• Phonetics and Speech Sciences
• /
• v.10 no.1
• /
• pp.33-38
• /
• 2018

Although deep neural network-based acoustic models have greatly improved the performance of automatic speech recognition (ASR), reverberation still degrades the performance of distant speech recognition in indoor environments. In this paper, we adopt the DenseNet, which has shown great performance results in image classification tasks, to improve the performance of reverberant speech recognition. The DenseNet enables the deep convolutional neural network (CNN) to be effectively trained by concatenating feature maps in each convolutional layer. In addition, we extend the concept of multi-resolution CNN to multi-resolution DenseNet for robust speech recognition in reverberant environments. We evaluate the performance of reverberant speech recognition on the single-channel ASR task in reverberant voice enhancement and recognition benchmark (REVERB) challenge 2014. According to the experimental results, the DenseNet-based acoustic models show better performance than do the conventional CNN-based ones, and the multi-resolution DenseNet provides additional performance improvement.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.81.2-81
• /
• 2002

$\textbullet$ Introduction- Concept and meaning of the emotional Recognition $\textbullet$ The feature of 4-emotions $\textbullet$ Pitch(approach) $\textbullet$ Simulator-structure, RNN(learning algorithm), evaluation function, solution search method $\textbullet$ Result

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.1375-1378
• /
• 1997

In this paper, we presents the algorithm which is to recognize the traffic sign on the road the traffic signal in a video image for autonomous navigation. First, the rocognition of traffic sign on the road can be detected using boundary point estimation form some scan-lines within the lane deducted. For this algorithm, index matrix method is used to detemine what sign is. Then, the traffic signal recognition is performed by usign the window minified by several scan-lines which position may be expected. For this algoritm, line profile concept is adopted.

• Korean Journal of Child Studies
• /
• v.34 no.4
• /
• pp.19-36
• /
• 2013

This study examined concept development in children in terms of sound propagation. The subjects consisted of 240 five-, seven-, and ten-year-old children from three kindergartens and two primary schools respectively. The instrument used for the purposes of this study was the examination instrument on sound as developed by the SPACE Project in the UK. The concepts related to sound were classified into five stages including : no recognition, egocentric concept, initial mental models, synthetic mental models, and scientific mental models. The results revealed the existence of significant differences in terms of the types of concept that children were aware of, according to age and context. Most five-, seven-, and ten-year-old children revealed egocentric concepts related to every sounds, drum, rubber band context, however, most five-, seven-, and ten-year-old children revealed their recognition of the synthetic models of sound propagation when using the string telephone context. These results have implications for the contents of science education for children when it comes to the development of concepts related to sound propagation.

• Journal of Gifted/Talented Education
• /
• v.24 no.3
• /
• pp.339-358
• /
• 2014

On the subjects of elementary 3-rd grade three child prodigies who learned primary concept of division, this study explored how they could compose schema and transformed schema through recognition of precise concepts and linking with the contents of division of fraction. That is to say, this study examined in depth what schema and transformed schema as primary concept of division they composed to get relational understanding of division of fraction, and how they used the schema and transformed schema composed by themselves to approach problem solving as well as how they transformed the schema in their concept composition and problem solving competence. As a result, it was found that learning of primary concept of division played a key role of composing schema and transformed schema needed for coping with division of fraction, and that at this time, composition of the transformed schema and transformed schema derived from the recognition of primary concept of division could play the inevitable role of problem solving for division of fraction.

• Journal of Institute of Control, Robotics and Systems
• /
• v.12 no.2
• /
• pp.130-137
• /
• 2006

In this paper, we propose a method of selecting new types of rectangle features that are suitable for facial expression recognition. The basic concept in this paper is similar to Viola's approach, which is used for face detection. Instead of previous Haar-like features we choose rectangle features for facial expression recognition among all possible rectangle types in a 3${\times}$3 matrix form using the AdaBoost algorithm. The facial expression recognition system constituted with the proposed rectangle features is also compared to that with previous rectangle features with regard to its capacity. The simulation and experimental results show that the proposed approach has better performance in facial expression recognition.

• Culinary science and hospitality research
• /
• v.11 no.2
• /
• pp.1-13
• /
• 2005

This study focused on the consumers' recognition of regional cuisine in Kyonggi province. Above all, the study was conducted being related to the tourism merchandising strategy based on the purposes of this study such as the establishment of concept of regional cuisine in Kyonggi province, its recognition and interest, and literature study was also conducted with empirical study. The results are as follows: First, to the question about whether they know the regional cuisine in Kyonggi province or not, 44% of respondents answered they don't know it well, which means recognition is very low. Second, to the question about the recognition of regional cuisine in Kyonggi province, Sujebi(a clear soup with wheat flakes in it) and Nengkonggugsu(iced bean noodles) of main dish, Samgyetang (chicken broth with ginseng and other ingredients) and Galbitang(beef-rib soup) of side dish, Moguachunghwachae(honeyed juice mixed with a Chinese quince as a punch) of drinks, and Kaedduck(a pie-shaped cake made of some rough flower) of desserts were highly recognized.

• 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.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.2
• /
• pp.757-770
• /
• 2020

We propose a hand gesture recognition method that is compatible with a head-up display (HUD) including small processing resource. For fast link adaptation with HUD, it is necessary to rapidly process gesture recognition and send the minimum amount of driver hand gesture data from the wearable device. Therefore, we use a method that recognizes each hand gesture with an inertial measurement unit (IMU) sensor based on revised correlation matching. The method of gesture recognition is executed by calculating the correlation between every axis of the acquired data set. By classifying pre-defined gesture values and actions, the proposed method enables rapid recognition. Furthermore, we evaluate the performance of the algorithm, which can be implanted within wearable bands, requiring a minimal process load. The experimental results evaluated the feasibility and effectiveness of our decomposed correlation matching method. Furthermore, we tested the proposed algorithm to confirm the effectiveness of the system using pre-defined gestures of specific motions with a wearable platform device. The experimental results validated the feasibility and effectiveness of the proposed hand gesture recognition system. Despite being based on a very simple concept, the proposed algorithm showed good performance in recognition accuracy.

• Journal of Families and Better Life
• /
• v.25 no.2 s.86
• /
• pp.59-69
• /
• 2007

The purpose of this analysis is to investigate the effect of leading characters in picture books on self concept in children. For this study, we selected a total of 55 picture books, all of which were best sellers in four internet book stores. The method used for the study was based on content analysis. The results of this analysis were as follows: First, the gender of leading characters was male more often than female, and the majority of picture books were creative story books and translation books. Second, subordinate parts related to the self-concept were most frequently found in the social self-concept. Social self-concept was positively expressed through the relationship between friend and family characters in story books. Cognitive self-concept was the least. Third, when the frequency of positive and negative expressions in picture books was analyzed, the social self-concept was found to be positively expressed in the subordinate parts of relationships between friends and family. The social self-concept was negatively expressed, however, in the subordinate parts of relationship between non-significant others, and the emotional self-concept was also negatively expressed in the subordinate parts of the emotional recognition and expression. In summary. instances of cognitive self-concept and positive expression were found to be more significant than negative ones.