• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.11
• /
• pp.411-418
• /
• 2022

Incorrectly loaded containers can easily knock down by strong winds. Container collapse accidents can lead to material damage and paralysis of the port system. In this paper, We propose a deep learning-based container loading state and accident risk detection technique. Using Darknet-based YOLO, the container load status identifies in real-time through corner casting on the top and bottom of the container, and the risk of accidents notifies the manager. We present criteria for classifying container alignment states and select efficient learning algorithms based on inference speed, classification accuracy, detection accuracy, and FPS in real embedded devices in the same environment. The study found that YOLOv4 had a weaker inference speed and performance of FPS than YOLOv3, but showed strong performance in classification accuracy and detection accuracy.

• International Journal of Computer Science & Network Security
• /
• v.23 no.9
• /
• pp.1-7
• /
• 2023

Demand prediction is an essential component of any business or supply chain. Large retailers need to keep track of tens of millions of items flows each day to ensure smooth operations and strong margins. The demand prediction is in the epicenter of this planning tornado. For business processes in retail companies that deal with a variety of products with short shelf life and foodstuffs, forecast accuracy is of the utmost importance due to the shifting demand pattern, which is impacted by an environment of dynamic and fast response. All sectors strive to produce the ideal quantity of goods at the ideal time, but for retailers, this issue is especially crucial as they also need to effectively manage perishable inventories. In light of this, this research aims to show how Machine Learning approaches can help with demand forecasting in retail and future sales predictions. This will be done in two steps. One by using historic data and another by using open data of weather conditions, fuel, Consumer Price Index (CPI), holidays, any specific events in that area etc. Several machine learning algorithms were applied and compared using the r-squared and mean absolute percentage error (MAPE) assessment metrics. The suggested method improves the effectiveness and quality of feature selection while using a small number of well-chosen features to increase demand prediction accuracy. The model is tested with a one-year weekly dataset after being trained with a two-year weekly dataset. The results show that the suggested expanded feature selection approach provides a very good MAPE range, a very respectable and encouraging value for anticipating retail demand in retail systems.

• International Journal of Computer Science & Network Security
• /
• v.23 no.8
• /
• pp.210-216
• /
• 2023

Demand prediction is an essential component of any business or supply chain. Large retailers need to keep track of tens of millions of items flows each day to ensure smooth operations and strong margins. The demand prediction is in the epicenter of this planning tornado. For business processes in retail companies that deal with a variety of products with short shelf life and foodstuffs, forecast accuracy is of the utmost importance due to the shifting demand pattern, which is impacted by an environment of dynamic and fast response. All sectors strive to produce the ideal quantity of goods at the ideal time, but for retailers, this issue is especially crucial as they also need to effectively manage perishable inventories. In light of this, this research aims to show how Machine Learning approaches can help with demand forecasting in retail and future sales predictions. This will be done in two steps. One by using historic data and another by using open data of weather conditions, fuel, Consumer Price Index (CPI), holidays, any specific events in that area etc. Several machine learning algorithms were applied and compared using the r-squared and mean absolute percentage error (MAPE) assessment metrics. The suggested method improves the effectiveness and quality of feature selection while using a small number of well-chosen features to increase demand prediction accuracy. The model is tested with a one-year weekly dataset after being trained with a two-year weekly dataset. The results show that the suggested expanded feature selection approach provides a very good MAPE range, a very respectable and encouraging value for anticipating retail demand in retail systems.

• Human Ecology Research
• /
• v.61 no.1
• /
• pp.15-28
• /
• 2023

The aim of this study was to develop, implement, and evaluate the use of Artificial Intelligence in the prevention of violence among middle-school students. The sample for this study consisted of 20 first-year middle-school students who participated in theme selection activities in a free semester program as part of their home economics studies. The data for the study consisted of nine class observation logs, four group activity outputs, 30 class results, an online survey, and in-depth interviews with three students. A program called "R.U.OK" was developed by setting problematic situation for school violence prevention linked to the contents of the Home Economics Education(HEE) curriculum. After the program was implemented, the survey on the students' class satisfaction content elements, with AI-based learning activities and PBL and interest, displayed high points, with an average of 4.0 or higher. Our qualitative analysis produced four significant results. First, students' concerns about school violence had increased and they showed a change in attitude, having more empathy with friends and more interest in their surroundings. Second, digital and AI literacy had improved, and students' interest in digital media learning had increased. Third, there had been an improvement in problem-solving ability in terms of being able to think more critically and independently. Fourth, the results also demonstrated that there had been a positive effect on self-direction and an improved capacity for teamwork. This study was significant in demonstrating the effectiveness of a program for the prevention of school violence based on the use of digital technology in the educational environment.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.48 no.5
• /
• pp.55-63
• /
• 2011

This paper describes new moving target detection technique combining two algorithms to detect targets and reject clutters in video frame images for surveillance system: One obtains the region of moving target using phase correlation method using $N{\times}M$ sub-block images in frequency domain. The other uses adaptive threshold using learning weight for extracting target candidates in subtracted image. The block region with moving target can be obtained using the characteristics that the highest value of phase correlation depends on the movement of largest image in block. This technique can be used in camera motion environment calculating and compensating camera movement using FFT phase correlation between input video frame images. The experimental results show that the proposed algorithm accurately detects target(s) with a low false alarm rate in variety environment using the receiver operating characteristics (ROC) curve.

• Proceedings of the KIEE Conference
• /
• 2002.11c
• /
• pp.163-168
• /
• 2002

The voting algorithm for action selection performs self-improvement by Reinforcement learning algorithm in the dynamic environment. The proposed voting algorithm improves the navigation of the robot by adapting the eligibility of the behaviors and determining the Command Set Generator (CGS). The Navigator that using a proposed voting algorithm corresponds to the CGS for giving the weight values and taking the reward values. It is necessary to decide which Command Set control the mobile robot at given time and to select among the candidate actions. The Command Set was learnt online by means as Q-learning. Action Selector compares Q-values of Navigator with Heterogeneous behaviors. Finally, real-world experimentation was carried out. Results show the good performance for the selection on command set as well as the convergence of Q-value.

• International Journal of Computer Science & Network Security
• /
• v.21 no.8
• /
• pp.127-136
• /
• 2021

Real-time human recognition is a challenging task, as the images are captured in an unconstrained environment with different poses, makeups, and styles. This limitation is addressed by generating several facial images with poses, makeup, and styles with a single reference image of a person using Generative Adversarial Networks (GAN). In this paper, we propose deep learning-based human recognition using integration of GAN and Spatial Domain Techniques. A novel concept of human recognition based on face depiction approach by generating several dissimilar face images from single reference face image using Domain Transfer Generative Adversarial Networks (DT-GAN) combined with feature extraction techniques such as Local Binary Pattern (LBP) and Histogram is deliberated. The Euclidean Distance (ED) is used in the matching section for comparison of features to test the performance of the method. A database of millions of people with a single reference face image per person, instead of multiple reference face images, is created and saved on the centralized server, which helps to reduce memory load on the centralized server. It is noticed that the recognition accuracy is 100% for smaller size datasets and a little less accuracy for larger size datasets and also, results are compared with present methods to show the superiority of proposed method.

• Journal of Korea Multimedia Society
• /
• v.24 no.8
• /
• pp.1026-1034
• /
• 2021

Tire wear and defect are important factors for safe driving condition. These defects are generally inspected by some specialized experts or very expensive equipments such as stereo depth camera and depth gauge. In this paper, we propose tire safety vision inspector based on deep neural network (DNN). The status of tire wear is categorized into three: 'safety', 'warning', and 'danger' based on depth of tire tread. We propose an attention mechanism for emphasizing the feature of tread area. The attention-based feature is concatenated to output feature maps of the last convolution layer of ResNet-101 to extract more robust feature. Through experiments, the proposed tire wear classification model improves 1.8% of accuracy compared to the existing ResNet-101 model. For detecting the tire defections, the developed tire defect detection model shows up-to 91% of accuracy using the Mask R-CNN model. From these results, we can see that the suggested models are useful for checking on the safety condition of working tire in real environment.

• International conference on construction engineering and project management
• /
• 2017.10a
• /
• pp.80-88
• /
• 2017

Although hazard identification is one of the most important steps of safety management process, numerous hazards remain unidentified in the construction workplace due to the dynamic environment of the construction site and the lack of available resource for visual inspection. To this end, our previous study proposed the collective sensing approach for safety hazard identification and showed the feasibility of identifying hazards by capturing collective abnormalities in workers' walking patterns. However, workers generally performed different activities during the construction task in the workplace. Thereby, an additional process that can identify the worker's walking activity is necessary to utilize the proposed hazard identification approach in real world settings. In this context, this study investigated the feasibility of identifying walking activities during construction task using Wearable Inertial Measurement Units (WIMU) attached to the worker's ankle. This study simulated the indoor masonry work for data collection and investigated the classification performance with three different machine learning algorithms (i.e., Decision Tree, Neural Network, and Support Vector Machine). The analysis results showed the feasibility of identifying worker's activities including walking activity using an ankle-attached WIMU. Moreover, the finding of this study will help to enhance the performance of activity recognition and hazard identification in construction.

• Journal of Bio-Environment Control
• /
• v.28 no.2
• /
• pp.95-103
• /
• 2019

Temperature and relative humidity are important factors in crop cultivation and should be properly controlled for improving crop yield and quality. In order to control the environment accurately, we need to predict how the environment will change in the future. The objective of this study was to predict air temperature and relative humidity at a future time by using a multilayer perceptron (MLP). The data required to train MLP was collected every 10 min from Oct. 1, 2016 to Feb. 28, 2018 in an eight-span greenhouse ($1,032m^2$) cultivating mango (Mangifera indica cv. Irwin). The inputs for the MLP were greenhouse inside and outside environment data, and set-up and operating values of environment control devices. By using these data, the MLP was trained to predict the air temperature and relative humidity at a future time of 10 to 120 min. Considering typical four seasons in Korea, three-day data of the each season were compared as test data. The MLP was optimized with four hidden layers and 128 nodes for air temperature ($R^2=0.988$) and with four hidden layers and 64 nodes for relative humidity ($R^2=0.990$). Due to the characteristics of MLP, the accuracy decreased as the prediction time became longer. However, air temperature and relative humidity were properly predicted regardless of the environmental changes varied from season to season. For specific data such as spray irrigation, however, the numbers of trained data were too small, resulting in poor predictive accuracy. In this study, air temperature and relative humidity were appropriately predicted through optimization of MLP, but were limited to the experimental greenhouse. Therefore, it is necessary to collect more data from greenhouses at various places and modify the structure of neural network for generalization.