• Journal of Digital Contents Society
• /
• v.19 no.1
• /
• pp.35-41
• /
• 2018

The Q-learning algorithm based on reinforcement learning is useful for learning the goal for one behavior at a time, using a combination of discrete states and actions. In order to learn multiple actions, applying a behavior-based architecture and using an appropriate behavior adjustment method can make a robot perform fast and reliable actions. Q-learning is a popular reinforcement learning method, and is used much for robot learning for its characteristics which are simple, convergent and little affected by the training environment (off-policy). In this paper, Q-learning algorithm is applied to a lamp robot to learn multiple behaviors (human recognition, desk object recognition). As the learning rate of Q-learning may affect the performance of the robot at the learning stage of multiple behaviors, we present the optimal multiple behaviors learning model by changing learning rate.

• The Journal of the Korea Contents Association
• /
• v.21 no.8
• /
• pp.1-9
• /
• 2021

Deep learning in computer vision has made accelerated improvement over a short period but large-scale learning data and computing power are still essential that required time-consuming trial and error tasks are involved to derive an optimal network model. In this study, we propose a similar image classification performance improvement method based on CR (Confusion Rate) that considers only the characteristics of the data itself regardless of network optimization or data reinforcement. The proposed method is a technique that improves the performance of the deep learning model by calculating the CRs for images in a dataset with similar characteristics and reflecting it in the weight of the Loss Function. Also, the CR-based recognition method is advantageous for image identification with high similarity because it enables image recognition in consideration of similarity between classes. As a result of applying the proposed method to the Resnet18 model, it showed a performance improvement of 0.22% in HanDB and 3.38% in Animal-10N. The proposed method is expected to be the basis for artificial intelligence research using noisy labeled data accompanying large-scale learning data.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.3
• /
• pp.268-272
• /
• 2009

In this paper, we propose a Gaussian Mixture Model (GMM) based incremental robust adaptation with a forgetting factor for the speaker verification. Speaker recognition system uses a speaker model adaptation method with small amounts of data in order to obtain a good performance. However, a conventional adaptation method has vulnerable to the outlier from the irregular utterance variations and the presence noise, which results in inaccurate speaker model. As time goes by, a rate in which new data are adapted to a model is reduced. The proposed algorithm uses an incremental robust adaptation in order to reduce effect of outlier and use forgetting factor in order to maintain adaptive rate of new data on GMM based speaker model. The incremental robust adaptation uses a method which registers small amount of data in a speaker recognition model and adapts a model to new data to be tested. Experimental results from the data set gathered over seven months show that the proposed algorithm is robust against outliers and maintains adaptive rate of new data.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.47 no.3
• /
• pp.95-103
• /
• 2024

The diversity of smart EV(electric vehicle)-related industries is increasing due to the growth of battery-based eco-friendly electric vehicle component material technology, and labor-intensive industries such as logistics, manufacturing, food, agriculture, and service have invested in and studied automation for a long time. Accordingly, various types of robots such as autonomous mobile robots and collaborative robots are being utilized for each process to improve industrial engineering such as optimization, productivity management, and work management. The technology that should accompany this unmanned automobile industry is unmanned automatic charging technology, and if autonomous mobile robots are manually charged, the utility of autonomous mobile robots will not be maximized. In this paper, we conducted a study on the technology of unmanned charging of autonomous mobile robots using charging terminal docking and undocking technology using an unmanned charging system composed of hardware such as a monocular camera, multi-joint robot, gripper, and server. In an experiment to evaluate the performance of the system, the average charging terminal recognition rate was 98%, and the average charging terminal recognition speed was 0.0099 seconds. In addition, an experiment was conducted to evaluate the docking and undocking success rate of the charging terminal, and the experimental results showed an average success rate of 99%.

• Journal of IKEEE
• /
• v.18 no.4
• /
• pp.471-477
• /
• 2014

In this paper, we propose a low-cost license plate recognition system based on smart cam system using Android. The proposed system consists of a portable device and server. Potable device Hardware consists of ARM Cortex-A9 (S5PV210) processor control unit, a power supply device, wired and wireless communication, input/output unit. We develope Linux kernel and dedicated device driver for WiFi module and camera. The license plate recognition algorithm is consisted of setting candidate plates areas with canny edge detector, extracting license plate number with Labeling, recognizing with template matching, etc. The number that is recognized by the device is transmitted to the remote server via the user mobile phone, and the server re-transfer the vehicle information in the database to the portable device. To verify the utility of the proposed system, user photographs the license plate of any vehicle in the natural environment. Confirming the recognition result, the recognition rate was 95%. The proposed system was suitable for low cost portable license plate recognition device, it enabled the stability of the system when used long time by using the Android operating system.

• The Journal of the Acoustical Society of Korea
• /
• v.35 no.1
• /
• pp.74-82
• /
• 2016

In this paper, a new automatic speech recognition (ASR) was proposed where ultrasonic doppler signals were used, instead of conventional speech signals. The proposed method has the advantages over the conventional speech/non-speech-based ASR including robustness against acoustic noises and user comfortability associated with usage of the non-contact sensor. In the method proposed herein, 40 kHz ultrasonic signal was radiated toward to the mouth and the reflected ultrasonic signals were then received. Frequency shift caused by the doppler effects was used to implement ASR. The proposed method employed multi-channel ultrasonic signals acquired from the various locations, which is different from the previous method where single channel ultrasonic signal was employed. The PCA(Principal Component Analysis) coefficients were used as the features of ASR in which hidden markov model (HMM) with left-right model was adopted. To verify the feasibility of the proposed ASR, the speech recognition experiment was carried out the 60 Korean isolated words obtained from the six speakers. Moreover, the experiment results showed that the overall word recognition rates were comparable with the conventional speech-based ASR methods and the performance of the proposed method was superior to the conventional signal channel ASR method. Especially, the average recognition rate of 90 % was maintained under the noise environments.

• Journal of Broadcast Engineering
• /
• v.23 no.2
• /
• pp.246-252
• /
• 2018

Hand gestures are attracting attention as a NUI (Natural User Interface) of wearable devices such as smart glasses. Recently, to support efficient media consumption in IoT (Internet of Things) and wearable environments, the standardization of IoMT (Internet of Media Things) is in the progress in MPEG. In IoMT, it is assumed that hand gesture detection and recognition are performed on a separate device, and thus provides an interoperable interface between these modules. Meanwhile, deep learning based hand gesture recognition techniques have been recently actively studied to improve the recognition performance. In this paper, we propose a method of hand gesture recognition based on CNN (Convolutional Neural Network) for various applications such as media consumption in wearable devices which is one of the use cases of IoMT. The proposed method detects hand contour from stereo images acquisitioned by smart glasses using depth information and color information, constructs data sets to learn CNN, and then recognizes gestures from input hand contour images. Experimental results show that the proposed method achieves the average 95% hand gesture recognition rate.

• Journal of the Institute of Convergence Signal Processing
• /
• v.11 no.4
• /
• pp.289-297
• /
• 2010

In this paper, A method to prevent the external acoustic noise from being misrecognized as the speech recognition object is presented in the speech activity detection process for the speech recognition. Also this paper confirmed besides the acoustic energy to the lip movement image signals. First of all, the successive images are obtained through the image camera for personal computer and the lip movement whether or not is discriminated. The next, the lip movement image signal data is stored in the shared memory and shares with the speech recognition process. In the mean time, the acoustic energy whether or not by the utterance of a speaker is verified by confirming data stored in the shared memory in the speech activity detection process which is the preprocess phase of the speech recognition. Finally, as a experimental result of linking the speech recognition processor and the image processor, it is confirmed to be normal progression to the output of the speech recognition result if face to the image camera and speak. On the other hand, it is confirmed not to the output the result of the speech recognition if does not face to the image camera and speak. Also, the initial feature values under off-line are replaced by them. Similarly, the initial template image captured while off-line is replaced with a template image captured under on-line, so the discrimination of the lip movement image tracking is raised. An image processing test bed was implemented to confirm the lip movement image tracking process visually and to analyze the related parameters on a real-time basis. As a result of linking the speech and image processing system, the interworking rate shows 99.3% in the various illumination environments.

• Journal of the Korean Society of Safety
• /
• v.33 no.6
• /
• pp.28-36
• /
• 2018

There was a difference in recognition of respirators according to the educational performance environment. they were showed higher recognition of respirators of group by internal and external mix trainer, less than 6 months, over 1hour, more than 5 times, variety of education. To identify the relationship between types of job classification(typical and atypical)and the levels of recognition of respirators, a total of 153 workers in a business workplace. mainly, typical workers showed higher recognition of respirators than atypical workers. Training of correct wearing showed high demands both typical and atypical workers. Descriptive statistics(SAS ver 9.2)was performed. the results of recognition of respirators were analyzed the mean and standard deviation by t-test, and anova, fit factor is used geometric means(geometric standard deviation), paired t-test, Wilcoxon analysis(P=0.05). Particulate filtering facepiece respirators (PFFR) is one of the most widely used items of personal protective equipments, and a tight fit of the respirators on the wearers is critical for the protection effectiveness. In order to effectively protect the workers through the respirators, it is important to find and evaluate the ways that can be readily applicable at the workplace to improve the fit of the respirators. This study was designed to evaluate effects of mask style (cup or foldable type) and donning training on fit factors (FF) of the respirators, since these are available at various workplace, especially at small business workplace. A total of 40 study subjects, comprised of employment type workers in metalworking industries, were enrolled in this study. The FF were quantitatively measured before and after training related to the proper donning and use of cup or foldable-type respirators. The pass/fail criterion of FF was set at 100. After the donning training for the cup-type mask, fit test were increased by 769%. but foldable-type mask was also increased after the donning training, the GM of FF for the foldable-type mask and it's increase rate were smaller as compared to the cup-type mask. Furthermore, the differences of the increase rates of the GM of FF in employment type of the subjects were not significantly for the foldable-type mask. These results imply that the raining on the donning and use of PFFR can enhance the protection effectiveness of cup or foldable-type mask, and that the training effects for the foldable-type mask is less significant than that for the cup-type mask. Therefore, it is recommended that the donning training and fit tests should be conducted before the use of the PFFR, and listening to workers opinion regularly.

• Journal of Convergence for Information Technology
• /
• v.11 no.6
• /
• pp.33-39
• /
• 2021

Speech recognition technology is being combined with deep learning and is developing at a rapid pace. In particular, voice recognition services are connected to various devices such as artificial intelligence speakers, vehicle voice recognition, and smartphones, and voice recognition technology is being used in various places, not in specific areas of the industry. In this situation, research to meet high expectations for the technology is also being actively conducted. Among them, in the field of natural language processing (NLP), there is a need for research in the field of removing ambient noise or unnecessary voice signals that have a great influence on the speech recognition recognition rate. Many domestic and foreign companies are already using the latest AI technology for such research. Among them, research using a convolutional neural network algorithm (CNN) is being actively conducted. The purpose of this study is to determine the non-voice section from the user's speech section through the convolutional neural network. It collects the voice files (wav) of 5 speakers to generate learning data, and utilizes the convolutional neural network to determine the speech section and the non-voice section. A classification model for discriminating speech sections was created. Afterwards, an experiment was conducted to detect the non-speech section through the generated model, and as a result, an accuracy of 94% was obtained.