• The Journal of Korean Institute of Next Generation Computing
• /
• v.15 no.6
• /
• pp.7-14
• /
• 2019

The automatic monitoring and detection of crowd behavior in the surveillance videos has obtained significant attention in the field of computer vision due to its vast applications such as security, safety and protection of assets etc. Also, the field of crowd analysis is growing upwards in the research community. For this purpose, it is very necessary to detect and analyze the crowd behavior. In this paper, we proposed a deep learning-based method which detects abnormal activities in surveillance cameras installed in a smart city. A fine-tuned VGG-16 model is trained on publicly available benchmark crowd dataset and is tested on real-time streaming. The CCTV camera captures the video stream, when abnormal activity is detected, an alert is generated and is sent to the nearest police station to take immediate action before further loss. We experimentally have proven that the proposed method outperforms over the existing state-of-the-art techniques.

• Smart Media Journal
• /
• v.7 no.4
• /
• pp.17-23
• /
• 2018

We proposed a LSTM-based RNN model that can effectively perform the automatic spacing characteristics. For those long or noisy sentences which are known to be difficult to handle within Neural Network Learning, we defined a proper input data format and decoding data format, and added dropout, bidirectional multi-layer LSTM, layer normalization, and attention mechanism to improve the performance. Despite of the fact that Sejong corpus contains some spacing errors, a noise-robust learning model developed in this study with no overfitting through a dropout method helped training and returned meaningful results of Korean word spacing and its patterns. The experimental results showed that the performance of LSTM sequence-to-sequence model is 0.94 in F1-measure, which is better than the rule-based deep-learning method of GRU-CRF.

• Smart Media Journal
• /
• v.10 no.2
• /
• pp.16-21
• /
• 2021

Financial time series analysis plays a very important role economically and socially in modern society and is an important task affecting global development, but due to difficulties such as a lot of noise and uncertainty, financial time series analysis prediction is a difficult research topic. In this paper, we propose a market prediction method (MPIL) by converting unstructured data and structured data into images. For market prediction, it analyzes SNS and news data, which is unstructured data for n days, and converts the market data, which is structured data, to an image with the GADF algorithm, and predicts an ultra-short market that predicts the price of n+1 days through image learning. MPIL has an average accuracy of 56%, which is higher than the 50% average accuracy of the model that predicts the market with LSTM by using sentiment analysis used for existing market forecasting.

• Smart Media Journal
• /
• v.12 no.4
• /
• pp.27-40
• /
• 2023

Many researchers have investigated ways to enhance recommender systems by integrating heterogeneous data to address the data sparsity problem. However, only a few studies have successfully integrated heterogeneous data using knowledge graph. Additionally, most of the knowledge graphs built in these studies only incorporate explicit relationships between entities and lack additional information. Therefore, we propose a method for expanding knowledge graphs by using deep learning to model latent relationships between heterogeneous data from multiple knowledge bases. Our extended knowledge graph enhances the quality of entity features and ultimately increases the accuracy of predicted user preferences. Experiments using real music data demonstrate that the expanded knowledge graph leads to an increase in recommendation accuracy when compared to the original knowledge graph.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.43 no.1
• /
• pp.99-107
• /
• 2023

Recently, the rate of death and safety accidents at construction sites is the highest among all kinds of industries. In order to apply artificial intelligence technology to construction sites, it is essential to secure a dataset which can be used as a basic training data. In this paper, a number of image data were collected through actual construction site, for which major construction equipment objects mainly operated in civil engineering sites were defined. The optimal training dataset construction was completed by annotation process of about 90,000 image dataset. Reliability of the dataset was verified with the mAP of over 90 % in use of YOLO, a representative model in the field of object detection. The construction equipment training dataset built in this study has been released which is currently available on the public data portal of the Ministry of Public Administration and Security. This dataset is expected to be freely used for any application of object detection technology on construction sites especially in the field of construction safety in the future.

• Smart Media Journal
• /
• v.12 no.5
• /
• pp.28-35
• /
• 2023

In this study, we propose a new light-weight model RoutingConvNet with fewer parameters to improve the applicability and practicality of speech emotion recognition. To reduce the number of learnable parameters, the proposed model connects bidirectional MFCCs on a channel-by-channel basis to learn long-term emotion dependence and extract contextual features. A light-weight deep CNN is constructed for low-level feature extraction, and self-attention is used to obtain information about channel and spatial signals in speech signals. In addition, we apply dynamic routing to improve the accuracy and construct a model that is robust to feature variations. The proposed model shows parameter reduction and accuracy improvement in the overall experiments of speech emotion datasets (EMO-DB, RAVDESS, and IEMOCAP), achieving 87.86%, 83.44%, and 66.06% accuracy respectively with about 156,000 parameters. In this study, we proposed a metric to calculate the trade-off between the number of parameters and accuracy for performance evaluation against light-weight.

• Smart Media Journal
• /
• v.12 no.5
• /
• pp.17-27
• /
• 2023

Recently, automatic text summarization, which automatically summarizes only meaningful information for users, is being studied steadily. Especially, research on text summarization using Transformer, an artificial neural network model, has been mainly conducted. Among various studies, the GSG method, which trains a model through sentence-by-sentence masking, has received the most attention. However, the traditional GSG has limitations in selecting a sentence to be masked based on the degree of overlap of tokens, not the meaning of a sentence. Therefore, in this study, in order to improve the quality of text summarization, we propose SbGSG (Semantic-based GSG) methodology that selects sentences to be masked by GSG considering the meaning of sentences. As a result of conducting an experiment using 370,000 news articles and 21,600 summaries and reports, it was confirmed that the proposed methodology, SbGSG, showed superior performance compared to the traditional GSG in terms of ROUGE and BERT Score.

• Smart Media Journal
• /
• v.13 no.1
• /
• pp.9-17
• /
• 2024

This study aims to develop an integrated agricultural distribution network management system to improve the quality, profit, and decision-making efficiency of agricultural products. We adopt two key techniques: crop maturity detection based on the YOLOX target detection algorithm and market price prediction based on the Prophet model. By training the target detection model, it was possible to accurately identify crops of various maturity stages, thereby optimizing the shipment timing. At the same time, by collecting historical market price data and predicting prices using the Prophet model, we provided reliable price trend information to shipping decision makers. According to the results of the study, it was found that the performance of the model considering the holiday factor was significantly superior to that of the model that did not, proving that the effect of the holiday on the price was strong. The system provides strong tools and decision support to farmers and agricultural distribution managers, helping them make smart decisions during various seasons and holidays. In addition, it is possible to optimize the distribution network of agricultural products and improve the quality and profit of agricultural products.

• Smart Media Journal
• /
• v.13 no.4
• /
• pp.23-32
• /
• 2024

Large-scale fires and their consequential damages are becoming increasingly common, but confidence in fire detection systems is waning. Recently, widely-used chemical fire detectors frequently generate lots of false alarms, while video-based deep learning fire detection is hampered by its time-consuming and expensive nature. To tackle these issues, this study proposes a fire detection model utilizing an autoencoder approach. The objective is to minimize false alarms while achieving swift and precise fire detection. The proposed model, employing an autoencoder methodology, can exclusively learn from normal data without the need for fire-related data, thus enhancing its adaptability to diverse environments. By amalgamating data from five distinct sensors, it facilitates rapid and accurate fire detection. Through experiments with various hyperparameter combinations, the proposed model demonstrated that out of 14 scenarios, only one encountered false alarm issues. Experimental results underscore its potential to curtail fire-related losses and bolster the reliability of fire detection systems.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.10a
• /
• pp.686-689
• /
• 2018

3D 손 포즈 추정(Hand Pose Estimation, HPE)은 스마트 인간 컴퓨터 인터페이스를 위해서 중요한 기술이다. 이 연구에서는 딥러닝 방법을 기반으로 하여 단일 RGB-Depth 카메라로 촬영한 양손의 3D 손 자세를 실시간으로 인식하는 손 포즈 추정 시스템을 제시한다. 손 포즈 추정 시스템은 4단계로 구성된다. 첫째, Skin Detection 및 Depth cutting 알고리즘을 사용하여 양손을 RGB와 깊이 영상에서 감지하고 추출한다. 둘째, Convolutional Neural Network(CNN) Classifier는 오른손과 왼손을 구별하는데 사용된다. CNN Classifier 는 3개의 convolution layer와 2개의 Fully-Connected Layer로 구성되어 있으며, 추출된 깊이 영상을 입력으로 사용한다. 셋째, 학습된 CNN regressor는 추출된 왼쪽 및 오른쪽 손의 깊이 영상에서 손 관절을 추정하기 위해 다수의 Convolutional Layers, Pooling Layers, Fully Connected Layers로 구성된다. CNN classifier와 regressor는 22,000개 깊이 영상 데이터셋으로 학습된다. 마지막으로, 각 손의 3D 손 자세는 추정된 손 관절 정보로부터 재구성된다. 테스트 결과, CNN classifier는 오른쪽 손과 왼쪽 손을 96.9%의 정확도로 구별할 수 있으며, CNN regressor는 형균 8.48mm의 오차 범위로 3D 손 관절 정보를 추정할 수 있다. 본 연구에서 제안하는 손 포즈 추정 시스템은 가상 현실(virtual reality, VR), 증강 현실(Augmented Reality, AR) 및 융합 현실 (Mixed Reality, MR) 응용 프로그램을 포함한 다양한 응용 분야에서 사용할 수 있다.