• Journal of Broadcast Engineering
• /
• v.25 no.3
• /
• pp.374-385
• /
• 2020

In this paper, we propose a preprocessing technique to solve the problems of action recognition with Equirectangular Projection (ERP) video. The preprocessing technique proposed in this paper assumes the person object as the subject of action, that is, the Object of Interest (OOI), and the surrounding area of the OOI as the ROI. The preprocessing technique consists of three modules. I) Recognize person object in the image with object recognition model. II) Create a saliency map from the input image. III) Select subject of action using recognized person object and saliency map. The subject boundary box of the selected action is input to the action recognition model in order to improve the action recognition performance. When comparing the performance of the proposed preprocessing method to the action recognition model and the performance of the original ERP image input method, the performance is improved up to 99.6%, and the action is obtained when only the OOI is detected. It can also see the effects of related video summaries.

• Journal of the Korean Society for Precision Engineering
• /
• v.22 no.4
• /
• pp.121-127
• /
• 2005

There is a necessity for the communication between intelligent robots and human beings because of wide spread use of them. Gesture recognition is currently being studied in regards to better conversing. On the basis of previous research, however, the gesture recognition algorithms appear to require not only complicated algorisms but also separate training process for high recognition rates. This study suggests a gesture recognition algorithm based on computer vision system, which is relatively simple and more efficient in recognizing various human gestures. After tracing the hand gesture using a marker, direction changes of the gesture trajectory were analyzed to determine the simple gesture code that has minimal information to recognize. A map is developed to recognize the gestures that can be expressed with different gesture codes. Through the use of numerical and geometrical trajectory, the advantages and disadvantages of the suggested algorithm was determined.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.5 no.1
• /
• pp.228-246
• /
• 2011

A 3D model based approach for a face representation and recognition algorithm has been investigated as a robust solution for pose and illumination variation. Since a generative 3D face model consists of a large number of vertices, a 3D model based face recognition system is generally inefficient in computation time and complexity. In this paper, we propose a novel 3D face representation algorithm based on a pixel to vertex map (PVM) to optimize the number of vertices. We explore shape and texture coefficient vectors of the 3D model by fitting it to an input face using inverse compositional image alignment (ICIA) to evaluate face recognition performance. Experimental results show that the proposed face representation and recognition algorithm is efficient in computation time while maintaining reasonable accuracy.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2009.01a
• /
• pp.173-176
• /
• 2009

본 논문에서는 피부색 정보와 눈, 입의 위치를 찾아 실시간으로 얼굴을 인식하는 랩을 제안한다. 먼저 노이즈를 제거하여 얼굴 후보 영역을 지정한다. 지정된 얼굴 후보 영역에서 눈과 입을 찾고, 찾은 눈과 입 사이의 영역에서 에지를 탐색하여 코의 존재 유무를 검증하고 이를 바탕으로 얼굴인지 판단하는 절차를 따른다. 제안한 기법은 피부색 검출을 위해 YCbCr 을 이용하여 피부 영역을 찾고 지정한 피부 영역에서 노이즈를 제거한 후, Eye Map의 EyeMapC 연산을 통해 눈을 Lip Map을 통해 입을 찾는다. 찾아낸 눈과 입의 사이의 영역에서 Canny Edge 연산을 수행하여 코의 존재 유무를 판단하여 최종적인 얼굴 영역을 판별하는 방법을 제안한다.

• Smart Media Journal
• /
• v.12 no.2
• /
• pp.9-14
• /
• 2023

In this paper, we proposed a face recognition network which attempts to use more facial features awhile using smaller number of training sets. When combining the neural network together for face recognition, we want to use networks that use different part of the facial features. However, the network training chooses randomly where these facial features are obtained. Other hand, the judgment basis of the network model can be expressed as a saliency map through gradCAM. Therefore, in this paper, we use gradCAM to visualize where the trained face recognition model has made a observations and recognition judgments. Thus, the network combination can be constructed based on the different facial features used. Using this approach, we trained a network for small face recognition problem. In an simple toy face recognition example, the recognition network used in this paper improves the accuracy by 1.79% and reduces the equal error rate (EER) by 0.01788 compared to the conventional approach.

• Journal of Korean Society of Rural Planning
• /
• v.16 no.4
• /
• pp.13-22
• /
• 2010

Rural landscape is an outcome of residents' life activity based on natural environment. Unlike city, rural residents make their own landscape over a period of time interacting with nature through cultivating and building houses and huts based on the background. Therefore, residents' role in rural area is of greater importance than city's and their recognition of landscape is a key factor to evaluate and manage rural landscape. Landscape Evaluation Map which utilizing Feeling Map method is a evaluation tool to [md out residents' recognition of landscape. In this tool, responses evaluate landscape around their living space and mark color dots which mean landscape grade on a map. This research is to examine effectiveness and applicability of the tool, Landscape Evaluation Map, which is recommended to estimate residents' evaluation of landscape. Through analyzing 7 cases of field application, the effectiveness of Landscape Evaluation Map has been verified and also demerits have been drawn. After modifying detailed techniques and developing resident education, Landscape evaluation map could be applied to [md out landscape resources rather than to evaluate whole rural landscape.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2005.05a
• /
• pp.91-96
• /
• 2005

In this study, researchers developed the estimative algorithm for artificial defects in semiconductor packages and performed it by pattern recognition technology. For this purpose, the estimative algorithm was included that researchers made software with MATLAB. The software consists of some procedures including ultrasonic image acquisition, equalization filtering, Self-Organizing Map and Backpropagation Neural Network. Self-Organizing Map and Backpropagation Neural Network are belong to methods of Neural Networks. And the pattern recognition technology has applied to classify three kinds of detective patterns in semiconductor packages: Crack, Delamination and Normal. According to the results, we were confirmed that estimative algorithm was provided the recognition rates of $75.7\%$ (for Crack) and $83_4\%$ (for Delamination) and $87.2\%$ (for Normal).

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.1E
• /
• pp.20-24
• /
• 1994

A new speech recognition model, DLSMM(Dynamically Localized Self-organizing Map Model) and its effective training algorithm are proposed in this paper. In DLSMM, temporal and spatial distortions of speech are efficiently normalized by dynamic programming technique and localized self-organizing maps, respectively. Experiments on Korean digits recognition have been carried out. DLSMM has smaller Experiments on Korean digits recognition have been carried out. DLSMM has smaller connections than predictive neural network models, but it has scored a little high recognition rate.

• International Journal of Precision Engineering and Manufacturing
• /
• v.5 no.3
• /
• pp.11-18
• /
• 2004

In this paper, the classification of artificial defects in semiconductor packages is studied by using pattern recognition technology. For this purpose, the pattern recognition algorithm includes the user made MATLAB code. And preprocess is made of the image process and self-organizing map, which is the input of the back-propagation neural network and the dimensionality reduction method, The image process steps are data acquisition, equalization, binary and edge detection. Image process and self-organizing map are compared to the preprocess method. Also the pattern recognition technology is applied to classify two kinds of defects in semiconductor packages: cracks and delaminations.

• Journal of Biomedical Engineering Research
• /
• v.28 no.4
• /
• pp.494-502
• /
• 2007

Ubiquitous location based services (u-LBS) will be interested to an important services. They can easily recognize object position at anytime, anywhere. At present, many researchers are making a study of the position recognition and tracking. This paper consists of postion recognition and user identification system. The position recognition is based on location under services (LBS) using a signal strength map, a database is previously made use of empirical measured received signal strength indicator (RSSI). The user identification system automatically controls instruments which is located in home. Moreover users are able to measures body signal freely. We implemented the multi-hop routing method using the Star-Mesh networks. Also, we use the sensor devices which are satisfied with the IEEE 802.15.4 specification. The used devices are the Nano-24 modules in Octacomm Co. Ltd. A RSSI is very important factor in position recognition analysis. It makes use of the way that decides position recognition and user identification in narrow indoor space. In experiments, we can analyze properties of the RSSI, draw the parameter about position recognition. The experimental result is that RSSI value is attenuated according to increasing distances. It also derives property of the radio frequency (RF) signal. Moreover, we express the monitoring program using the Microsoft C#. Finally, the proposed methods are expected to protect a sudden death and an accident in home.