• International Journal of Contents
• /
• 제15권4호
• /
• pp.59-64
• /
• 2019

In recent years, emotion recognition has been an interesting and challenging topic. Compared to facial expressions and speech modality, gesture-based emotion recognition has not received much attention with only a few efforts using traditional hand-crafted methods. These approaches require major computational costs and do not offer many opportunities for improvement as most of the science community is conducting their research based on the deep learning technique. In this paper, we propose an end-to-end deep learning approach for classifying emotions based on bodily gestures. In particular, the informative keyframes are first extracted from raw videos as input for the 3D-CNN deep network. The 3D-CNN exploits the short-term spatiotemporal information of gesture features from selected keyframes, and the convolutional LSTM networks learn the long-term feature from the features results of 3D-CNN. The experimental results on the FABO dataset exceed most of the traditional methods results and achieve state-of-the-art results for the deep learning-based technique for gesture-based emotion recognition.

• Journal of Information Processing Systems
• /
• 제16권5호
• /
• pp.1145-1157
• /
• 2020

A convolution neural networks (CNNs) has demonstrated outstanding performance compared to other algorithms in the field of face recognition. Regarding the over-fitting problem of CNN, researchers have proposed a residual network to ease the training for recognition accuracy improvement. In this study, a novel face recognition model based on game theory for call-over in the classroom was proposed. In the proposed scheme, an image with multiple faces was used as input, and the residual network identified each face with a confidence score to form a list of student identities. Face tracking of the same identity or low confidence were determined to be the optimisation objective, with the game participants set formed from the student identity list. Game theory optimises the authentication strategy according to the confidence value and identity set to improve recognition accuracy. We observed that there exists an optimal mapping relation between face and identity to avoid multiple faces associated with one identity in the proposed scheme and that the proposed game-based scheme can reduce the error rate, as compared to the existing schemes with deeper neural network.

• 한국통신학회논문지
• /
• 제24권12B호
• /
• pp.2447-2454
• /
• 1999

본 논문은 기존 분류형 신경망의 인식성능을 향상시키기 위하여 프레임 정규화와 비선형 사후확률 추정법(N-APPEM)을 제안하고 한국어 숫자음에 대하여 예측형과 분류형 신경망으로 인식성능을 평가하였다. 실험결과 예측형 신경망에서 최고 98.0%의 인식률을 얻었다. 예측형 신경망은 네트워크가 입력패턴의 카테고리 수만큼 마련되는 복잡한 네트워크를 가지는 반면에 분류형 신경망은 단일 네트워크로 구성되며 프레임 정규화와 비선형 사후확률 추정법으로 85.5%까지 인식률을 향상시킬 수 있었으며 이는 기존의 방법보다 인식률이 12.0% 향상된 것이다.

• 지능정보연구
• /
• 제10권3호
• /
• pp.105-114
• /
• 2004

최근 생체인식에 대한 관심이 증가하면서 출입 통제나 사용자 인증과 같은 보안 분야에 적용이 활발히 진행되고 있다. 특히 얼굴인식은 생체인식 기술 중 사용자 편의성과 접촉 거부감이 적어 활용성이 증대되고 있으나 타 인식기술에 비해 인식 결과의 정확성과 재시도율(Re-attempt Rate)에 취약한 단점이 있다. 본 논문에서는 이러한 단점을 보완하기 위해 데이터 분류 방법(Data Classification Algorithm)으로 인식 결과를 재분류(Re-Classification)하는 접근법에 대해서 제안하고자 한다. 본 실험을 위해서 대표적인 형상 기반(Appearance-based) 알고리즘인 PCA를 사용하였고, 200명(총 얼굴 영상 200장)을 대상으로 제안한 재분류 접근법을 적용한 결과 재인식의 경우 성능이 향상되었음을 확인하였다.

• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2007년도 추계학술대회
• /
• pp.439-443
• /
• 2007

Nowadays, activity recognition becomes a hot topic in context-aware computing. In activity recognition, machine learning techniques have been widely applied to learn the activity models from labeled activity samples. Most of the existing work uses only one learning method for activity learning and is focused on how to effectively utilize the labeled samples by refining the learning method. However, not much attention has been paid to the use of multiple classifiers for boosting the learning performance. In this paper, we use two methods to generate multiple classifiers. In the first method, the basic learning algorithms for each classifier are the same, while the training data is different (ASTD). In the second method, the basic learning algorithms for each classifier are different, while the training data is the same (ADTS). Experimental results indicate that ADTS can effectively improve activity recognition performance, while ASTD cannot achieve any improvement of the performance. We believe that the classifiers in ADTS are more diverse than those in ASTD.

• 로봇학회논문지
• /
• 제16권3호
• /
• pp.276-282
• /
• 2021

The system for recognizing text in natural scenes has been applied in various industries. However, due to the change in brightness that occurs in nature such as light reflection and shadow, the text recognition performance significantly decreases. To solve this problem, we propose an adversarial shadow generation and training algorithm that is robust to shadow changes. The adversarial shadow generation and training algorithm divides the entire image into a total of 9 grids, and adjusts the brightness with 4 trainable parameters for each grid. Finally, training is conducted in a adversarial relationship between the text recognition model and the shaded image generator. As the training progresses, more and more difficult shaded grid combinations occur. When training with this curriculum-learning attitude, we not only showed a performance improvement of more than 3% in the ICDAR2015 public benchmark dataset, but also confirmed that the performance improved when applied to our's android application text recognition dataset.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권4호
• /
• pp.2078-2093
• /
• 2019

Visual smoke recognition is a challenging task due to large variations in shape, texture and color of smoke. To improve performance, we propose a novel smoke recognition method by combining dual-encoded features that are extracted from both spatial and Curvelet domains. A Curvelet transform is used to filter an image to generate fifty sub-images of Curvelet coefficients. Then we extract Local Binary Pattern (LBP) maps from these coefficient maps and aggregate histograms of these LBP maps to produce a histogram map. Afterwards, we encode the histogram map again to generate Dual-encoded Local Binary Patterns (Dual-LBP). Histograms of Dual-LBPs from Curvelet domain and Completed Local Binary Patterns (CLBP) from spatial domain are concatenated to form the feature for smoke recognition. Finally, we adopt Gaussian Kernel Optimization (GKO) algorithm to search the optimal kernel parameters of Support Vector Machine (SVM) for further improvement of classification accuracy. Experimental results demonstrate that our method can extract effective and reasonable features of smoke images, and achieve good classification accuracy.

• 한국컴퓨터정보학회논문지
• /
• 제27권9호
• /
• pp.41-47
• /
• 2022

본 논문에서는 자동표정인식을 위하여 얼굴 이미지 배열의 가운데 이미지를 예측하는 새롭고 간단한 자기주도학습 방법을 제안한다. 자동표정인식은 딥러닝 모델을 통해 높은 성능을 달성할 수 있으나 일반적으로 큰 비용과 시간이 투자된 대용량의 데이터 세트가 필요하고, 데이터 세트의 크기와 알고리즘의 성능이 비례한다. 제안하는 방법은 추가적인 데이터 세트 구축 없이 기존의 데이터 세트를 활용하여 자기주도학습을 통해 얼굴의 잠재적인 심층표현방법을 학습하고 학습된 파라미터를 전이시켜 자동표정인식의 성능을 향상한다. 제안한 방법은 CK+와 AFEW 8.0 두가지 데이터 세트에 대하여 높은 성능 향상을 보여주었고, 간단한 방법으로 큰 효과를 얻을 수 있음을 보여주었다.

• 한국응급구조학회지
• /
• 제26권1호
• /
• pp.37-56
• /
• 2022

Purpose: This study aimed to identify and present suitable recognition types of policy alternative for before and after response, according to the recognition types of problems in response to violence. Methods: This study investigated 36 EMT's of 17 cities and provinces nationwide. The study was approved by the Kongju National University Institute Review Board (KNU_IRB_2021-17). Data were collected from May 1, 2021 to August 30, 2021 and analyzed by Q factor analysis using the PC-QUNAL program. Results: Recognition types of the problem in 119 EMT's response to violence were described as "I type; lack of professional manpower," "II type; inadequate policy on violence," and "III type; lack of awareness on the emergency field." Recognition types of policy alternative on response to violence by 119 EMT's were described as "Itype; training and public relations oriented," "II type; work environment improvement," "III type; violence handling specialization demand," and "IV type; recovery support seeker." Conclusion: This study provides the foundation required to develop and implement the policies regarding the response to violence; therefore, contributing to EMT's provision.

• Journal of Information Processing Systems
• /
• 제19권6호
• /
• pp.730-744
• /
• 2023

The topic of this paper is the recognition of human activities using egocentric vision, particularly captured by body-worn cameras, which could be helpful for video surveillance, automatic search and video indexing. This being the case, it could also be helpful in assistance to elderly and frail persons for revolutionizing and improving their lives. The process throws up the task of human activities recognition remaining problematic, because of the important variations, where it is realized through the use of an external device, similar to a robot, as a personal assistant. The inferred information is used both online to assist the person, and offline to support the personal assistant. With our proposed method being robust against the various factors of variability problem in action executions, the major purpose of this paper is to perform an efficient and simple recognition method from egocentric camera data only using convolutional neural network and deep learning. In terms of accuracy improvement, simulation results outperform the current state of the art by a significant margin of 61% when using egocentric camera data only, more than 44% when using egocentric camera and several stationary cameras data and more than 12% when using both inertial measurement unit (IMU) and egocentric camera data.