• 한국산학기술학회:학술대회논문집
• /
• 한국산학기술학회 2006년도 추계학술발표논문집
• /
• pp.122-128
• /
• 2006

The intent of this paper is to describe a neural network structure called multi dynamic neural network(MDNN), and examine how it can be used in developing a learning scheme for computing robot inverse kinematic transformations. The architecture and learning algorithm of the proposed dynamic neural network structure, the MDNN, are described. Computer simulations are demonstrate the effectiveness of the proposed learning using the MDNN.

• International Journal of Internet, Broadcasting and Communication
• /
• 제11권4호
• /
• pp.76-85
• /
• 2019

We introduce a vision-based object detection method for real-time video surveillance system in low-end edge computing environments. Recently, the accuracy of object detection has been improved due to the performance of approaches based on deep learning algorithm such as Region Convolutional Neural Network(R-CNN) which has two stage for inferencing. On the other hand, one stage detection algorithms such as single-shot detection (SSD) and you only look once (YOLO) have been developed at the expense of some accuracy and can be used for real-time systems. However, high-performance hardware such as General-Purpose computing on Graphics Processing Unit(GPGPU) is required to still achieve excellent object detection performance and speed. To address hardware requirement that is burdensome to low-end edge computing environments, We propose sub-frame analysis method for the object detection. In specific, We divide a whole image frame into smaller ones then inference them on Convolutional Neural Network (CNN) based image detection network, which is much faster than conventional network designed forfull frame image. We reduced its computationalrequirementsignificantly without losing throughput and object detection accuracy with the proposed method.

• 산업공학
• /
• 제21권1호
• /
• pp.65-74
• /
• 2008

In the future, the ubiquitous computing environment will provide users with context-aware services, intelligently interacting with omnipresent resources and surrounding environments at any location and time. Therefore, the ubiquitous computing environment requires context-aware applications in order to gather and analyze context information in various situations. However, existing context-aware applications have mainly focused on providing services and adapting themselves to users based on user related contexts. But this environment requires not only user related contexts but also Quality-of-Service (QoS) related contexts. In this paper we propose a frame-work for supporting user-oriented QoS in ubiquitous environments. The proposed approach adopts a semantic ontology and a neural network algorithm in order to reason about explicit and uncertain contexts. We also show the possibility of applying the framework to real environments such as collaborative engineering service, automobile maintenance service and ubiquitous home service.

• International Journal of Computer Science & Network Security
• /
• 제23권5호
• /
• pp.33-40
• /
• 2023

The mainstay of current image recovery frameworks is Content-Based Image Retrieval (CBIR). The most distinctive retrieval method involves the submission of an image query, after which the system extracts visual characteristics such as shape, color, and texture from the images. Most of the techniques use RGB color space to extract and classify images as it is the default color space of the images when those techniques fail to change the color space of the images. To determine the most effective color space for retrieving images, this research discusses the transformation of RGB to different color spaces, feature extraction, and usage of Convolutional Neural Networks for retrieval.

• Journal of Information Processing Systems
• /
• 제14권5호
• /
• pp.1075-1086
• /
• 2018

In this paper a cross-validation algorithm for training probabilistic neural networks (PNNs) is presented in order to be applied to automatic face identification. Actually, standard PNNs perform pretty well for small and medium sized databases but they suffer from serious problems when it comes to using them with large databases like those encountered in biometrics applications. To address this issue, we proposed in this work a new training algorithm for PNNs to reduce the hidden layer's size and avoid over-fitting at the same time. The proposed training algorithm generates networks with a smaller hidden layer which contains only representative examples in the training data set. Moreover, adding new classes or samples after training does not require retraining, which is one of the main characteristics of this solution. Results presented in this work show a great improvement both in the processing speed and generalization of the proposed classifier. This improvement is mainly caused by reducing significantly the size of the hidden layer.

• 한국통계학회:학술대회논문집
• /
• 한국통계학회 2004년도 학술발표논문집
• /
• pp.65-66
• /
• 2004

The aim of this article is to illustrate what soft computing is and how important it is.

• International Journal of Internet, Broadcasting and Communication
• /
• 제10권2호
• /
• pp.25-30
• /
• 2018

This paper is about wine quality classification with multilayer perceptron using the deep neural network. Wine complexity is an issue when predicting the quality. And the deep neural network is considered when using complex dataset. Wine Producers always aim high to get the highest possible quality. They are working on how to achieve the best results with minimum cost and efforts. Deep learning is the possible solution for them. It can help them to understand the pattern and predictions. Although there have been past researchers, which shows how artificial neural network or data mining can be used with different techniques, in this paper, rather not focusing on various techniques, we evaluate how a deep learning model predicts for the quality using two different activation functions. It will help wine producers to decide, how to lead their business with deep learning. Prediction performance could change tremendously with different models and techniques used. There are many factors, which, impact the quality of the wine. Therefore, it is a good idea to use best features for prediction. However, it could also be a good idea to test this dataset without separating these features. It means we use all features so that the system can consider all the feature. In the experiment, due to the limited data set and limited features provided, it was not possible for a system to choose the effective features.

• 제어로봇시스템학회논문지
• /
• 제7권11호
• /
• pp.918-926
• /
• 2001

In this paper, intelligent control method using neural network as a nonlinear controller is presented. Simulation studies for three link rotary robot are performed. Neural network controller is implemented on DSP board in PC to make real time computing possible. On-line training algorithms for neural network control are proposed. As a test-bed, a large x-y table was build and interface with PC has been implemented. Experiments such as inverted pendulum control and large x-y table position control are performed. The results for different PD controller gains with neural network show excellent position tracking for circular trajectory compared with those for PD controller only. Neural control scheme also works better for controlling inverted pendulum on x-y table.

• 로봇학회논문지
• /
• 제17권2호
• /
• pp.133-141
• /
• 2022

This paper introduces model compression algorithms which make a deep neural network smaller and faster for embedded systems. The model compression algorithms can be largely categorized into pruning, quantization and knowledge distillation. In this study, gradual pruning, quantization aware training, and knowledge distillation which learns the activation boundary in the hidden layer of the teacher neural network are integrated. As a large deep neural network is compressed and accelerated by these algorithms, embedded computing boards can run the deep neural network much faster with less memory usage while preserving the reasonable accuracy. To evaluate the performance of the compressed neural networks, we evaluate the size, latency and accuracy of the deep neural network, DenseNet201, for image classification with CIFAR-10 dataset on the NVIDIA Jetson Xavier.

• 문화기술의 융합
• /
• 제8권5호
• /
• pp.613-618
• /
• 2022

SNN(Spiking Neural Networks) 기반의 인공지능 연구는 현재 유행하는 DNN(Deep Neural Networks) 기반의 인공지능의 한계를 극복할 수 있는 차세대 인공지능으로서 주목받고 있다. 본 논문에서는 SNN 형태의 입력을 뉴로모픽 컴퓨팅 시스템에서 구동시킬 수 있는 시스템 SW인 SNN 컴파일러의 구조에 대하여 설명한다. 또한 컴파일러 구현을 위하여 사용된 알고리즘을 소개하고 매핑 알고리즘의 동작 형태에 따라 뉴로모픽 컴퓨팅 시스템에서 수행시간이 어떻게 달라지는지에 대한 실험결과를 제시한다. 본문에서 제안한 매핑 알고리즘은 랜덤 매핑에 비해 최대 3.96배의 수행속도 향상이 있었다. 해당 연구 결과를 통해 SNN들을 다양한 뉴로모픽 하드웨어에서 적용할 수 있을 것이다.