• Journal of Sensor Science and Technology
• /
• v.33 no.5
• /
• pp.298-304
• /
• 2024

In Korea, fires are the second most common type of disaster, causing large-scale damages. The installation of fire detectors is legislated to prevent fires and minimize damage. Conventional fire detectors have limitations in initial suppression of failures because they detect fires when large amounts of smoke and heat are generated. Additionally, frequent malfunctions in fire detectors may cause users to turn them off. To address these issues, recent studies focus on accurately detecting even small-scale fires using multi-sensor and deep-learning technologies. They also aim at quick fire detection and thermal decomposition using gas. However, these studies are not practical because they overlook the heavy computations involved. Therefore, we propose a fast and accurate fire detection system based on multi-sensor and deep-learning technologies. In addition, we propose a computation-reduction method for selecting sensors suitable for detection using the Pearson correlation coefficient. Specifically, we use a moving average to handle outliers and two-stage labeling to reduce false detections during preprocessing. Subsequently, a deep-learning model is selected as LSTM for analyzing the temporal sequence. Then, we analyze the data using a correlation analysis. Consequently, the model using a small data group with low correlation achieves an accuracy of 99.88% and a false detection rate of 0.12%.

• International Journal of Advanced Culture Technology
• /
• v.5 no.2
• /
• pp.74-81
• /
• 2017

Recently, there are many companies that use systems based on artificial intelligence. The accuracy of artificial intelligence depends on the amount of learning data and the appropriate algorithm. However, it is not easy to obtain learning data with a large number of entity. Less data set have large generalization errors due to overfitting. In order to minimize this generalization error, this study proposed DGA(Dropout Genetic Algorithm) which can expect relatively high accuracy even though data with a less data set is applied to machine learning based genetic algorithm to deep learning based dropout. The idea of this paper is to determine the active state of the nodes. Using Gradient about loss function, A new fitness function is defined. Proposed Algorithm DGA is supplementing stochastic inconsistency about Dropout. Also DGA solved problem by the complexity of the fitness function and expression range of the model about Genetic Algorithm As a result of experiments using MNIST data proposed algorithm accuracy is 75.3%. Using only Dropout algorithm accuracy is 41.4%. It is shown that DGA is better than using only dropout.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.5
• /
• pp.171-177
• /
• 2018

Recently there are many image datasets which has variety of data class and point to extract general features. But in order to this variety data class and point, deep learning model trained this dataset has not good performance in heterogeneous data feature local area. In this paper, we propose the structure which use sub-category and openset object detection methods to train more robust model, named multi-branch tree using ASSL. By using this structure, we can have more robust object detection deep learning model in heterogeneous data feature environment.

• The Journal of Korea Robotics Society
• /
• v.14 no.4
• /
• pp.357-362
• /
• 2019

This paper proposes a model and train method that can real-time detect objects and distances estimation based on a monocular camera by applying deep learning. It used YOLOv2 model which is applied to autonomous or robot due to the fast image processing speed. We have changed and learned the loss function so that the YOLOv2 model can detect objects and distances at the same time. The YOLOv2 loss function added a term for learning bounding box values x, y, w, h, and distance values z as 클래스ification losses. In addition, the learning was carried out by multiplying the distance term with parameters for the balance of learning. we trained the model location, recognition by camera and distance data measured by lidar so that we enable the model to estimate distance and objects from a monocular camera, even when the vehicle is going up or down hill. To evaluate the performance of object detection and distance estimation, MAP (Mean Average Precision) and Adjust R square were used and performance was compared with previous research papers. In addition, we compared the original YOLOv2 model FPS (Frame Per Second) for speed measurement with FPS of our model.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.3
• /
• pp.85-90
• /
• 2021

In this paper, fall detection using individual wearable devices for older people is considered. To design a low-cost wearable device for reliable fall detection, we present a comprehensive analysis of two representative models. One is a machine learning model composed of a decision tree, random forest, and Support Vector Machine(SVM). The other is a deep learning model relying on a one-dimensional(1D) Convolutional Neural Network(CNN). By considering data segmentation, preprocessing, and feature extraction methods applied to the input data, we also evaluate the considered models' validity. Simulation results verify the efficacy of the deep learning model showing improved overall performance.

• Journal of Korea Multimedia Society
• /
• v.24 no.8
• /
• pp.1000-1011
• /
• 2021

In recent years, using Deep Learning methods to apply for medical and biomedical image analysis has seen many advancements. In clinical, using Deep Learning-based approaches for cancer image analysis is one of the key applications for cancer detection and treatment. However, the scarcity and shortage of labeling images make the task of cancer detection and analysis difficult to reach high accuracy. In 2015, the Unet model was introduced and gained much attention from researchers in the field. The success of Unet model is the ability to produce high accuracy with very few input images. Since the development of Unet, there are many variants and modifications of Unet related architecture. This paper proposes a new approach of using Unet++ with pretrained EfficientNet as backbone architecture for breast tumor cell nuclei segmentation and uses the multi-organ transfer learning approach to segment nuclei of breast tumor cells. We attempt to experiment and evaluate the performance of the network on the MonuSeg training dataset and Triple Negative Breast Cancer (TNBC) testing dataset, both are Hematoxylin and Eosin (H & E)-stained images. The results have shown that EfficientUnet++ architecture and the multi-organ transfer learning approach had outperformed other techniques and produced notable accuracy for breast tumor cell nuclei segmentation.

• Transactions of Materials Processing
• /
• v.31 no.4
• /
• pp.229-239
• /
• 2022

Convolution Neural Network(CNN) is a class of deep learning algorithms and can be used for image analysis. In particular, it has excellent performance in finding the pattern of images. Therefore, CNN is commonly applied for recognizing, learning and classifying images. In this study, the surface defect classification performance of Al 6061 extruded material using CNN-based algorithms were compared and evaluated. First, the data collection criteria were suggested and a total of 2,024 datasets were prepared. And they were randomly classified into 1,417 learning data and 607 evaluation data. After that, the size and quality of the training data set were improved using data augmentation techniques to increase the performance of deep learning. The CNN-based algorithms used in this study were VGGNet-16, VGGNet-19, ResNet-50 and DenseNet-121. The evaluation of the defect classification performance was made by comparing the accuracy, loss, and learning speed using verification data. The DenseNet-121 algorithm showed better performance than other algorithms with an accuracy of 99.13% and a loss value of 0.037. This was due to the structural characteristics of the DenseNet model, and the information loss was reduced by acquiring information from all previous layers for image identification in this algorithm. Based on the above results, the possibility of machine vision application of CNN-based model for the surface defect classification of Al extruded materials was also discussed.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.5
• /
• pp.171-176
• /
• 2022

With the recent development of deep learning technology, the areas of recommendation systems have also diversified. This paper studied algorithms to improve the learning rate and studied the significance results according to words through comparison with the performance characteristics of the Word2Vec model. The problem recommendation algorithm was implemented with the values expressed through the reflection of meaning and similarity test between texts, which are characteristics of the Word2Vec model. Through Word2Vec's learning results, problem recommendations were conducted using text similarity values, and problems with high similarity can be recommended. In the experimental process, it was seen that the accuracy decreased with the quantitative amount of data, and it was confirmed that the larger the amount of data in the data set, the higher the accuracy.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.5
• /
• pp.63-68
• /
• 2022

Autonomous driving technology has recently made great progress with the introduction of deep reinforcement learning. In order to effectively use deep reinforcement learning, it is important to select the appropriate activation function. In the meantime, many activation functions have been presented, but they show different performance depending on the environment to be applied. This paper compares and evaluates the performance of 12 activation functions to see which activation functions are effective when using reinforcement learning to learn autonomous driving on highways. To this end, a performance evaluation method was presented and the average reward value of each activation function was compared. As a result, when using GELU, the highest average reward could be obtained, and SiLU showed the lowest performance. The average reward difference between the two activation functions was 20%.

• Fourth Industrial Review
• /
• v.1 no.1
• /
• pp.31-40
• /
• 2021

Purpose - According to the importance of business simulation learning as a new type of business learning tool, this study reviews the dimensions of business education and a brief history of business education simulation. At the end Capsim strategic management simulation program is introduce with its feature. Research design, data, and methodology - This study has been analyzed in a way that reviews at previous literature on simulation learning and looks at examples and features of Capsim simulation, online business simulation tools which has been used in the global market. Result - Capsim simulations are designed to offer focused opportunities for deep practice. That's why they are often more effective than passive tools such as textbooks, videos, or lectures. By the way, 'deep practice' is very different from 'ordinary practice'. After commuters who drive to school or work can accumulate thousands of hours of driving, but that doesn't make them expert drivers. The key to deep practice is self-awareness. That is, paying attention to what you are doing well and not so well. This is so important to learn that scientists use a specific term for it: 'metacognition', or thinking about the way you think and learn. Conclusion - The use of business simulation learning, such as Capsim, which is a given case, can create similar local systems by potentially engaging a large number of users in the virtual market. It could also be used as an individual to complete business training for students and those who are active in the business field of business.