• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.5
• /
• pp.1295-1302
• /
• 2010

This paper proposed indoor location recognition method based on RSSI(received signal strength indication) using the LVQ network. In order to verify the effectiveness of the proposed method, we performed experiments, and then compared to the conventional triangularity measurement method. In the experiments, we set up the system to the laboratory, divided the 40 section, and installed 6 nodes as a reference node. We obtained the log-normal path loss model of wireless channels, RSSI converted into the distance. The distance values used as the input of LVQ. To learn the LVQ network, we set the target values as section indices. In the experiments, we determined the optimal number of subclass, and confirmed that the success rate of training phase was 96%, test phase was 91%.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.3
• /
• pp.89-100
• /
• 2023

Recently, on-device artificial intelligence (AI) solutions using mobile devices and embedded edge devices have emerged in various fields, such as computer vision, to address network traffic burdens, low-energy operations, and security problems. Although vision transformer deep learning models have outperformed conventional convolutional neural network (CNN) models in computer vision, they require more computations and parameters than CNN models. Thus, they are not directly applicable to embedded edge devices with limited hardware resources. Many researchers have proposed various model compression methods or lightweight architectures for vision transformers; however, there are only a few studies evaluating the effects of model compression techniques of vision transformers on performance. Regarding this problem, this paper presents a performance evaluation of vision transformers on embedded platforms. We investigated the behaviors of three vision transformers: DeiT, LeViT, and MobileViT. Each model performance was evaluated by accuracy and inference time on edge devices using the ImageNet dataset. We assessed the effects of the quantization method applied to the models on latency enhancement and accuracy degradation by profiling the proportion of response time occupied by major operations. In addition, we evaluated the performance of each model on GPU and EdgeTPU-based edge devices. In our experimental results, LeViT showed the best performance in CPU-based edge devices, and DeiT-small showed the highest performance improvement in GPU-based edge devices. In addition, only MobileViT models showed performance improvement on EdgeTPU. Summarizing the analysis results through profiling, the degree of performance improvement of each vision transformer model was highly dependent on the proportion of parts that could be optimized in the target edge device. In summary, to apply vision transformers to on-device AI solutions, either proper operation composition and optimizations specific to target edge devices must be considered.

• Korean Management Science Review
• /
• v.16 no.1
• /
• pp.157-171
• /
• 1999

In a increasingly competitive marketplace, the manufacturing companies have no choice but looking for ways to improve productivity to sustain their competitiveness and survive in the industry. Recently cellular manufacturing has been under discussion as an option to be easily implemented without burdensome capital investment. The objective of cellular manufacturing is to realize many aspects of efficiencies associated with mass production in the less repetitive job-shop production systems. The very first step for cellular manufacturing is to group the sets of parts having similar processing requirements into part families, and the equipment needed to process a particular part family into machine cells. The underlying problem to determine the part and machine assignments to each manufacturing cell is called the cell formation. The purpose of this study is to develop a clustering algorithm based on the neural network approach which overcomes the drawbacks of ART1 algorithm for cell formation problems. In this paper, a generalized learning vector quantization(GLVQ) algorithm was devised in order to transform a 0/1 part-machine assignment matrix into the matrix with diagonal blocks in such a way to increase clustering performance. Furthermore, an assignment problem model and a rearrangement procedure has been embedded to increase efficiency. The performance of the proposed algorithm has been evaluated using data sets adopted by prior studies on cell formation. The proposed algorithm dominates almost all the cell formation reported so far, based on the grouping index($\alpha$ = 0.2). Among 27 cell formation problems investigated, the result by the proposed algorithm was superior in 11, equal 15, and inferior only in 1.

• The Journal of the Acoustical Society of Korea
• /
• v.17 no.3E
• /
• pp.12-21
• /
• 1998

In this paper, we describe the one channel five-way, V/U/M/N/S (Voice/Unvoice/Nasal/Silent), classification algorithm for automatically classifying speech. The decision making process is viewed as a pattern viewed as a pattern recognition problem. Two aspects of the algorithm are developed: feature selection and classifier type. The feature selection procedure is studied for identifying a set of features to make V/U/M/N/S classification. The classifiers used are a vector quantization (VQ), a neural network(NN), and a decision tree method. Actual five sentences spoken by six speakers, three male and three female, are tested with proposed classifiers. From a set of measurement tests, the proposed classifiers show fairly good accuracy for V/U/M/N/S decision.

• Journal of the Korea Society of Computer and Information
• /
• v.6 no.2
• /
• pp.41-47
• /
• 2001

In this paper, I propose a detection algorithm that can reliably separate moving objects from noisy background in the image sequence received from a camera at the fixed position. The proposed algorithm consists of four processes: generation of the difference image between the input image and the reference image. multilevel quantization of the difference image, and multistage merging in the quantized image, detection of the moving object using a back propagation in a neural network. The test results show that the proposed algorithm can detect moving objects very effectively in noisy environment.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2005.11a
• /
• pp.101-105
• /
• 2005

본 논문에서는 LVQ(Learning Vector Quantization)을 퍼지화한 새로운 퍼지 학습 법칙을 제안하였다. 퍼지 LVQ 학습 법칙 3은 기존의 학습률 대신에 퍼지 학습률을 사용하였는데, 기존의 LVQ와는 달리 비대칭인 학습률을 사용하였다. 기본의 LVQ에서는 분류가 맞거나 틀렸을 때 같은 학습률을 사용하고 부호만 달랐으나, 새로운 퍼지 학습 법칙에서는 분류가 맞거나 틀렸을 때 부호가 다를 뿐만 아니라 학습률도 다르다. 이 새로운 퍼지 학습 법칙을 무감독 신경회로망인 improved IAFC(Integrated Adaptive Fuzzy Clustering) 신경회로망에 적용하여 감독 신경회로망으로 변형하였다. Improved IAFC 신경회로망은 유연성이 있으면서도 안정성이 있다. 제안한 supervised IAFC 신경회로망 3의 성능과 오류 역전파 신경회로망의 성능을 비교하기 위하여 iris 데이터를 사용하였는데 Supervised IAFC 신경회로망 3가 오류 역전파 신경회로망보다 성능이 우수하였다.

• Proceedings of the KIEE Conference
• /
• 2000.11d
• /
• pp.848-850
• /
• 2000

본 논문에서는 신경회로망을 이용하여 On-Line상에서 EEG(Electroencephalogram) 신호를 분류하는 방법을 제안한다. EEG 신호란 인간의 두뇌활동에서 발생하는 전기적 신호로서 고도의 비선형과 시변 특성을 지니고 있어 정량적인 분석이 어려운 신호로 여겨진다. 이를 분석하기 위해 본 논문에서는 입력 벡터들을 서브클래스로 분류하는 경쟁 레이어와 서브클래스를 모아 정해진 클래스를 선택하는 선형 레이어로 이루어진 LVQ (Learning Vector Quantization) 신경망을 구성하고 On-Line 분석결과를 제시한다. 이러한 On-line 분석방법은 EEG 신호를 실시간으로 분석하여 컴퓨터를 인간의 생각만으로 제어될 수 있는 BCI(Brain Computer Interface)의 구현에 사용될 것이다.

• Proceedings of the KIEE Conference
• /
• 1989.07a
• /
• pp.481-485
• /
• 1989

The object of this parer is on the design of an broad board classifier for connected. Korean digit. Many approaches have been applied in speech recognition systems: parametric vector quantization, dynamic programming and hiden Markov model. In the 80's the neural network method, which is expected to solve complex speech recognition problems, came bach. We have chosen the rule based system for our model. The phoneme-groups that we wish to classify are vowel_like, plosive_like fricative_like, and stop_like.The data used are 1380 connected digits spoken by three untrained male speakers. We have seen 91.5% classification rate.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2000.06a
• /
• pp.1133-1138
• /
• 2000

This paper describes a signal recognition method for diagnosing the rotating machinery using wavelet-aided Self-Organizing Feature Map(SOFM). The SOFM specialized from neural network is a new and effective algorithm for interpreting large and complex data sets. It converts high-dimensional data items into simple order relationships with low dimension. Additionally the Learning Vector Quantization(LVQ) is used for reducing the error from SOFM. Multi-resolution and wavelet transform are used to extract salient features from the primary vibration signals. Since it decomposes the raw timebase signal into two respective parts in the time space and frequency domain, it does not lose either information unlike Fourier transform. This paper is focused on the development of advanced signal classifier in order to automatize vibration signal pattern recognition. This method is verified by the experiment and several abnormal vibrations such as unbalance and rubbing are classified with high flexibility and reliability by the proposed methods.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.11a
• /
• pp.57-59
• /
• 2020

본 논문에서는 컨볼루션 신경망을 이용하여 다시점 비디오의 중간 시점 양자화 노이즈를 제거하는 방안을 제안한다. 다시점 비디오에서 중간 시점의 화질을 개선하기 위한 방안으로 인접 시점의 정보를 활용하였다. 제안하는 알고리즘을 적용하여 중간 시정에서의 양자화 노이즈를 제거할 수 있으며, 화질 (PSNR, peak-to-noise ratio)를 개선할 수 있다. 인접 시접의 정보를 활용할 경우, 일반적인 양자화 노이즈에 대해서 학습한 결과 대비 성능 향상을 제공한다.