• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.5
• /
• pp.72-83
• /
• 2011

In this paper, we propose a method to detect and recover the occluded parts of face images using the correlation between pairs of pixels. In a training stage, correlation coefficients between every pairs of pixels are calculated using the occlusion-free face images. Once a new occluded face image is shown, the occluded area is detected and recovered using the correlation coefficients obtained in the training stage. We compare the performance of the proposed method with the conventional method based on PCA. The results show that the proposed method detects and recovers occluded area with much smaller noises than the conventional PCA based method. Moreover, recovered images by the proposed method were more smooth with reduced blurring effect.

• International Journal of Computer Science & Network Security
• /
• v.21 no.1
• /
• pp.97-106
• /
• 2021

Social networking platforms have become a smart way for people to interact and meet on internet. It provides a way to keep in touch with friends, families, colleagues, business partners, and many more. Among the various social networking sites, Twitter is one of the fastest-growing sites where users can read the news, share ideas, discuss issues etc. Due to its vast popularity, the accounts of legitimate users are vulnerable to the large number of threats. Spam and Malware are some of the most affecting threats found on Twitter. Therefore, in order to enjoy seamless services it is required to secure Twitter against malicious users by fixing them in advance. Various researches have used many Machine Learning (ML) based approaches to detect spammers on Twitter. This research aims to devise a secure system based on Hybrid Similarity Cosine and Soft Cosine measured in combination with Genetic Algorithm (GA) and Artificial Neural Network (ANN) to secure Twitter network against spammers. The similarity among tweets is determined using Cosine with Soft Cosine which has been applied on the Twitter dataset. GA has been utilized to enhance training with minimum training error by selecting the best suitable features according to the designed fitness function. The tweets have been classified as spammer and non-spammer based on ANN structure along with the voting rule. The True Positive Rate (TPR), False Positive Rate (FPR) and Classification Accuracy are considered as the evaluation parameter to evaluate the performance of system designed in this research. The simulation results reveals that our proposed model outperform the existing state-of-arts.

• IEIE Transactions on Smart Processing and Computing
• /
• v.5 no.4
• /
• pp.256-266
• /
• 2016

Concurrently detected and annotated abnormal events can have a significant impact on surveillance systems. By considering the specific domain of pedestrian trajectories, this paper presents two main contributions. First, as introduced in much of the work on trajectory-based anomaly detection in the literature, only information about pedestrian paths, such as direction and speed, is considered. Differing from previous work, this paper proposes a framework that deals with additional types of trajectory-based anomalies. These abnormal events take places when a person enters prohibited areas. Those restricted regions are constructed by an online learning algorithm that uses surrounding information, including detected pedestrians and background scenes. Second, a simple data-boosting technique is introduced to overcome a lack of training data; such a problem particularly challenges all previous work, owing to the significantly low frequency of abnormal events. This technique only requires normal trajectories and fundamental information about scenes to increase the amount of training data for both normal and abnormal trajectories. With the increased amount of training data, the conventional abnormal trajectory classifier is able to achieve better prediction accuracy without falling into the over-fitting problem caused by complex learning models. Finally, the proposed framework (which annotates tracks that enter prohibited areas) and a conventional abnormal trajectory detector (using the data-boosting technique) are integrated to form a united detector. Such a detector deals with different types of anomalous trajectories in a hierarchical order. The experimental results show that all proposed detectors can effectively detect anomalous trajectories in the test phase.

• Journal of the Korea Society of Computer and Information
• /
• v.16 no.7
• /
• pp.25-34
• /
• 2011

In this paper, we propose a new golf putting training system which we call "X-Putt". X-Putt analyzes putting strokes by measuring putter face angle and path. To do this, we improved the sonar-based localization scheme used by previous localization techniques. As a result, X-Putt can measure putter's location within the error range, ${\pm}0.9cm$ and putter face angle within ${\pm}1.5^{\circ}$. Additionally, we built an user application that has an easy-to-use interface for analyzing the strokes after training.

• The Journal of Korean Association of Computer Education
• /
• v.22 no.3
• /
• pp.113-121
• /
• 2019

The purpose of this study is to develop TPACK-P teachers' training program based on backward design included understanding of backward design principles in order to enhance teachers' professionalism of software education and informatics education. This teachers' training program was conducted for non-informatics teachers in 32 hours. As a results, teachers showed that post-test result of SE-TEBI was higher than pre-test and the t-test results showed that the total value of SE-TE was t=-3.541(p< .01), subscale PSETE was t=-3.559( p< .01) and SEOE was t=-2.258(p< .05). Therefore, we demonstrate that it helps improve the teaching professionalism of teachers. However, as a result of the test, teachers' PSETE tend to be lower in SEOE, which is interpreted as a lack of confidence in the lessons. Therefore, in further study, it is necessary to develop a teacher education program that can promote the confidence of teachers.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.1
• /
• pp.50-56
• /
• 2020

With the rapid development of deep neural network theory and application research, the effect of face detection has been improved. However, due to the complexity of deep neural network calculation and the high complexity of the detection environment, how to detect face quickly and accurately becomes the main problem. This paper is based on the relatively simple model of the MTCNN model, using FDDB (Face Detection Dataset and Benchmark Homepage), LFW (Field Label Face) and FaceScrub public datasets as training samples. At the same time of sorting out and introducing MTCNN(Multi-Task Cascaded Convolutional Neural Network) model, it explores how to improve training speed and Increase performance at the same time. In this paper, the dynamic image pyramid technology is used to replace the traditional image pyramid technology to segment samples, and OHEM (the online hard example mine) function in MTCNN model is deleted in training, so as to improve the training speed.

• Journal of Korea Multimedia Society
• /
• v.25 no.10
• /
• pp.1375-1385
• /
• 2022

In this paper, we propose a semi-supervised domain adaptation solution to deal with practical face recognition (FR) scenarios where a single face image for each target identity (to be recognized) is only available in the training phase. Main goal of the proposed method is to reduce the discrepancy between the target and the source domain face images, which ultimately improves FR performances. The proposed method is based on the Domain Adatation network (DAN) using an MMD loss function to reduce the discrepancy between domains. In order to train more effectively, we develop a novel loss function learning strategy in which MMD loss and cross-entropy loss functions are adopted by using different weights according to the progress of each epoch during the learning. The proposed weight adoptation focuses on the training of the source domain in the initial learning phase to learn facial feature information such as eyes, nose, and mouth. After the initial learning is completed, the resulting feature information is used to training a deep network using the target domain images. To evaluate the effectiveness of the proposed method, FR performances were evaluated with pretrained model trained only with CASIA-webface (source images) and fine-tuned model trained only with FERET's gallery (target images) under the same FR scenarios. The experimental results showed that the proposed semi-supervised domain adaptation can be improved by 24.78% compared to the pre-trained model and 28.42% compared to the fine-tuned model. In addition, the proposed method outperformed other state-of-the-arts domain adaptation approaches by 9.41%.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.9 no.3
• /
• pp.473-479
• /
• 2005

A key issue toward mobile multimedia communications is to create technologies for broadband signal transmission that ran support high quality services. Such a broadband mobile communications system should be able to overcome severe distortion caused by time-varying multi-path fading channel, while providing high spectral efficiency and low power consumption. For these reasons, an adaptive suboptimum decision feedback equalize. (DFE) for the single-carrier short-burst transmissions system is considered as one of the feasible solutions. For the performance improvement of the system with the short-burst format including the short training sequence, in this paper, the multiple-training least mean square (MTLMS) based DFE scheme with soft decision feedback is proposed and its performance is investigated in mobile wireless channels throughout computer simulation.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2006.05a
• /
• pp.950-953
• /
• 2006

A key issue toward mobile multimedia communications is to create technologies for broadband signal transmission that can support high quality services. Such a broadband mobile communications system should be able to overcome severe distortion caused by time-varying multi-path fading channel, while providing high spectral efficiency and low power consumption. For these reasons, an adaptive suboptimum decision feedback equalizer (DFE) for the single-carrier short-burst transmissions system is considered as one of the feasible solutions. For the performance improvement of the system with the short-burst format including the short training sequence, in this paper, the multiple-training least mean square (MTLMS) based DFE scheme with soft decision feedback is proposed and its performance is investigated in mobile wireless channels throughout computer simulation.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.10
• /
• pp.3378-3393
• /
• 2014

We have witnessed the rapid development of information technology in recent years. One of the key phenomena is the fast, near-exponential increase of data. Consequently, most of the traditional data classification methods fail to meet the dynamic and real-time demands of today's data processing and analyzing needs--especially for continuous data streams. This paper proposes an improved incremental learning algorithm for a large-scale data stream, which is based on SVM (Support Vector Machine) and is named DS-IILS. The DS-IILS takes the load condition of the entire system and the node performance into consideration to improve efficiency. The threshold of the distance to the optimal separating hyperplane is given in the DS-IILS algorithm. The samples of the history sample set and the incremental sample set that are within the scope of the threshold are all reserved. These reserved samples are treated as the training sample set. To design a more accurate classifier, the effects of the data volumes of the history sample set and the incremental sample set are handled by weighted processing. Finally, the algorithm is implemented in a cloud computing system and is applied to study user behaviors. The results of the experiment are provided and compared with other incremental learning algorithms. The results show that the DS-IILS can improve training efficiency and guarantee relatively high classification accuracy at the same time, which is consistent with the theoretical analysis.