• Smart Media Journal
• /
• v.11 no.4
• /
• pp.46-55
• /
• 2022

It is very important to first detect and remove defective fruits with scratches or bruised areas in the automatic fruit quality screening system. This paper proposes a method of detecting defective areas in fruits using the latest artificial intelligence technology, the Yolo V4 deep learning model in order to overcome the limitations of the method of detecting fruit's defective areas using the existing image processing techniques. In this study, a total of 2,400 defective fruits, including 1,000 defective apples and 1,400 defective fruits with scratch or decayed areas, were learned using the Yolo V4 deep learning model and experiments were conducted to detect defective areas. As a result of the performance test, the precision of apples is 0.80, recall is 0.76, IoU is 69.92% and mAP is 65.27%. The precision of pears is 0.86, recall is 0.81, IoU is 70.54% and mAP is 68.75%. The method proposed in this study can dramatically improve the performance of the existing automatic fruit quality screening system by accurately selecting fruits with defective areas in real time rather than using the existing image processing techniques.

• Journal of Information Processing Systems
• /
• v.16 no.5
• /
• pp.1034-1047
• /
• 2020

The personal authentication technique is an essential tool in this complex and modern digital information society. Traditionally, the most general mechanism of personal authentication was using alphanumeric passwords. However, passwords that are hard to guess or to break, are often hard to remember. There are demands for a technology capable of replacing the text-based password system. Graphical passwords can be an alternative, but it is vulnerable to shoulder-surfing attacks. This paper looks through a number of recently developed graphical password systems and introduces a personal authentication system using a machine learning technique with electroencephalography (EEG) signals as a new type of personal authentication system which is easier for a person to use and more difficult for others to steal than other preexisting authentication systems.

• Journal of Advanced Navigation Technology
• /
• v.24 no.2
• /
• pp.155-162
• /
• 2020

Recently, with the development of computer vision and deep learning technology, research on human action recognition has been actively conducted for video analysis, video surveillance, interactive multimedia, and human machine interaction applications. Diverse techniques have been introduced for human action understanding and classification by many researchers using RGB image, depth image, skeleton and inertial data. However, skeleton-based action discrimination is still a challenging research topic for human machine-interaction. In this paper, we propose an end-to-end skeleton joints mapping of action for generating spatio-temporal image so-called dynamic image. Then, an efficient deep convolution neural network is devised to perform the classification among the action classes. We use publicly accessible UTD-MHAD skeleton dataset for evaluating the performance of the proposed method. As a result of the experiment, the proposed system shows better performance than the existing methods with high accuracy of 97.45%.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.6
• /
• pp.1209-1216
• /
• 2020

In this study, we propose a method for detecting coastal surface wastes using an UAV(Unmanned Aerial Vehicle) remote sensing method and an object detection algorithm based on deep learning. An object detection algorithm based on deep neural networks was proposed to detect coastal debris in aerial images. A deep neural network model was trained with image datasets of three classes: PET, Styrofoam, and plastics. And the detection accuracy of each class was compared with Darknet-53. Through this, it was possible to monitor the wastes landing on the shore by type through unmanned aerial vehicles. In the future, if the method proposed in this study is applied, a complete enumeration of the whole beach will be possible. It is believed that it can contribute to increase the efficiency of the marine environment monitoring field.

• International Journal of Computer Science & Network Security
• /
• v.24 no.10
• /
• pp.1-16
• /
• 2024

Due to its complexity and high diagnosis and treatment costs, heart attack (HA) is the top cause of death globally. Heart failure's widespread effect and high morbidity and death rates make accurate and fast prognosis and diagnosis crucial. Due to the complexity of medical data, early and accurate prediction of HA is difficult. Healthcare providers must evaluate data quickly and accurately to intervene. This novel hybrid approach predicts HA using Long Short-Term Memory (LSTM) networks, Deep belief networks (DBNs) with attention mechanism, and robust data mining to fill this essential gap. HA is predicted using Kaggle, PhysioNet, and UCI datasets. Wearable sensor data, ECG signals, and demographic and clinical data provide a solid analytical base. To maintain consistency, ECG signals are normalized and segmented after thorough cleaning to remove missing values and noise. Feature extraction employs complex approaches like Principal Component Analysis (PCA) and Autoencoders to pick time-domain (MNN, SDNN, RMSSD, PNN50) and frequency-domain (PSD at VLF, LF, HF bands) characteristics. The hybrid model architecture uses LSTM networks for sequence learning and DBNs for feature representation and selection to create a robust and comprehensive prediction model. Accuracy, precision, recall, F1-score, and ROC-AUC are measured after cross-entropy loss and SGD optimization. The LSTM-DBN model outperforms predictive methods in accuracy, sensitivity, and specificity. The findings show that several data sources and powerful algorithms can improve heart attack predictions. The proposed architecture performed well on many datasets, with an accuracy rate of 96.00%, sensitivity of 98%, AUC of 0.98, and F1-score of 0.97. High performance proves this system's dependability. Moreover, the proposed approach is outperformed compared to state-of-the-art systems.

• Journal of Korea Multimedia Society
• /
• v.23 no.7
• /
• pp.850-861
• /
• 2020

Hangul is a language that is composed of initial, medial, and final syllables. It has 11,172 characters. For this reason, the current method of designing all the characters by hand is very expensive and time-consuming. In order to solve the problem, this paper proposes an automatic Hangul font generation system and evaluates the standards for mapping Hangul characters to produce an effective automated Hangul font generation system. The system was implemented using character generation engine based on deep learning CycleGAN. In order to evaluate the criteria when mapping characters in pairs, each criterion was designed based on Hangul structure and character shape, and the quality of the generated characters was evaluated. As a result of the evaluation, the standards designed based on the Hangul structure did not affect the quality of the automated Hangul font generation system. On the other hand, when tried with similar characters, the standards made based on the shape of Hangul characters produced better quality characters than when tried with less similar characters. As a result, it is better to generate automated Hangul font by designing a learning method based on mapping characters in pairs that have similar character shapes.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.360-363
• /
• 2022

According to the 2018 Kaggle ML & DS Survey, among the proportions of frameworks for machine learning and data science, TensorFlow and Keras each account for 41.82%. It was found to be 34.09%, and in the case of development programming, it is confirmed that about 82% use Python. A significant number of machine learning and deep learning structures utilize the Keras framework and Python, but in the case of Python, distribution and execution are limited to the Python script environment due to the script language, so it is judged that it is difficult to operate in various environments. This paper implemented a machine learning and deep learning system using C# and Keras running in Visual Studio 2019. Using the Mnist dataset, 100 tests were performed in Python 3.8,2 and C# .NET 5.0 environments, and the minimum time for Python was 1.86 seconds, the maximum time was 2.38 seconds, and the average time was 1.98 seconds. Time 1.78 seconds, maximum time 2.11 seconds, average time 1.85 seconds, total time 37.02 seconds. As a result of the experiment, the performance of C# improved by about 6% compared to Python, and it is expected that the utilization will be high because executable files can be extracted.

• Journal of Multimedia Information System
• /
• v.7 no.3
• /
• pp.205-214
• /
• 2020

Crime has become one of the major problems in modern society. Even though visual surveillances through closed-circuit television (CCTV) is extensively used for solving crime, the number of crimes has not decreased. This is because there is insufficient workforce for performing 24-hour surveillance. In addition, CCTV surveillance by humans is not efficient for detecting dangerous situations owing to accuracy issues. In this paper, we propose the autonomous detection of dangerous situations in CCTV scenes using a deep learning model with relational inference. The main feature of the proposed method is that it can simultaneously perform object detection and relational inference to determine the danger of the situations captured by CCTV. This enables us to efficiently classify dangerous situations by inferring the relationship between detected objects (i.e., distance and position). Experimental results demonstrate that the proposed method outperforms existing methods in terms of the accuracy of image classification and the false alarm rate even when object detection accuracy is low.

• Journal of Multimedia Information System
• /
• v.6 no.4
• /
• pp.235-238
• /
• 2019

In real world object detection is an active research topic for understanding different objects from images. There are different models presented in past and had significant results. In this paper we are presenting vehicle logo detection using previous object detection models such as You only look once (YOLO) and Faster Region-based CNN (F-RCNN). Both the front and rear view of the vehicles were used for training and testing the proposed method. Along with deep learning an image pre-processing algorithm called perspective transformation is proposed for all the test images. Using perspective transformation, the top view images were transformed into front view images. This algorithm has higher detection rate as compared to raw images. Furthermore, YOLO model has better result as compare to F-RCNN model.

• Journal of Multimedia Information System
• /
• v.4 no.4
• /
• pp.225-232
• /
• 2017

Understanding actions in videos is an important task. It helps in finding the anomalies present in videos such as fights. Detection of fights becomes more crucial when it comes to sports. This paper focuses on finding fight scenes in Hockey sport videos using blur & radon transform and convolutional neural networks (CNNs). First, the local motion within the video frames has been extracted using blur information. Next, fast fourier and radon transform have been applied on the local motion. The video frames with fight scene have been identified using transfer learning with the help of pre-trained deep learning model VGG-Net. Finally, a comparison of the methodology has been performed using feed forward neural networks. Accuracies of 56.00% and 75.00% have been achieved using feed forward neural network and VGG16-Net, respectively.