• Korean Journal of Computational Design and Engineering
• /
• v.13 no.6
• /
• pp.403-411
• /
• 2008

The alignment of facial images is crucial for 2D face recognition. This is the same to facial meshes for 3D face recognition. Most of the 3D face recognition methods refer to 3D alignment but do not describe their approaches in details. In this paper, we focus on describing an automatic 3D alignment in viewpoint of quantitative analysis. This paper presents a framework of 3D face alignment and normalization based on feature points obtained by Active Shape Models (ASMs). The positions of eyes and mouth can give possibility of aligning the 3D face exactly in three-dimension space. The rotational transform on each axis is defined with respect to the reference position. In aligning process, the rotational transform converts an input 3D faces with large pose variations to the reference frontal view. The part of face is flopped from the aligned face using the sphere region centered at the nose tip of 3D face. The cropped face is shifted and brought into the frame with specified size for normalizing. Subsequently, the interpolation is carried to the face for sampling at equal interval and filling holes. The color interpolation is also carried at the same interval. The outputs are normalized 2D and 3D face which can be used for face recognition. Finally, we carry two sets of experiments to measure aligning errors and evaluate the performance of suggested process.

• Journal of Internet Computing and Services
• /
• v.18 no.5
• /
• pp.47-53
• /
• 2017

In past few years, high-throughput sequencing, big-data generation, cloud computing, and computational biology are revolutionary. RNA sequencing is emerging as an attractive alternative to DNA microarrays. And the methods for constructing Gene Regulatory Network (GRN) from RNA-Seq are extremely lacking and urgently required. Because GRN has obtained substantial observation from genomics and bioinformatics, an elementary requirement of the GRN has been to maximize distinguishable genes. Despite of RNA sequencing techniques to generate a big amount of data, there are few computational methods to exploit the huge amount of the big data. Therefore, we have suggested a novel gene selection algorithm combining Support Vector Machines and Intensity-dependent normalization, which uses log differential expression ratio in RNAseq. It is an extended variation of support vector machine recursive feature elimination (SVM-RFE) algorithm. This algorithm accomplishes minimum relevancy with subsets of Big-Data, such as NCBI-GEO. The proposed algorithm was compared to the existing one which uses gene expression profiling DNA microarrays. It finds that the proposed algorithm have provided as convenient and quick method than previous because it uses all functions in R package and have more improvement with regard to the classification accuracy based on gene ontology and time consuming in terms of Big-Data. The comparison was performed based on the number of genes selected in RNAseq Big-Data.

• The Journal of the Acoustical Society of Korea
• /
• v.33 no.1
• /
• pp.68-74
• /
• 2014

In a previous work, we proposed a novel approach of histogram equalization using a supplement set which is composed of centroids of Fuzzy C-Means of the background utterances. The performance of the proposed method is affected by the size of the supplement set, but it is difficult to find the best size at the point of recognition. In this paper, we propose a histogram equalization using a supplement set for majority voting based speaker identification. The proposed method identifies test utterances using a majority voting on the histogram equalization methods with various sizes of supplement sets. The proposed method is compared with the conventional feature normalization methods such as CMN(Cepstral Mean Normalization), MVN(Mean and Variance Normalization), and HEQ(Histogram Equalization) and the histogram equalization method using a supplement set.

• The Journal of the Acoustical Society of Korea
• /
• v.22 no.4
• /
• pp.299-307
• /
• 2003

There have been many studies on speaker normalization which aims to minimize the effects of speaker's vocal tract length on the recognition performance of the speaker independent speech recognition system. In this paper, we propose a simple vector quantizer based linear warping speaker normalization method based on the observation that the vector quantizer can be successfully used for speaker verification. For this purpose, we firstly generate an optimal codebook which will be used as the basis of the speaker normalization, and then the warping factor of the unknown speaker will be extracted by comparing the feature vectors and the codebook. Finally, the extracted warping factor is used to linearly warp the Mel scale filter bank adopted in the course of MFCC calculation. To test the performance of the proposed method, a series of recognition experiments are conducted on discrete HMM with thirteen mono-syllabic Korean number utterances. The results showed that about 29% of word error rate can be reduced, and that the proposed warping factor extraction method is useful due to its simplicity compared to other line search warping methods.

• The KIPS Transactions:PartB
• /
• v.12B no.4 s.100
• /
• pp.437-442
• /
• 2005

For the purpose of speaker normalization in speaker independent speech recognition systems, experiments are conducted on a method based on Gaussian mixture model(GMM). The method, which is an improvement of the previous study based on vector quantizer, consists of modeling the probability distribution of canonical feature vectors by a GMM with an appropriate number of clusters, and of estimating the warp factor of a test speaker by making use of the obtained probabilistic model. The purpose of this study is twofold: improving the existing ML based methods, and comparing the performance of what is called 'soft decision' method with that of the previous study based on vector quantizer. The effectiveness of the proposed method is investigated by recognition experiments on the TIMIT corpus. The experimental results showed that a little improvement could be obtained tv adjusting the number of clusters in GMM appropriately.

• Journal of Korea Multimedia Society
• /
• v.23 no.8
• /
• pp.977-985
• /
• 2020

Aiming at the problem of complex feature extraction and low accuracy in human action recognition, this paper proposed a network structure combining batch normalization algorithm with GoogLeNet network model. Applying Batch Normalization idea in the field of image classification to action recognition field, it improved the algorithm by normalizing the network input training sample by mini-batch. For convolutional network, RGB image was the spatial input, and stacked optical flows was the temporal input. Then, it fused the spatio-temporal networks to get the final action recognition result. It trained and evaluated the architecture on the standard video actions benchmarks of UCF101 and HMDB51, which achieved the accuracy of 93.42% and 67.82%. The results show that the improved convolutional neural network has a significant improvement in improving the recognition rate and has obvious advantages in action recognition.

• Phonetics and Speech Sciences
• /
• v.4 no.2
• /
• pp.79-86
• /
• 2012

In this paper, we propose a novel approach to improve histogram equalization for speaker identification. Our method collects all speech features of UBM training data to make a reference distribution. The ranks of the feature vectors are calculated in the sorted list of the collection of the UBM training data and the test data. We use the ranks to perform order-based histogram equalization. The proposed method improves the accuracy of the speaker recognition system with short utterances. We use four kinds of speech databases to evaluate the proposed speaker recognition system and compare the system with cepstral mean normalization (CMN), mean and variance normalization (MVN), and histogram equalization (HEQ). Our system reduced the relative error rate by 33.3% from the baseline system.

• Proceedings of the IEEK Conference
• /
• 2005.11a
• /
• pp.367-370
• /
• 2005

This paper proposes a novel method for content-based watermarking based on feature points of an image. At each feature point, watermark is embedded after affine normalization according to the local characteristic scale and orientation. The characteristic scale is the scale at which the normalized scale-space representation of an image attains a maximum value, and the characteristic orientation is the angle of the principal axis of an image. By binding watermarking with the local characteristics of an image, resilience against affine transformations can be obtained. Experimental results show that the proposed method is robust against various image processing steps including affine transformations, cropping, filtering, and JPEG compression.

• Proceedings of the KIEE Conference
• /
• 1991.07a
• /
• pp.900-904
• /
• 1991

In this paper, our study for the recognition of Hand-written Korean characters, Arabic numbers and alphabets by neural netwoks. This System extracts feature of character by using the MESH feature point of handwritten character, Arabic numbers and alphabets. To reduce the input image data, features are extracted from each input images. A MLP(multi-layer perceptron) with one hidden layer was trained with a modified BEP(back error propagation) algorithm. This method extracts feature sets of the characters directly from the scanner and can enhance computation speed without using the special preprocesses such as size normalization, smoothing, and thinning.

• Journal of Information Processing Systems
• /
• v.7 no.3
• /
• pp.425-434
• /
• 2011

The success of iris recognition depends mainly on two factors: image acquisition and an iris recognition algorithm. In this study, we present a system that considers both factors and focuses on the latter. The proposed algorithm aims to find out the most efficient wavelet family and its coefficients for encoding the iris template of the experiment samples. The algorithm implemented in software performs segmentation, normalization, feature encoding, data storage, and matching. By using the Haar and Biorthogonal wavelet families at various levels feature encoding is performed by decomposing the normalized iris image. The vertical coefficient is encoded into the iris template and is stored in the database. The performance of the system is evaluated by using the number of degrees of freedom, False Reject Rate (FRR), False Accept Rate (FAR), and Equal Error Rate (EER) and the metrics show that the proposed algorithm can be employed for an iris recognition system.