, SPSS frequency analysis and the correlation between each feature was analyzed. Frequently presented instructional contents among 23 virtual manipulatives apps were geometry, arithmetic operation, number concept and measurement. The frequently presented level of instructional contents was lower grade elementary school and kindergarten age. The frequently presented instructional type was the simulation. Regarding the design features, instructional design was rated as the highest (mean = 3.7); interactive design (mean = 3.6) and interface design (mean = 3.3) were also rated higher than neural. In addition, as the learning strategy was appropriately presented, it was evaluated that there was less screen linkage and content error.

Interactive Spatial Augmented Reality Book on Cultural Heritage of Myanmar

  • Hta, Aye Chan Zay;Lee, Yunli
    • Journal of information and communication convergence engineering
    • /
    • v.18 no.2
    • /
    • pp.69-74
    • /
    • 2020
  • Myanmar, also known as Burma, has a rich cultural heritage, and its historical tourist attractions well known around the world. Therefore, we designed and developed an interactive spatial augmented reality (iSAR) book on the cultural heritage of Myanmar. This iSAR book has total of 18 pages with rich media content including videos, animations, audio, and images featuring the cultural heritage of Myanmar in a digital format. In addition to virtual content, navigational features such as virtual buttons and touch-based hand gestures were implemented using Leap Motion and VVVV. Therefore, the developed iSAR book allows virtual content and navigational features to merge seamlessly into a physical book. Five participants were recruited to evaluate the prototype iSAR book, and interviews were conducted to gather their feedback based on its immersive qualities. Thus, the developed iSAR book on Myanmar effectively shares the cultural heritage of Myanmar, and ultimately allows users to explore and gain more insight into the country.

A Study on the Performance Analysis of Content-based Image & Video Retrieval Systems (내용기반 이미지 및 비디오 검색 시스템 성능분석에 관한 연구)

  • Kim, Seong-Hee
    • Journal of the Korean BIBLIA Society for library and Information Science
    • /
    • v.15 no.2
    • /
    • pp.97-115
    • /
    • 2004
  • The paper examined the concepts and features of content-based Image and Video retrieval systems. It then analyzed the retrieval performance of on five content_based retrieval systems in terms of usability and retrieval features. The results showed that the combination of content_based retrieval techniques and meta-data based retrieval will be able to improve the retrieval effectiveness.

  • PDF

An Acoustic Study of the Pronunciation of English Pitch. Accents Uttered by Korean Speakers (한국인의 영어피치악센트 발음에 관한 연구)

  • Koo, Hee-San
    • Speech Sciences
    • /
    • v.10 no.2
    • /
    • pp.223-236
    • /
    • 2003
  • The purpose of this experimental study is to investigate characteristics of English pitch accents uttered by Korean speakers. Six English sentences were uttered five times by fifteen male undergraduate and graduate students from three groups, Seoul, Yongnam and Honam dialect speakers. We compared the subjects' data with the data of a native speaker of English as model pronunciation of English pitch accents. Acoustic features(Fo, duration, amplitude) were measured from sound spectrograms made by the PC Works. Results showed that (1) acoustic features of English pitch accents are Fo and duration for the native speaker and Korean speakers altogether, (2) Seoul dialect speakers uttered English pitch accents more similarly to the English native speaker than the other dialect speakers and (3) Korean speakers generally have difficulties in pronouncing L* accents. It appears that Korean speakers have more problems in pronouncing L* accents than H* accents.

  • PDF

Experimental Study on the Mean Flow Characteristics of Forward-Curved Centrifugal Fans

  • Kwon, Eui-Yong;Cho, Nam-Hyo
    • Journal of Mechanical Science and Technology
    • /
    • v.15 no.12
    • /
    • pp.1728-1738
    • /
    • 2001
  • Measurements have been made in an automotive HVAC b1ower for two different centrifugal fans. This work is directed at improving the performance of a conventional forward-curved centrifugal fan for a given small blower casing. Mean velocities and pressure have been measured using a miniature five-hole probe and a pressure scanning unit connected to an online data acquisition system. First, we obtained the fan performance versus flow rates showing a significant attenuation of unstable nature achieved with the new fan rotor in the surging operation range. Second, aerodynamic characterizations were carried out by investigating the velocity and pressure fields in the casing flow passage for different fan operating conditions. The measurements stowed that performance coefficients are strongly influenced by flow characteristics at the throat region. The main flow features ware common in both fans, but improved performance is achieved with tole new fan rotor, particularly in lower flow rate legions. Based on the measured results, design improvements were carried out in an acceptable operation range, which gave considerable insight into what features of flow behavior ware most important.

  • PDF