• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.6
• /
• pp.175-186
• /
• 2004

We present a hybrid approach of texture-based method and connected component (CC)-based method for text extraction in complex images. Two primary methods, which are mainly utilized in this area, are sequentially merged for compensating for their weak points. An automatically constructed MLP-based texture classifier can increase recall rates for complex images with small amount of user intervention and without explicit feature extraction. CC-based filtering based on the shape information using NMF enhances the precision rate without affecting overall performance. As a result, a combination of texture and CC-based methods leads to not only robust but also efficient text extraction. We also enhance the processing speed by adopting appropriate region marking methods for each input image category.

• ALGAE
• /
• v.23 no.2
• /
• pp.141-150
• /
• 2008

Populations of Codium amplivesiculatum were found in association with rhodolith beds. We present population trends of C. amplivesiculatum from surveys conducted in the southwestern Gulf of California based on monthly sampling from June 1999 to August 2000. Surveys were conducted at 2 sites in the San Lorenzo Channel at depths of 8 - 12 m. The parameters measured were percent cover, biomass and structure in relation to size of the thalli. From each thallus the following measurements were taken: length and diameter, proportion of each thallus with reproductive structures and the degree of fragmentation (measured as number of small thalli). Mean biomass and cover varied seasonally with the maximum (1036.6 gm$^{-2}$ and 100% cover) present in June 2000 and the minimum in winter (0 gm$^{-2}$ and 0% cover). The longest fronds (3 m) occurred during July 1999 and August 2000, while the maximum diameters of 1.0 cm were present during October and November 1999. There were no significant linear relationships between length and diameter of the fronds, suggesting independent growth. Reproductive structures were common in June and July 1999 with monoic and dioic thalli present. Most fronds were monoic, and this is the first record of this feature for this species. Fragmentation occurred in September, and may be associated with stress from early gametangial reproduction and the combination of high temperatures and low nutrients. A hypothetical life cycle for the species is presented.

• Fisheries and Aquatic Sciences
• /
• v.12 no.4
• /
• pp.265-275
• /
• 2009

G Protein-coupled receptors (GPCRs) mediating wide ranges of physiological responses is one of the most attractive targets for drug development. Rhodopsin, a dim-light photoreceptor, has been extensively used as a model system for structural and functional study of GPCRs. Fish have rhodopsin finely-tuned to their habitats where the intensity and the wavelength of lights are changed depending on its water-depth. To study the detailed molecular characteristics of GPCR architecture and to understand the fishery light-sensing system, genes encoding rod opsins were isolated from fishes living under different photic environments. Full-length rod opsin genes were obtained by combination of PCR amplification and DNA walking strategy of genomic DNA isolated from olive flounder, P. olivaceus, Japanese eel, A. japonica, and Common carp C. carpio. Deduced amino acid sequences showed a typical feature of rod opsins including the sites for Schiffs base formation (Lys296) and its counter ion (Glu113), disulfide formation (Cys110 and Cys187), and palmitoylation (Cys322 and Cys323) although Cys322 is replaced by Phe in Japanese eel. Comparison of opsins by amino acid sequence alignment indicated the closest similarity between P. olivaceus and H. hippoglossus (94%), A. japonica and A. anguilla (98%), and C. carpio and C. auratus (95%), respectively.

• Journal of Korean Society for Geospatial Information Science
• /
• v.13 no.3 s.33
• /
• pp.33-40
• /
• 2005

The aim of remotely sensed data classification is to produce the best accuracy map of the earth surface assigning each pixel to its appropriate category of the real-world. The classification of satellite multi-spectral image data has become tool for generating ground cover map. Many classification methods exist. In this study, MLC(Maximum Likelihood Classification), ANN(Artificial neural network), SVM(Support Vector Machine), Naive Bayes classifier algorithms are compared using IKONOS image of the part of Dalsung Gun, Daegu area. Two preprocessing methods are performed-PCA(Principal component analysis), ICA(Independent Component Analysis). Boosting algorithms also performed. By the combination of appropriate feature selection pre-processing and classifier, the best results were obtained.

• Journal of the Korean Society of Costume
• /
• v.53 no.5
• /
• pp.13-22
• /
• 2003

The purposes of this study are to develop a Korean image clothing design for cultural products by using the literary artist´s works and to satisfy the demand of the local self-government and of Gwangju ㆍChonnam area´s small and medium-sized companies. The image of traditional Korean clothing can be expressed with adjectives like Pure, plain, rhythmical, neatness, plentiful, and calm etc. These attributes are similar to the characteristics of the literary artist´s style, which is expressed as suggestive, animated, spatial, possibility of stationary, serene and still. The features of the traditional Korean clothing design can be described as ´The Beauty of Elegance´, ´The Beauty of Plainness´, ´The Beauty of Asymmetric Balance´ and ´The Beauty of rhythm´. It can be expressed in the modern fashion design, which is developed through the combination of the most important feature like ´The Beauty of Space´ in the literary artist´s style. At the time of production we tried to express spatial beauty and plentiful taste of the literary artist's style in the clothing design of Korean image for cultural products by considering the balance between them. Six works of the literary artist were selected which satisfied the purpose of the developed clothing design item. The results were nine clothing designs of Korean image, like a skirt suit made up of a blouse and a skirt, a pants suit made up of pants and a jacket, two evening dresses with a shawl, and also T-shirts and one-pieces etc.

• Journal of the Korean Society of Costume
• /
• v.63 no.1
• /
• pp.1-15
• /
• 2013

This study deals with the aesthetic characteristics in Gareth Pugh's fashion film. As fashion films are becoming a newly emerging communication tool in fashion, the aim of this study will be to build a database on these films. For the specific method in this work, literature reviews were performed along with the empirical study of seven pieces of fashion film. The results of the study showed that Gareth Pugh released the fashion films from 2008 to 2012, and he collaborated with SHOW Studio member and photographer Nick Knight and Ruth Hogben. Common themes in their films were fear, fantasy, dynamics, and going off-limits, and these can be summed up in the following ways. First, a sense of fear was expressed via black colors, bizarre mechanical sounds, decalcomania shape in abstract images, and transformed physical body images. Second, an image of fantasy was presented with diverse expression techniques including colorful light presentations and abstract image setups from repeatedly changed image reproduction. Third, dynamic feature was denoted with a combination of fast development of images and quick reproduced layouts, as well as quick beat sounds, and big and powerful action movements. Fourth, characteristic of going off-limits was reflected, by showing the collections with a communication tool called fashion film instead of existing collection concept.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.3
• /
• pp.55-62
• /
• 2017

Non-negative matrix factorization (NMF) is one of the typical parts-based representation in which images are expressed as a linear combination of basis vectors that show the lcoal features or objects in the images. In this paper, we represent face images using various NMF methods and recognize their face identities based on extracted features using a learning vector quantization. We analyzed the various NMF methods by comparing extracted basis vectors. Also we confirmed the availability of NMF to the face recognition by verification of recognition rate of the various NMF methods.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.17 no.1
• /
• pp.199-206
• /
• 2017

Environmental noise can degrade the performance of speech recognition system. This paper presents a procedure for performing cepstrum based feature compensation to make recognition system robust to noise. The approach is based on direct compensation of spectral tilt to remove effects of additive noise. The noise compensation scheme operates in the cepstral domain by means of calculating spectral tilt of the log power spectrum. Spectral compensation is applied in combination with SNR-dependent cepstral mean compensation. Experimental results, in the presence of white Gaussian noise, subway noise and car noise, show that the proposed compensation method achieves substantial improvements in recognition accuracy at various SNR's.

• Journal of Institute of Control, Robotics and Systems
• /
• v.19 no.6
• /
• pp.513-519
• /
• 2013

The current vision-based approaches for emotion recognition, such as facial expression analysis, have many technical limitations in real circumstances, and are not suitable for applications that use them solely in practical environments. In this paper, we propose an approach for emotion recognition by combining extrinsic representations and intrinsic activities among the natural responses of humans which are given specific imuli for inducing emotional states. The intrinsic activities can be used to compensate the uncertainty of extrinsic representations of emotional states. This combination is done by using PRMs (Probabilistic Relational Models) which are extent version of bayesian networks and are learned by greedy-search algorithms and expectation-maximization algorithms. Previous research of facial expression-related extrinsic emotion features and physiological signal-based intrinsic emotion features are combined into the attributes of the PRMs in the emotion recognition domain. The maximum likelihood estimation with the given dependency structure and estimated parameter set is used to classify the label of the target emotional states.

• Journal of Korea Multimedia Society
• /
• v.16 no.5
• /
• pp.601-609
• /
• 2013

One of the most serious factors constraining the next generation cellular mobile consumer communication systems will be the severe co-channel interference experienced at the cell edge. Such a capacity-degrading impairment combined with the limited available spectrum resource makes it essential to develop more spectrally efficient solutions to enhance the system performance and enrich the mobile user's application services. This paper proposes a unique hybrid method of frequency hopping (FH) and subcarrier-reuse-partitioning that can maximize the system capacity by efficiently utilizing the available spectrum while at the same time reduce the co-channel interference effect. The main feature of the proposed method is that it applies an optimal combination of different frequency reuse factors (FRF) and FH-subcarrier allocation patterns into the partitioned cell regions. From the simulation results, it is shown that the proposed method can achieve the optimum number of subcarrier subsets according to the frequency-reuse distance and results in better performance than the fixed FRF methods, for a given partitioning arrangement. The results are presented in the context of both blocking probability and BER performances. It will also be shown how the proposed scheme is well suited to FH-OFDMA based cellular systems aiming at low co-channel interference performance and optimized number of subcarriers.