• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.6
• /
• pp.1081-1090
• /
• 2022

The face recognition model is used for identity recognition of smartphones, providing convenience to many users. As a result, the security review of the DNN model is becoming important, with adversarial attacks present as a well-known vulnerability of the DNN model. Adversarial attacks have evolved to decision-based attack techniques that use only the recognition results of deep learning models to perform attacks. However, existing decision-based attack technique[14] have a problem that requires a large number of queries when generating adversarial examples. In particular, it takes a large number of queries to approximate the gradient. Therefore, in this paper, we propose a method of generating adversarial examples using orthogonal space sampling and dimensionality reduction sampling to avoid wasting queries that are consumed to approximate the gradient of existing decision-based attack technique[14]. Experiments show that our method can reduce the perturbation size of adversarial examples by about 2.4 compared to existing attack technique[14] and increase the attack success rate by 14% compared to existing attack technique[14]. Experimental results demonstrate that the adversarial example generation method proposed in this paper has superior attack performance.

• Journal of the Korea Society for Simulation
• /
• v.18 no.1
• /
• pp.17-25
• /
• 2009

This paper suggests fuzzy-based Segment-Boost method and an effective method for face recognition using the fuzzy-based Segment-Boost. Fuzzy-based Segment-Boost eliminates the limitations of Segment-Boost, and it guarantees improved learning performance and the stability of the performance. By using the fuzzy theory, fuzzy-based Segment-Boost optimizes the selection number of sub-vectors, and leads the optimized learning performance. The fuzzy controller designed in this paper measures learning performance of the fuzzy-based Segment-Boost, and it controls the selection number of sub-vectors by inferring the optimized selection number. The simulation results show that the fuzzy controller inferred the selection number which is very approximate to the true optimized value. As a result, fuzzy-based Segment-Boost showed higher face recognition rate than compared boosting methods and it preserves the velocity of feature selection as fast as that of Segment-Boost. From the experimental results, it was proved that fuzzy-based Segment-Boost has improved and stable performances of learning, feature selection and face recognition.

• Convergence Security Journal
• /
• v.22 no.1
• /
• pp.29-38
• /
• 2022

The importance of prevention of epidemics is increasing due to the serious spread of infectious diseases. For prevention of epidemics, we need to focus on the non-contact industry. Therefore, in this paper, a face recognition door lock that controls access through non-contact is designed and developed. First very simple features are combined to find objects and face recognition is performed using Haar-based cascade algorithm. Then the texture of the image is binarized to find features using LBPH. An non-contact door lock system which composed of Raspberry PI 3B+ board, an ultrasonic sensor, a camera module, a motor, etc. are suggested. To verify actual performance and ascertain the impact of light sources, various experiment were conducted. As experimental results, the maximum value of the recognition rate was about 85.7%.

• Convergence Security Journal
• /
• v.24 no.2
• /
• pp.201-207
• /
• 2024

Voice is a critical element of human communication, and the development of speech recognition models is one of the significant achievements in artificial intelligence, which has recently been applied in various aspects of human life. The application of speech recognition models in the military field is also inevitable. However, before artificial intelligence models can be applied in the military, it is necessary to research their vulnerabilities. In this study, we evaluates the military applicability of the multilingual speech recognition model "Whisper" by examining its vulnerabilities to battlefield noise, white noise, and adversarial attacks. In experiments involving battlefield noise, Whisper showed significant performance degradation with an average Character Error Rate (CER) of 72.4%, indicating difficulties in military applications. In experiments with white noise, Whisper was robust to low-intensity noise but showed performance degradation under high-intensity noise. Adversarial attack experiments revealed vulnerabilities at specific epsilon values. Therefore, the Whisper model requires improvements through fine-tuning, adversarial training, and other methods.

• International Journal of Computer Science & Network Security
• /
• v.24 no.10
• /
• pp.43-52
• /
• 2024

A medium scale Urdu speakers' and English speakers' database with multiple accents and dialects has been developed to use in Urdu Speaker Verification Systems, English Speaker Verification Systems, accents and dialect verification systems. Urdu is the national language of Pakistan and English is the official language. Majority of the people are non-native Urdu speakers and non-native English in all regions of Pakistan in general and Gilgit-Baltistan region in particular. In order to design Urdu and English speaker verification systems for security applications in general and telephone banking in particular, two databases has been designed one for foreign accent of Urdu and another for foreign accent of English language. For the design of databases, voice data is collected from 180 speakers from GB region of Pakistan who could speak Urdu as well as English. The speakers include both genders (males and females) with different age groups ranging from 18 to 69 years. Finally, using a subset of the data, Multilayer Perceptron based speaker verification system has been designed. The designed system achieved overall accuracy rate of 83.4091% for English dataset and 80.0454% for Urdu dataset. It shows slight differences (4.0% with English and 7.4% with Urdu) in recognition accuracy if compared with the recently proposed multilayer perceptron (MLP) based SIS achieved 87.5% recognition accuracy

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.1
• /
• pp.82-89
• /
• 2004

This paper describes how to reduce the effect of an occupation threshold by that the transform of mixture components of HMM parameters is controlled in hierarchical tree structure to prevent from over-adaptation. To reduce correlations between data elements and to remove elements with less variance, we employ PCA (principal component analysis) and ICA (independent component analysis) that would give as good a representation as possible, and decline the effect of over-adaptation. When we set lower occupation threshold and increase the number of transformation function, ordinary WLLR adaptation algorithm represents lower recognition rate than SI models, whereas the proposed MLLR adaptation algorithm represents the improvement of over 2% for the word recognition rate as compared to performance of SI models.

• Journal of the Korean Society for Precision Engineering
• /
• v.15 no.12
• /
• pp.226-235
• /
• 1998

This study proposes the analysis and evaluation of method of time series of ultrasonic signal using the chaotic feature extraction for ultrasonic pattern recognition. The features are extracted from time series data for analysis of weld defects quantitatively. For this purpose, analysis objectives in this study are fractal dimension, Lyapunov exponent, and strange attractor on hyperspace. The Lyapunov exponent is a measure of rate in which phase space diverges nearby trajectories. Chaotic trajectories have at least one positive Lyapunov exponent, and the fractal dimension appears as a metric space such as the phase space trajectory of a dynamical system. In experiment, fractal(correlation) dimensions and Lyapunov exponents show the mean value of 4.663, and 0.093 relatively in case of learning, while the mean value of 4.926, and 0.090 in case of testing in slag inclusion(weld defects) are shown. Therefore, the proposed chaotic feature extraction can be enhancement of precision rate for ultrasonic pattern recognition in defecting signals of weld zone, such as slag inclusion.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.16 no.2
• /
• pp.180-195
• /
• 1991

In this paper, we studied on new method of recognition using structural information to recognize character pattern in orginal shape of Hangeul. First, for the purpose of knowing location of character in input image. it processed Making block. Second, after we investigated. whether vertical vowel exited or not in character image accordingly the center of gravity of Hangeul. each character was classified into Type of Hangeul by searching location and length for horizontal vowel and short pole. Last, we processed it by means of template matching which calculate Uclid's distance on each Jaso in accordance to type classified. This paper made an experiment on 2350 characters and obtained 98.3% classifing rate and 95.2% recognizing rate.

• Journal of Korea Multimedia Society
• /
• v.2 no.4
• /
• pp.490-495
• /
• 1999

In this paper, a method of improving the learning time and convergence rate is proposed to exploit the advantages of artificial neural networks and fuzzy theory to neuron structure. This method is applied to the XOR Problem, n bit parity problem which is used as the benchmark in neural network structure, and recognition of digit image in the vehicle plate image for practical image application. As a result of the experiments, it does not always guarantee the convergence. However, the network showed improved the teaming time and has the high convergence rate. The proposed network can be extended to an arbitrary layer Though a single layer structure Is considered, the proposed method has a capability of high speed 3earning even on large images.

• The Journal of the Acoustical Society of Korea
• /
• v.21 no.2
• /
• pp.190-198
• /
• 2002

In general, we have used the Viterbi algorithm of Speech recognition for decoding. But a decoder in speaker verification has to recognize same word of every speaker differently. In this paper, we propose a noble decoding algorithm that could replace the typical Viterbi algorithm for the speaker verification system. We utilize for the proposed algorithm the speaker adaptation algorithms that transform feature vectors into the region of the client' characteristics in the speech recognition. There are many adaptation algorithms, but we take MLLR (Maximum Likelihood Linear Regression) and MAP (Maximum A-Posterior) adaptation algorithms for proposed algorithm. We could achieve improvement of performance about 30% of EER (Equal Error Rate) using proposed algorithm instead of the typical Viterbi algorithm.