• International Journal of Computer Science & Network Security
• /
• v.23 no.4
• /
• pp.48-54
• /
• 2023

Nowadays, a blind man finds it very difficult to cross the roads. They should be very vigilant with every step they take. To resolve this problem, Convolutional Neural Networks(CNN) is a best method to analyse the data and automate the model without intervention of human being. In this work, a traffic signal recognition system is designed using CNN for the visually impaired. To provide a safe walking environment, a voice message is given according to light state and timer state at that instance. The developed model consists of two phases, in the first phase the CNN model is trained to classify different images captured from traffic signals. Common Objects in Context (COCO) labelled dataset is used, which includes images of different classes like traffic lights, bicycles, cars etc. The traffic light object will be detected using this labelled dataset with help of object detection model. The CNN model detects the color of the traffic light and timer displayed on the traffic image. In the second phase, from the detected color of the light and timer value a text message is generated and sent to the text-to-speech conversion model to make voice guidance for the blind person. The developed traffic light recognition model recognizes traffic light color and countdown timer displayed on the signal for safe signal crossing. The countdown timer displayed on the signal was not considered in existing models which is very useful. The proposed model has given accurate results in different scenarios when compared to other models.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.35-40
• /
• 2021

People counting data is crucial for most business owners, since they can derive meaningful information about customers movement within their businesses. For example, owners of the supermarkets can increase or decrease the number of checkouts counters depending on number of occupants. Also, it has many applications in smart buildings, too. Where it can be used as a smart controller to control heating and cooling systems depending on a number of occupants in each room. There are advanced technologies like camera-based people counting system, which can give more accurate counting result. But they are expensive, hard to deploy and privacy invasive. In this paper, we propose a method and a hardware sensor for counting people passing through a passage or an entrance using IR Infrared sensors. Proposed sensor operates at low voltage, so low power consumption ensure long duration on batteries. Moreover, we propose a new method that distinguishes human body and other objects. Proposed method is inexpensive, easy to install and most importantly, it is real-time. The evaluation of our proposed method showed that when counting people passing one by one without overlapping, recall was 96% and when people carrying handbag like objects, the precision was 88%. Our proposed method outperforms IR Infrared based people counting systems in term of counting accuracy.

• Journal of the Korean Society of Industry Convergence
• /
• v.19 no.1
• /
• pp.1-9
• /
• 2016

In this paper, we propose a new learning control scheme for various walk motion control of biped robot with same learning-base by neural network. We show that learning control algorithm based on the neural network is significantly more attractive intelligent controller design than previous traditional forms of control systems. A multi layer back propagation neural network identification is simulated to obtain a dynamic model of biped robot. Once the neural network has learned, the other neural network control is designed for various trajectory tracking control with same learning-base. The biped robots have been received increased attention due to several properties such as its human like mobility and the high-order dynamic equation. These properties enable the biped robots to perform the dangerous works instead of human beings. Thus, the stable walking control of the biped robots is a fundamentally hot issue and has been studied by many researchers. However, legged locomotion, it is difficult to control the biped robots. Besides, unlike the robot manipulator, the biped robot has an uncontrollable degree of freedom playing a dominant role for the stability of their locomotion in the biped robot dynamics. From the simulation and experiments the reliability of iterative learning control was illustrated.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.12 no.6
• /
• pp.437-444
• /
• 2002

This paper is to provide the basic way of a acoustical evaluation and efficient control noise by investigating the limits of perceptual loudness of living environment and by finding out any correlation between Physical characteristics of noise and psychoacoustic parameters. The limits of perceptual loudness were selected by the subjects in a chamber for residential and working environment. And the noise sources were analyzed to find out whether there is any correlation with Zwicker parameters and ACF factors. In this study especially, to set up the domestic evaluation grade about floor impact noise. we'd like to suggest the loudness Perception research result as fundamental resource for setting up the evaluation grade through the result that is based on annoyance. In the result of this research, upper limit of heavy-weight impact noise was L-60, and lower limit of it was L-50. On the other hand, upper limit of light-weight impact noise was L-70, and lower limit of it was L-55. It seemed that the loudness of noise from vacuum cleaner noise does not affect its perceived noisiness. Noises implicated In human such as floor walking noise and talking sound, are the most irritating noise in office environment.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.32 no.2
• /
• pp.11-24
• /
• 2004

The purpose of this study is to present data on the human activities responding to the physical environment of ‘Ulsan’s Grand Parks by evaluating user behavior and activity, visiting motivations, and user satisfaction. This study was conducted using multi-methods such as interviews and questionnaires surveys. The main findings of this study can be summarized in six parts as follows: \circled1 Behavior patterns showed that the users visited the park left within 30 minutes and the frequency of visits was 1 or 2 times per week. They spent their time mostly on ‘walking’ and ‘picnicking’. The users visited regardless of the seasons or the day of the week. \circled2 The priorities for improvements were analyzed as follows: the users expressed their demands for ‘shadowing facilities’ like shelters and pergolas because they used these facilities frequently. Also, the users would like an ‘event program’, ‘sign system’ and ‘guide program’. \circled3 Analysis of the the user’s perception of the park showed that they perceive the park as ‘representative source of the landscape and open space in Ulsan and place for making contact with nature. \circled4 In examining the visiting motivations of the users of Ulsan Grand Park, it was revealed that people use the park for ‘time with family and friends’, ‘to escape from city life’, ‘to relieve fatigue. As a result of factor analysis, 4 factors were identified such as ‘physical motive(MF1)’, ‘exploratory motive(MF2)’, ‘social motive(MF3)’ and ‘emotional motive(MF4). \circled5 Park users’ evaluation for park facilities showed that people are satisfied with most of the facilities and especially, they have high level of satisfaction for ‘footpaths’, ‘squares’ and ‘picnicspace’. The evaluation of the park user’s of activity reveals that they are content with nearly all the variables. Especially, they have high level of satisfaction for the variables of ‘convenience for dynamic activities’, ‘making of a beautiful atmosphere, ‘accessibility from the outside’ and, ‘convenience in group activitie. Factor analysis of the park user’s of activity revealed 5 factors such as ‘convenience and interest factor (AF1)’, ‘park maintenance, management and use program(AF2)’, ‘visual beauty(AF3)’, ‘safety and accessibility(AF4)’ and ‘crowding(AF5)’. \circled6 Regression analysis was employed to get the predictor factors of overall satisfaction with a result of 60.0%($R^2$). The variance was explained as ‘quality of the picnic space’, ‘convenience and interest factor while using the park’, ‘park program for maintain and management in the park’, ‘visually beauty while using the park’, ‘safety and accessibility of the parks’, ‘quality of the pond’, ‘crowding’, ‘quality of the square’.