• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.877-885
• /
• 2022

Vision-based activity recognition has been widely attempted at construction sites to estimate productivity and enhance workers' health and safety. Previous studies have focused on extracting an individual worker's postural information from sequential image frames for activity recognition. However, various trades of workers perform different tasks with similar postural patterns, which degrades the performance of activity recognition based on postural information. To this end, this research exploited a concept of human-object interaction, the interaction between a worker and their surrounding objects, considering the fact that trade workers interact with a specific object (e.g., working tools or construction materials) relevant to their trades. This research developed an approach to understand the context from sequential image frames based on four features: posture, object, spatial features, and temporal feature. Both posture and object features were used to analyze the interaction between the worker and the target object, and the other two features were used to detect movements from the entire region of image frames in both temporal and spatial domains. The developed approach used convolutional neural networks (CNN) for feature extractors and activity classifiers and long short-term memory (LSTM) was also used as an activity classifier. The developed approach provided an average accuracy of 85.96% for classifying 12 target construction tasks performed by two trades of workers, which was higher than two benchmark models. This experimental result indicated that integrating a concept of the human-object interaction offers great benefits in activity recognition when various trade workers coexist in a scene.

• Korean Journal of Child Studies
• /
• v.29 no.5
• /
• pp.301-319
• /
• 2008

This study investigated the effectiveness of reading intervention on at-risk readers from first through third grade. The 34 children below the 20th percentile on an oral reading fluency test participated in the reading intervention program for 15 weeks. They received small group instruction in one 40-minute session per week. Data were analyzed with one-way ANOVA, paired t-test and effect size for individual cases. Upon completion of the intervention, at-risk first graders showed significantly higher performance in print concept, word recognition, oral reading fluency and reading comprehension. The at-risk second and third graders improved only in oral reading fluency. Most of children who received the intervention demonstrated a large effect in oral reading fluency.

• Korean Institute of Interior Design Journal
• /
• v.20 no.6
• /
• pp.71-78
• /
• 2011

Although the recognition and emotion about space is subjective and individual, if standard is proposed through common factor, objective, quantified space image evaluation will be available. In addition, space image evaluation standard caused by recognitive-emotional factor can meet requests of space users and increase psychological satisfactions. The purpose of this study is to grasp the space image caused by recognitive-emotional factor in space with PAD model and analyze the evaluation of space image giving visual, recognitive and emotional effects. The analysis result revealed that 'joyfulness' and access-avoidance had a very similar distribution. The result means that space is evaluated with the degree of 'joyfulness' for space and it is led by approach-avoidance behavior. The recognition factor that forms and evaluates space image and decides approach-avoidance is expressed as adjective images such as 'fresh, joyful, light and static and its emotional factors are adjective images such as 'calm, allowable, joyful and quiet'.

• Journal of Institute of Control, Robotics and Systems
• /
• v.16 no.6
• /
• pp.517-523
• /
• 2010

In this paper, we proposed an algorithm to detect aggressive driving status by analysing six kinds of driving patterns, which was achieved by comparing for the feature vectors using mahalanobis distance. The first step is to construct feature matrix of $6{\times}2$ size using frequency response of the time-series accelerometer data. Singular value decomposition makes it possible to find the dominant eigenvalue and its corresponding eigenvector. We use the eigenvector as the feature vector of the driving pattern. We conducted real experiments using three drivers to see the effects of recognition. Although there exists differences from individual drivers, we showed that driving patterns can be recognized with about 80% accuracy. Further research topics will include the development of aggressive driving warning system by improving the proposed technique and combining with post-processing of accelerometer signals.

• Journal of the Korean Institute of Rural Architecture
• /
• v.8 no.1
• /
• pp.80-89
• /
• 2006

This study is derived from relation between the natural landscape and architectural landscape. The type of landscape in Donghae seaside consist of three; Road, Mountain, Seaside. And we selected three landscapes about individual housing; Western, Traditional, Modern. This paper is analyzed 18-simulation scenes, which evaluated with semantic differential method in using 12-bipolar adjectives. The results of this study are as follows(ref: table 6). 1)The housing of western style do not correspond with landscape of Road, but landscape of mountain and seaside were suitable to the western style. 2)Mountain in Donghae seaside harmonizes with housing of traditional style. 3)Even though the housing of modern style were marked low assessment in three landscape, we found out relation, modern housing was well-matched with load landscape.

• Journal of information and communication convergence engineering
• /
• v.6 no.1
• /
• pp.100-104
• /
• 2008

Generally, it is difficult to find constant patterns on identifiers in a container image, since the identifiers are not normalized in color, size, and position, etc. and their shapes are damaged by external environmental factors. This paper distinguishes identifier areas from background noises and removes noises by using an ART2-based quantization method and general morphological information on the identifiers such as color, size, ratio of height to width, and a distance from other identifiers. Individual identifier is extracted by applying the 8-directional contour tracking method to each identifier area. This paper proposes a refined ART2-based RBF network and applies it to the recognition of identifiers. Through experiments with 300 container images, the proposed algorithm showed more improved accuracy of recognizing container identifiers than the others proposed previously, in spite of using shorter training time.

• Proceedings of the KIEE Conference
• /
• 1992.07a
• /
• pp.427-430
• /
• 1992

In this paper, we proposed an algorithm for the musical note recognition. Firstly, a given bit-mapped music score image is converted to a set of individual note pattern images via vertical projection. Then, the pitch of a note is determinal by comparison in the note-head position with the reference five-lines. Also, the length of a note is found via leader clustering with a set of normalized note patterns. Finally, a datafile to play the music is obtained using the pitch and length of musical notes. Experimental results with a simple musical score image show that the proposed scheme is performed well.

• Journal of Institute of Control, Robotics and Systems
• /
• v.13 no.8
• /
• pp.749-753
• /
• 2007

In this paper, a generalized version of the Motion Silhouette Image(MSI) called the Generic Motion Silhouette Image (GMSI) is proposed for gait recognition. The GMSI is a gray-level image and involves the spatiotemporal information of individual motion. The GMSI not only generalizes the MSI but also reflects a flexible feature of a gait sequence. Along with the GMSI, we use the Principal Component Analysis(PCA) to reduce the dimensionality of the GMSI and the Nearest Neighbor(NN) for classification. We apply the proposed feature to NLPR database and compare it with the conventional MSI. Experimental results show the effectiveness of the GMSI.

• Phonetics and Speech Sciences
• /
• v.3 no.3
• /
• pp.115-121
• /
• 2011

This paper examines how L2 speech production varies according to learner's L2 proficiency level. L2 speech production variations are analyzed by quantitative measures at word and phone levels using Korean learners' English corpus. Word-level variations are analyzed using correctness to explain how speech realizations are different from the canonical forms, while accuracy is used for analysis at phone level to reflect phone insertions and deletions together with substitutions. The results show that speech production of learners with different L2 proficiency levels are considerably different in terms of performance and individual realizations at word and phone levels. These results confirm that speech production of non-native speakers varies according to their L2 proficiency levels, even though they share the same L1 background. Furthermore, they will contribute to improve non-native speech recognition performance of ASR-based English language educational system for Korean learners of English.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2003.05b
• /
• pp.90-95
• /
• 2003

For the safety driving of an automobile which is become individual requisites, a new Neural Network algorithm which recognized the load vehicles in real time is proposed. The proposed neural network technique is the real time computation method through the inter-node diffusion. In the network, a node corresponds to a state in the quantized input space. Each node is composed of a processing unit and fixed weights from its neighbor nodes as well as its input terminal. The most reliable algorithm derived for real time recognition of vehicles, is a dynamic programming based algorithm based on sequence matching techniques that would process the data as it arrives and could therefore provide continuously updated neighbor information estimates. Through several simulation experiments, real time reconstruction of the nonlinear image information is processed. 1-D LIPN hardware has been composed and various experiments with static and dynamic signals have been implemented.