• 제어로봇시스템학회논문지
• /
• 제4권3호
• /
• pp.378-383
• /
• 1998

A FOV, that stands for "Field Of View", refers to the maximum area where a camera could be wholly seen. If a FOV of CCD camera cannot the cover overall inspection area, the overall inspection area should be divided into sub-areas of size FOV. In this paper, we propose a new neural network-based FOV generation method by using a newly modified self-organizing map(SOM) which has multiple structure based on a self-organizing map, and uses new training rule that is composed of the movement, creation and deletion terms. Then, experiment results using real PCB indicate the superiority of the method developed in this study to the existing sequential method.al method.

• 전자공학회논문지B
• /
• 제33B권9호
• /
• pp.88-96
• /
• 1996

Computer vision-based gesture recognition systems consist of image segmentation, object tracking and decision. However, it is difficult to segment an object from image for gesture in computer systems because of vaious illuminations and backgrounds. In this paper, we describe a method to learn features for segmentation, which improves the performance of computer vision-based hand-gesture recognition systems. Systems interact with a user to acquire exact training data and segment information according to a predefined plan. System provides some models to the user, takes pictures of the user's response and then analyzes the pictures with models and a prior knowledge. The system sends messages to the user and operates learning module to extract information with the analyzed result.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2014년도 추계학술대회
• /
• pp.268-270
• /
• 2014

Object extraction problem is one of the most important topics in the research area of computer vision, this type of technique can be widely used in practical, such as image processing, robot vision, automatically traffic guide and so on. In this paper, we propose a different way to estimate the background and foreground without any previous training procedure, this approach can be used for automatic object extraction in the future. A simple experiment result shows that our approach has a good potential for the further more practical application.

• 센서학회지
• /
• 제30권5호
• /
• pp.286-289
• /
• 2021

Image classification is one of the fundamental applications of computer vision. It enables a system to identify an object in an image. Recently, image classification applications have broadened their scope from computer applications to edge devices. The convolutional neural network (CNN) is the main class of deep learning neural networks that are widely used in computer tasks, and it delivers high accuracy. However, CNN algorithms use a large number of parameters and incur high computational costs, which hinder their implementation in edge hardware devices. To address this issue, this paper proposes a lightweight image classifier that provides good accuracy while using fewer parameters. The proposed image classifier diverts the input into three paths and utilizes different scales of receptive fields to extract more feature maps while using fewer parameters at the time of training. This results in the development of a model of small size. This model is tested on the CIFAR-10 dataset and achieves an accuracy of 90% using .26M parameters. This is better than the state-of-the-art models, and it can be implemented on edge devices.

• 도시과학
• /
• 제11권2호
• /
• pp.1-8
• /
• 2022

This study proposed how to identify window defects by using photos uploaded by occupants during the pre-handover inspection of mulch-family housing. A total of 1168 door images were acquired to generate training data and validation data. Subsequently, through the proposed algorithms, every pixel in images labeled a door was binarized using the OTSU threshold, and then dark pixels were identified as defects. Experimental results demonstrated that our computer vision-based defects identification method detects the door with a recall of 57.9%, and door defects with 63.6%. Although it is still a challenge to automatically identify building defects because of the distortion and brightness of photos, this study has the potential to support better defects management. Ultimately, the improved pre-handover inspection may lead to increased customer satisfaction.

• Journal of information and communication convergence engineering
• /
• 제21권2호
• /
• pp.139-144
• /
• 2023

Engineers prefer deep neural networks (DNNs) for solving computer vision problems. However, DNNs pose two major problems. First, neural networks require large amounts of well-labeled data for training. Second, the covariate shift problem is common in computer vision problems. Domain adaptation has been proposed to mitigate this problem. Recent work on adversarial-learning-based unsupervised domain adaptation (UDA) has explained transferability and enabled the model to learn robust features. Despite this advantage, current methods do not guarantee the distinguishability of the latent space unless they consider class-aware information of the target domain. Furthermore, source and target examples alone cannot efficiently extract domain-invariant features from the encoded spaces. To alleviate the problems of existing UDA methods, we propose the mixup regularization in adversarial discriminative domain adaptation (ADDA) method. We validated the effectiveness and generality of the proposed method by performing experiments under three adaptation scenarios: MNIST to USPS, SVHN to MNIST, and MNIST to MNIST-M.

• 반도체디스플레이기술학회지
• /
• 제22권2호
• /
• pp.77-80
• /
• 2023

Since fires in uncontrolled environments pose serious risks to society and individuals, many researchers have been investigating technologies for early detection of fires that occur in everyday life. Recently, with the development of deep learning vision technology, research on fire detection models using neural network backbones such as Transformer and Convolution Natural Network has been actively conducted. Vision-based fire detection systems can solve many problems with physical sensor-based fire detection systems. This paper proposes a fire detection method using the latest YOLOv8, which improves the existing fire detection method. The proposed method develops a system that detects sparks and smoke from input images by training the Yolov8 model using a universal fire detection dataset. We also demonstrate the superiority of the proposed method through experiments by comparing it with existing methods.

• 대한간호학회지
• /
• 제30권5호
• /
• pp.1318-1332
• /
• 2000

The purpose of this study was to determine the effect of walk training on leg strength, flexibility, postural stability, balance and gait in home bound elderly women. Eighteen elderly women of the experimental group aged between 70 and 90 years image who have normal vision, hearing and Romberg test. They participated in the 12 week walk training. The subjects of the experimental group practiced walk training 3 times a week for during 12 weeks. During the 40 minute workout, the subjects practiced 5 minutes of warming-up exercises, 30 minutes of conditioning exercises and 10 minutes of a cool-down exercise. The intensity for the conditioning phase was determined by subject' heart rates, which ranged from 60% to 70% of age-adjusted maximum heart rates. The body composition, leg strength, flexibility, postural stability, balance and gait were measured prior to and after the experimental treatment. The body fat, lean body mass, leg strength (ankle dorsiflexor, plantarflexor, inversor and eversir, knee flexor, extensior), flexibility (range of motion of ankle dorsiflexion, plantarflexion, inversion and eversion), and postural stability of the experimental group were significantly greater than those of the control group. Duration of standing on the right foot and that of standing on the left foot of the experimental group was greater than that of the control group. Total balance scores of the experimental group were significantly higher than those of the control group. Among 13 items for balance, the scores of experimental group in balance with eyes closes, turning balance, sternal nudge, neck turning, one leg standing balance and back extension were higher than those of the control group. Total scores of gait of the experimental group were significantly higher than those of the control group following the walking training. Scores of experimental group in step height, step length and walk stance while walking among 9 items for gait were significantly higher than those of the control group. The results suggest that walk training can improve physical fitness for prevention in home bound elderly women.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.1413-1415
• /
• 2004

This paper proposes a SVM(Support Vector Machine) training algorithm to control a service robot with voice command. The service robot with a stereo vision system and dual manipulators of four degrees of freedom implements a User-Dependent Voice Control System. The training of SVM algorithm that is one of the statistical learning theories leads to a QP(quadratic programming) problem. In this paper, we present an efficient SVM speech recognition scheme especially based on less learning data comparing with conventional approaches. SVM discriminator decides rejection or acceptance of user's extracted voice features by the MFCC(Mel Frequency Cepstrum Coefficient). Among several SVM kernels, the exponential RBF function gives the best classification and the accurate user recognition. The numerical simulation and the experiment verified the usefulness of the proposed algorithm.

• 한국산학기술학회:학술대회논문집
• /
• 한국산학기술학회 2009년도 추계학술발표논문집
• /
• pp.930-933
• /
• 2009

가입도 1.00D 미만의 초기 노안을 대상으로 푸쉬업과 플리퍼 방법에 의한 조절훈련이 가입도 측정값을 감소시킬 수 있는지 알아보고자 하였다. 조절훈련은 12주간 매일 home vision training을 실시하였고, 일주일 간격으로 조절력, 조절지체, 조절용이성 및 가입도 검사를 실시하였다. 푸쉬업 훈련과 플리퍼 훈련 결과, 가입도는 0.125D~0.375D 유의하게 감소되었고, 푸쉬업 훈련이 플리퍼 훈련보다 가입도 감소효과가 큰 것으로 나타났다. 푸쉬업 훈련과 플리퍼 훈련은 조절력을 향상시켜 초기 노안이 필요로 하는 가입도를 감소시킬 수 있으며, 이와 같은 상태를 유지하기 위해서는 지속적인 훈련이 필요할 것으로 나타났다.