• The Journal of Korean Institute of Information Technology
• /
• v.16 no.11
• /
• pp.105-113
• /
• 2018

In this paper, we proposed a method for effective classification of eye, nose, and mouth of human face. Most recent image classification uses Convolutional Neural Network(CNN). However, the features extracted by CNN are not sufficient and the classification effect is not too high. We proposed a new algorithm to improve the classification effect. The proposed method can be roughly divided into three parts. First, the Haar feature extraction algorithm is used to construct the eye, nose, and mouth dataset of face. The second, the model extracts CNN features of image using AlexNet. Finally, Haar-CNN features are extracted by performing convolution after Haar feature extraction. After that, CNN features and Haar-CNN features are fused and classify images using softmax. Recognition rate using mixed features could be increased about 4% than CNN feature. Experiments have demonstrated the performance of the proposed algorithm.

• International Journal of Computer Science & Network Security
• /
• v.22 no.9
• /
• pp.195-207
• /
• 2022

COVID-19 has remained one of the most serious health crises in recent history, resulting in the tragic loss of lives and significant economic impacts on the entire world. The difficulty of controlling COVID-19 poses a threat to the global health sector. Considering that Artificial Intelligence (AI) has contributed to improving research methods and solving problems facing diverse fields of study, AI algorithms have also proven effective in disease detection and early diagnosis. Specifically, acoustic features offer a promising prospect for the early detection of respiratory diseases. Motivated by these observations, this study conceptualized a speech-based diagnostic model to aid in COVID-19 diagnosis. The proposed methodology uses speech signals from confirmed positive and negative cases of COVID-19 to extract features through the pre-trained Visual Geometry Group (VGG-16) model based on Mel spectrogram images. This is used in addition to the K-means algorithm that determines effective features, followed by a Genetic Algorithm-Support Vector Machine (GA-SVM) classifier to classify cases. The experimental findings indicate the proposed methodology's capability to classify COVID-19 and NOT COVID-19 of varying ages and speaking different languages, as demonstrated in the simulations. The proposed methodology depends on deep features, followed by the dimension reduction technique for features to detect COVID-19. As a result, it produces better and more consistent performance than handcrafted features used in previous studies.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.8
• /
• pp.365-370
• /
• 2023

This paper aims to compare the performance of speech data classification into two groups, adult and elderly, based on the acoustic and linguistic characteristics that change due to aging, such as changes in respiratory patterns, phonation, pitch, frequency, and language expression ability. For acoustic features we used attributes related to the frequency, amplitude, and spectrum of speech voices. As for linguistic features, we extracted hidden state vector representations containing contextual information from the transcription of speech utterances using KoBERT, a Korean pre-trained language model that has shown excellent performance in natural language processing tasks. The classification performance of each model trained based on acoustic and linguistic features was evaluated, and the F1 scores of each model for the two classes, adult and elderly, were examined after address the class imbalance problem by down-sampling. The experimental results showed that using linguistic features provided better performance for classifying adult and elderly than using acoustic features, and even when the class proportions were equal, the classification performance for adult was higher than that for elderly.

• The Bulletin of The Korean Astronomical Society
• /
• v.46 no.2
• /
• pp.40.3-41
• /
• 2021

Minor mergers leave behind long lived, but extremely faint and extended tidal features including tails, streams, loops and plumes. These act as a fossil record for the host galaxy's past interactions, allowing us to infer recent accretion histories and place constraints on the properties and nature of a galaxy's dark matter halo. However, shallow imaging or small homogeneous samples of past surveys have resulted in weak observational constraints on the role of galaxy mergers and interactions in galaxy assembly. The Rubin Observatory, which is optimised to deliver fast, wide field-of-view imaging, will enable deep and unbiased observations over the 18,000 square degrees of the Legacy Survey of Space and Time (LSST), resulting in samples of potentially of millions of objects undergoing tidal interactions. Using realistic mock images produced with state-of-the-art cosmological simulations we perform a comprehensive theoretical investigation of the extended diffuse light around galaxies and galaxy groups down to low stellar mass densities. We consider the nature, frequency and visibility of tidal features and debris across a range of environments and stellar masses as well as their reliability as an indicator of galaxy accretion histories. We consider how observational biases such as projection effects, the point-spread-function and survey depth may effect the proper characterisation and measurement of tidal features, finding that LSST will be capable of recovering much of the flux found in the outskirts of L* galaxies at redshifts beyond local volume. In our simulated sample, tidal features are ubiquitous In L* galaxies and remain common even at significantly lower masses (M*>10^10 Msun). The fraction of stellar mass found in tidal features increases towards higher masses, rising to 5-10% for the most massive objects in our sample (M*~10^11.5 Msun). Such objects frequently exhibit many distinct tidal features often with complex morphologies, becoming increasingly numerous with increased depth. The interpretation and characterisation of such features can vary significantly with orientation and imaging depth. Our findings demonstrate the importance of accounting for the biases that arise from projection effects and surface-brightness limits and suggest that, even after the LSST is complete, much of the discovery space in low surface-brightness Universe will remain to be explored.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.3D
• /
• pp.437-444
• /
• 2008

In the paper, a methodology is verified to integrate IKONOS-2 satellite imagery and ALS dataset by compensating biases of RPC models. To achieve this, conjugate features from both data should be extracted in advance. For this purpose, linear features are chosen as conjugate features because they can be accurately extracted from man-made structures in urban area and more easily extracted than point features from ALS data. Then, observation equations are established from similarity measurements of the extracted features. During the process, several kinds of transformation functions were selected and used to register them. In addition, it was also analyzed how the number of linear features used as control features affects the accuracy of registration results. Finally, the results were evaluated by using check-points obtained from DGPS surveying techniques and it was clearly demonstrated that the proposed algorithms are appropriate to integrate these data.

• International Journal of Computer Science & Network Security
• /
• v.24 no.5
• /
• pp.11-20
• /
• 2024

While artificial neural networks are adept at identifying patterns, they can struggle to distinguish between actual correlations and false associations between extracted facial features and criminal behavior within the training data. These associations may not indicate causal connections. Socioeconomic factors, ethnicity, or even chance occurrences in the data can influence both facial features and criminal activity. Consequently, the artificial neural network might identify linked features without understanding the underlying cause. This raises concerns about incorrect linkages and potential misclassification of individuals based on features unrelated to criminal tendencies. To address this challenge, we propose a novel region-based training approach for artificial neural networks focused on criminal propensity detection. Instead of solely relying on overall facial recognition, the network would systematically analyze each facial feature in isolation. This fine-grained approach would enable the network to identify which specific features hold the strongest correlations with criminal activity within the training data. By focusing on these key features, the network can be optimized for more accurate and reliable criminal propensity prediction. This study examines the effectiveness of various algorithms for criminal propensity classification. We evaluate YOLO versions YOLOv5 and YOLOv8 alongside VGG-16. Our findings indicate that YOLO achieved the highest accuracy 0.93 in classifying criminal and non-criminal facial features. While these results are promising, we acknowledge the need for further research on bias and misclassification in criminal justice applications

• Proceedings of the Safety Management and Science Conference
• /
• 2010.04a
• /
• pp.519-524
• /
• 2010

The research presents implementation strategy and features for experimental design of quality. The MINITAB models of experimental design classfies by the kinds of factors, orthogonality, blocking, confounding and rotationability. The results discussed in this paper can be easily used for practictioners.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2008.05a
• /
• pp.151-154
• /
• 2008

In this paper, this program which is Retrieval System using Image Gray-scale histogram and Edge features is used to reduce the errors incurred by inputting methods which are used in a current Retrieval System. The Retrieval Algorithm is proceeding with several steps which are extracting features of images quality, extracting edge features and refining images, analysing extracted features, retaining important information from analyzed features, retrieving retained information from database, extracting and comparing among images from retrieved database. The proposed Retrieval System is used for a fast retrieval with accuracy and it is confirmed through simulations.

• Journal of Electrical Engineering and Technology
• /
• v.7 no.4
• /
• pp.570-575
• /
• 2012

In a wireless communication RF transmitter, the output of a quadrature modulator (QM) is distorted by not only the linear imperfection features such as in/quadrature-phase (I/Q) input gain imbalance, local phase imbalance, and local gain imbalance but also the nonlinear imperfection features such as direct current (DC) offset and mixer nonlinearity related to in-band spurious signal. In this paper, we propose the unified QM model to analyze the combined effects of the linear and nonlinear imperfection features on the performance of the QM. The unified QM model consists of two identical nonlinear systems and modified I/Q inputs based on the two-port nonlinear mixer model. The unified QM model shows that the output signals can be expressed by mixer circuit parameters such as intercept point and gain as well as the imperfection features. The proposed approach is validated by not only simulation but also measurement.

• ETRI Journal
• /
• v.34 no.6
• /
• pp.847-857
• /
• 2012

Intrusion detection systems (IDSs) have an important effect on system defense and security. Recently, most IDS methods have used transformed features, selected features, or original features. Both feature transformation and feature selection have their advantages. Neighborhood component analysis feature transformation and genetic feature selection (NCAGAFS) is proposed in this research. NCAGAFS is based on soft computing and data mining and uses the advantages of both transformation and selection. This method transforms features via neighborhood component analysis and chooses the best features with a classifier based on a genetic feature selection method. This novel approach is verified using the KDD Cup99 dataset, demonstrating higher performances than other well-known methods under various classifiers have demonstrated.