• Smart Media Journal
• /
• v.11 no.8
• /
• pp.84-92
• /
• 2022

Precision livestock monitoring promises greater management efficiency for farmers and higher welfare standards for animals. Recent studies on video-based animal activity recognition and tracking have shown promising solutions for understanding animal behavior. To achieve that, surveillance cameras are installed diagonally above the barn in a typical cattle farm setup to monitor animals constantly. Under these circumstances, tracking individuals requires addressing challenges such as occlusion and visual appearance, which are the main reasons for track breakage and increased misidentification of animals. This paper presents a framework for multi-cattle tracking in closed barns with appearance and motion models. To overcome the above challenges, we modify the DeepSORT algorithm to achieve higher tracking accuracy by three contributions. First, we reduce the weight of appearance information. Second, we use an Ensemble Kalman Filter to predict the random motion information of cattle. Third, we propose a supplementary matching algorithm that compares the absolute cattle position in the barn to reassign lost tracks. The main idea of the matching algorithm assumes that the number of cattle is fixed in the barn, so the edge of the barn is where new trajectories are most likely to emerge. Experimental results are performed on our dataset collected on two cattle farms. Our algorithm achieves 70.37%, 77.39%, and 81.74% performance on HOTA, AssA, and IDF1, representing an improvement of 1.53%, 4.17%, and 0.96%, respectively, compared to the original method.

• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.1
• /
• pp.53-63
• /
• 2022

In the paper, we propose a TSCH-based scheduling scheme for IEEE 802.15.4e, which is able to perform the scheduling of its own network by avoiding collision from interference network cluster (INC). Firstly, we model a bipartite graph structure for presenting the slot-frame (channel-slot assignment) of TSCH. Then, based on the bipartite graph edge weight, we utilize the Hungarian assignment algorithm to implement a scheduling scheme. We have employed two features (maximization and minimization) of the Hungarian-based assignment algorithm, which can perform the assignment in terms of minimizing the throughput of INC and maximizing the throughput of own network. Further, in this work, we called the scheme "dual-stage Hungarian-based assignment algorithm". Furthermore, we also propose deep learning (DL) based deep neural network (DNN)scheme, where the data were generated by the dual-stage Hungarian-based assignment algorithm. The performance of the DNN scheme is evaluated by simulations. The simulation results prove that the proposed DNN scheme providessimilar performance to the dual-stage Hungarian-based assignment algorithm while providing a low execution time.

• Geomechanics and Engineering
• /
• v.34 no.1
• /
• pp.43-49
• /
• 2023

Knowledge of minimum horizontal stress (S_hmin) is a significant step in determining full stress tensor. It provides crucial information for the production of sand, hydraulic fracturing, determination of safe mud weight window, reservoir production behavior, and wellbore stability. Calculating the S_hmin using indirect methods has been proved to be awkward because a lot of data are required in all of these models. Also, direct techniques such as hydraulic fracturing are costly and time-consuming. To figure these problems out, this work aims to apply the long-short-term memory (LSTM) algorithm to S_hmin time-series prediction. 13956 datasets obtained from an oil well logging operation were applied in the models. 80% of the data were used for training, and 20% of the data were used for testing. In order to achieve the maximum accuracy of the LSTM model, its hyper-parameters were optimized significantly. Through different statistical indices, the LSTM model's performance was compared with with other machine learning methods. Finally, the optimized LSTM model was recommended for S_hmin prediction in the well logging operation.

• Journal of the Semiconductor & Display Technology
• /
• v.22 no.4
• /
• pp.179-184
• /
• 2023

We propose the modeling methodology of CMOS inverter made of LTPO TFT using a machine learning. LTPO can achieve advantages of LTPS TFT with high electron mobility as a driving TFT and IGZO TFT with low off-current as a switching TFT. However, since the unified model of both LTPS and IGZO TFTs is still lacking, it is necessary to develop a SPICE-compatible compact model to simulate the LTPO current-voltage characteristics. In this work, a generic framework for combining the existing formula of I-V characteristics with artificial neural network is presented. The weight and bias values of ANN for LTPS and IGZO TFTs is obtained and implemented into PSPICE circuit simulator to predict CMOS inverter. This methodology enables efficient modeling for predicting LTPO TFT circuit characteristics.

• Journal of Korea Artificial Intelligence Association
• /
• v.2 no.1
• /
• pp.1-6
• /
• 2024

In this paper, we develop an AI-based recommendation system that matches the specifications of smartphones from company 'S'. The system aims to simplify the complex decision-making process of consumers and guide them to choose the smartphone that best suits their daily needs. The recommendation system analyzes five specifications of smartphones (price, battery capacity, weight, camera quality, capacity) to help users make informed decisions without searching for extensive information. This approach not only saves time but also improves user satisfaction by ensuring that the selected smartphone closely matches the user's lifestyle and needs. The system utilizes unsupervised learning, i.e. clustering (K-MEANS, DBSCAN, Hierarchical Clustering), and provides personalized recommendations by evaluating them with silhouette scores, ensuring accurate and reliable grouping of similar smartphone models. By leveraging advanced data analysis techniques, the system can identify subtle patterns and preferences that might not be immediately apparent to consumers, enhancing the overall user experience. The ultimate goal of this AI recommendation system is to simplify the smartphone selection process, making it more accessible and user-friendly for all consumers. This paper discusses the data collection, preprocessing, development, implementation, and potential impact of the system using Pandas, crawling, scikit-learn, etc., and highlights the benefits of helping consumers explore the various options available and confidently choose the smartphone that best suits their daily lives.

• The Journal of Korean Physical Therapy
• /
• v.35 no.6
• /
• pp.213-219
• /
• 2023

Purpose: In this study, based on the error augmentation, we performed walking training with increased rhythmic auditory stimulation speed on the affected side (IRAS) and walking training with decreased rhythmic auditory stimulation speed on the unaffected side (DRAS). The purpose of this study was to verify whether motor learning was effective in improving balance ability. Methods: Twenty-eight subjects with chronic stroke were recruited from a rehabilitation center. The subjects were divided into three groups: an IRAS group (10 subjects), a DRAS group (9 subjects), and control group (9 subjects). They received 30minutes of neuro-developmental therapy and walking training for 30minutes, five times a week for three weeks. Static and functional balance ability were measured before and after the training period. Static balance was measured by balancia software. Functional balance was measured by the timed up and go test (TUG) and the berg balance scale (BBS). Results: After the training periods, the IRAS group showed a significant improvement in TUG, BBS, area 95% COP, and weight distribution on the affected side when compared to both the DRAS group and control group (p<0.05). Conclusion: Based on the results of this study, it is possible to consider error augmentation methods of motor learning if rhythmic auditory stimulation is applied to stroke patients in clinical practice. If the affected side is shorter than the unaffected side, the affected side should be adjusted to the increased rhythmic auditory stimulation speed, which is considered to be an effective intervention to improve balance ability.

• The Journal of Pediatrics of Korean Medicine
• /
• v.14 no.1
• /
• pp.9-38
• /
• 2000

It has been widely said in Korea that early administrations of Cervi cornu parvum (deer antler) to febrile infants affect brain functions. Traditional Oriental Medicine states that the head is easily affected by fever and only an excess of heat causes headaches. Traditional Oriental Medicine also states that Cervi cornu parvum cannot be used in febrile conditions. With the aim of investigating different febrile response to LPS, experiments using intravenous injection of LPS have been carried out on Cervi comu parvum(CCP) and Soahbohyul - tang combined with Cervi comu parvum(SB-CCP) administered rabbits. Experiments were also conducted to evaluate the effects of early administration of CCP on learning and memory in 3 week old rats with LPS fever. These were evaluated by using the Morris water maze and the radial arm maze. Changes in body weight were also observed during this period. The results of these experiments are as follows. 1. In the experiments with febrile rabbits, the CCP and SB-CCP administered group showed statistically significant reductions of fever (p<0.05). 2. In the experiments with febrile rabbits, CCP and SB-CCP administered rabbits resulted in the tendency of lower body temperatures and shorter fever periods than the control group. 3. There were no differences of mean body weight and fever patterns among the 4 groups in the experiments on young rats with LPS fever. 4. There was no statistical difference of mean response latencies among the rats in Group I (DDW administered), GroupIII (CCP administered), and groupIV (SB-CCP administered) in the Morris water maze. However, Group Ⅱ (the scopolamine administered group) showed delayed latencies on the second day of the first session (p<0.05), and the second and third day of the second session (p< 0.05). 5. There were no statistical differences of mean response latencies among the rats in Group I, III and Ⅳ in the radial arm maze, but Group Ⅱ showed delayed latencies on the first and third day of the first session (p<0.05). 6. There was no influence from the administration of CCP and SB-CCP on the general behavior of the rats in Irwin´s test. These results suggest that Cervi cornu parvum and Soahbohyul - tang combined with Cervi comu parvum have anti-pyretic actions on LPS fever. The results also suggest that these drugs have no influence on learning and memory in young rats with LPS fever in the Morris water maze and the radial arm maze.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.562-565
• /
• 2022

Recently, the increasing number of overloaded vehicles on the road poses a risk to traffic safety, such as falling objects, road damage, and chain collisions due to the abnormal weight distribution, and can cause great damage once an accident occurs. However, this irregular weight distribution is not possible to be recognized with the current weight measurement system for vehicles on roads. To address this limitation, we propose to build an object detection-based AI model to identify overloaded vehicles that cause such social problems. In addition, we present a simple yet effective method to construct an object detection model for the large-scale vehicle images. In particular, we utilize the large-scale of vehicle image sets provided by open AI-Hub, which include the overloaded vehicles from the CCTV, black box, and hand-held camera point of view. We inspected the specific features of sizes of vehicles and types of image sources, and pre-processed these images to train a deep learning-based object detection model. Finally, we demonstrated that the detection performance of the overloaded vehicle was improved by about 23% compared to the one using raw data. From the result, we believe that public big data can be utilized more efficiently and applied to the development of an object detection-based overloaded vehicle detection model.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.24 no.7 s.178
• /
• pp.1661-1672
• /
• 2000

An immune system has powerful abilities such as memory, recognition and learning how to respond to invading antigens, and has been applied to many engineering algorithms in recent year. In this pap er, the combined optimization algorithm (Immune- Genetic Algorithm: IGA) is proposed for multi-optimization problems by introducing the capability of the immune system that controls the proliferation of clones to the genetic algorithm. The optimizing ability of the proposed combined algorithm is identified by comparing the result of optimization with simple genetic algorithm for two dimensional multi-peak function which have many local optimums. Also the new combined algorithm is applied to minimize the total weight of the shaft and the transmitted forces at the bearings. The inner diameter oil the shaft and the bearing stiffness are chosen as the design variables. The dynamic characteristics are determined by applying the generalized FEM. The results show that the combined algorithm and reduce both the weight of the shaft and the transmitted forces at the bearing with dynamic conatriants.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.12 no.4
• /
• pp.59-69
• /
• 2016

We proposed a specialized model of performance measurement to measure the training performance of the trainees in cyber practical training. Cyber security professionals are cultivating their expertise, skills, and competencies through cyber practical training in specialized education and training institutions. The our proposed process of trainee evaluation is consisted of an evaluation component discovery, evaluation item selection, evaluation index catalog, ratings and criteria decision, and calculation formula. The trainee evaluation is consisted of a formative evaluation during the training and an overall evaluation after finished training. Formative evaluation includes progress evaluation and participation evaluation, and overall evaluation includes practice evaluation and learning evaluation. The evaluation is weighted according to the importance of evaluation type. Because it is evaluated actual skills and abilities, competencies are assigned a high weight, while knowledge and attitudes are assigned a low weight. If cyber security trainees are evaluated by the proposed evaluation model, cyber security professionals can be cultivated by each skill and knowledge level and can be deployed by importance of security task.