• Transactions on Semiconductor Engineering
• /
• v.2 no.1
• /
• pp.17-20
• /
• 2024

This paper presents a novel lightweight object detection model tailored for low-powered edge devices, addressing the limitations of traditional resource-intensive computer vision models. Our proposed detector, inspired by the Single Shot Detector (SSD), employs a compact yet robust network design. Crucially, it integrates an 'enhancer block' that significantly boosts its efficiency in detecting smaller objects. The model comprises two primary components: the Light_Block for efficient feature extraction using Depth-wise and Pointwise Convolution layers, and the Enhancer_Block for enhanced detection of tiny objects. Trained from scratch on the Udacity Annotated Dataset with image dimensions of 300x480, our model eschews the need for pre-trained classification weights. Weighing only 5.5MB with approximately 0.43M parameters, our detector achieved a mean average precision (mAP) of 27.7% and processed at 140 FPS, outperforming conventional models in both precision and efficiency. This research underscores the potential of lightweight designs in advancing object detection for edge devices without compromising accuracy.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.3
• /
• pp.570-590
• /
• 2024

Breast cancer ranks among the most prevalent forms of malignancy and foremost cause of death by cancer worldwide. It is not preventable. Early and precise detection is the only remedy for lowering the rate of mortality and improving the probability of survival for victims. In contrast to present procedures, thermography aids in the early diagnosis of cancer and thereby saves lives. But the accuracy experiences detrimental impact by low sensitivity for small and deep tumours and the subjectivity by physicians in interpreting the images. Employing deep learning approaches for cancer detection can enhance the efficacy. This study explored the utilization of thermography in early identification of breast cancer with the use of a publicly released dataset known as the DMR-IR dataset. For this purpose, we employed a novel approach that entails the utilization of a pre-trained MobileNetV2 model and fine tuning it through transfer learning techniques. We created three models using MobileNetV2: one was a baseline transfer learning model with weights trained from ImageNet dataset, the second was a fine-tuned model with an adaptive learning rate, and the third utilized early stopping with callbacks during fine-tuning. The results showed that the proposed methods achieved average accuracy rates of 85.15%, 95.19%, and 98.69%, respectively, with various performance indicators such as precision, sensitivity and specificity also being investigated.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.191-206
• /
• 2022

Recently, it is a de-facto approach to utilize a pre-trained language model(PLM) to achieve the state-of-the-art performance for various natural language tasks(called downstream tasks) such as sentiment analysis and question answering. However, similar to any other machine learning method, PLM tends to depend on the data distribution seen during the training phase and shows worse performance on the unseen (Out-of-Distribution) domain. Due to the aforementioned reason, there have been many efforts to develop domain-specified PLM for various fields such as medical and legal industries. In this paper, we discuss the training of a finance domain-specified PLM for the Korean language and its applications. Our finance domain-specified PLM, KB-BERT, is trained on a carefully curated financial corpus that includes domain-specific documents such as financial reports. We provide extensive performance evaluation results on three natural language tasks, topic classification, sentiment analysis, and question answering. Compared to the state-of-the-art Korean PLM models such as KoELECTRA and KLUE-RoBERTa, KB-BERT shows comparable performance on general datasets based on common corpora like Wikipedia and news articles. Moreover, KB-BERT outperforms compared models on finance domain datasets that require finance-specific knowledge to solve given problems.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.1
• /
• pp.147-165
• /
• 2021

In this work, we empirically evaluated the efficiency of the recent EfficientNetB0 model to identify and diagnose malaria parasite infections in blood smears. The dataset used was collected and classified by relevant experts from the Lister Hill National Centre for Biomedical Communications (LHNCBC). We prepared our samples with minimal image transformations as opposed to others, as we focused more on the feature extraction capability of the EfficientNetB0 baseline model. We applied transfer learning to increase the initial feature sets and reduced the training time to train our model. We then fine-tuned it to work with our proposed layers and re-trained the entire model to learn from our prepared dataset. The highest overall accuracy attained from our evaluated results was 94.70% from fifty epochs and followed by 94.68% within just ten. Additional visualization and analysis using the Gradient-weighted Class Activation Mapping (Grad-CAM) algorithm visualized how effectively our fine-tuned EfficientNetB0 detected infections better than other recent state-of-the-art DCNN models. This study, therefore, concludes that when fine-tuned, the recent EfficientNetB0 will generate highly accurate deep learning solutions for the identification of malaria parasites in blood smears without the need for stringent pre-processing, optimization, or data augmentation of images.

• Journal of Sensor Science and Technology
• /
• v.30 no.6
• /
• pp.429-435
• /
• 2021

In this study, the creation of emotional dialogue was investigated within the process of developing a robot's natural language understanding and emotional dialogue processing. Unlike an English-based dataset, which is the mainstay of natural language processing, the Korean-based dataset has several shortcomings. Therefore, in a situation where the Korean language base is insufficient, the Korean dataset should be dealt with in detail, and in particular, the unique characteristics of the language should be considered. Hence, the first step is to base this study on a specific Korean dataset consisting of conversations on emotional topics. Subsequently, a model was built that learns to extract the continuous dialogue features from a pre-trained language model to generate sentences while maintaining the context of the dialogue. To validate the model, a chatbot system was implemented and meaningful results were obtained by collecting the external subjects and conducting experiments. As a result, the proposed model was influenced by the dataset in which the conversation topic was consultation, to facilitate free and emotional communication with users as if they were consulting with a chatbot. The results were analyzed to identify and explain the advantages and disadvantages of the current model. Finally, as a necessary element to reach the aforementioned ultimate research goal, a discussion is presented on the areas for future studies.

• Annual Conference of KIPS
• /
• 2023.05a
• /
• pp.706-708
• /
• 2023

In this study, we embark on a journey to uncover the essence of emotions by exploring the depths of transfer learning on three pre-trained transformer models. Our quest to classify five emotions culminates in discovering the KLUE (Korean Language Understanding Evaluation)-BERT (Bidirectional Encoder Representations from Transformers) model, which is the most exceptional among its peers. Our analysis of F1 scores attests to its superior learning and generalization abilities on the experimental data. To delve deeper into the mystery behind its success, we employ the powerful SHAP (Shapley Additive Explanations) method to unravel the intricacies of the KLUE-BERT model. The findings of our investigation are presented with a mesmerizing text plot visualization, which serves as a window into the model's soul. This approach enables us to grasp the impact of individual tokens on emotion classification and provides irrefutable, visually appealing evidence to support the predictions of the KLUE-BERT model.

• Smart Media Journal
• /
• v.13 no.8
• /
• pp.16-25
• /
• 2024

Recently, there has been a growing emphasis on identifying both known and unknown diseases in plant disease recognition. In this task, a model trained only on images of known classes is required to classify an input image into either one of the known classes or into an unknown class. Consequently, the capability to recognize unknown diseases is critical for model deployment. To enhance this capability, we are considering three factors. Firstly, we propose a new logits-based scoring function for unknown scores. Secondly, initial experiments indicate that a compact feature space is crucial for the effectiveness of logits-based methods, leading us to employ the AM-Softmax loss instead of Cross-entropy loss during training. Thirdly, drawing inspiration from the efficacy of transfer learning, we utilize a large plant-relevant dataset, PlantCLEF2022, for pre-training a model. The experimental results suggest that our method outperforms current algorithms. Specifically, our method achieved a performance of 97.90 CSA, 91.77 AUROC, and 90.63 OSCR with the ResNet50 model and a performance of 98.28 CSA, 92.05 AUROC, and 91.12 OSCR with the ConvNext base model. We believe that our study will contribute to the community.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.3
• /
• pp.33-43
• /
• 2022

In this paper, we propose an ensemble model facilitated by multi-channel palm images with attention U-Net models and pretrained convolutional neural networks (CNNs) for establishing a contactless palm-based user identification system using conventional inexpensive camera sensors. Attention U-Net models are used to extract the areas of interest including hands (i.e., with fingers), palms (i.e., without fingers) and palm lines, which are combined to generate three channels being ped into the ensemble classifier. Then, the proposed palm information-based user identification system predicts the class using the classifier ensemble with three outperforming pre-trained CNN models. The proposed model demonstrates that the proposed model could achieve the classification accuracy, precision, recall, F1-score of 98.60%, 98.61%, 98.61%, 98.61% respectively, which indicate that the proposed model is effective even though we are using very cheap and inexpensive image sensors. We believe that in this COVID-19 pandemic circumstances, the proposed palm-based contactless user identification system can be an alternative, with high safety and reliability, compared with currently overwhelming contact-based systems.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.4
• /
• pp.159-172
• /
• 2023

In general, social problem-solving research aims to create important social value by offering meaningful answers to various social pending issues using scientific technologies. Not surprisingly, however, although numerous and extensive research attempts have been made to alleviate the social problems and issues in nation-wide, we still have many important social challenges and works to be done. In order to facilitate the entire process of the social problem-solving research and maximize its efficacy, it is vital to clearly identify and grasp the important and pressing problems to be focused upon. It is understandable for the problem discovery step to be drastically improved if current social issues can be automatically identified from existing R&D resources such as technical reports and articles. This paper introduces a comprehensive dataset which is essential to build a machine learning model for automatically detecting the social problems and solutions in various national research reports. Initially, we collected a total of 700 research reports regarding social problems and issues. Through intensive annotation process, we built totally 24,022 sentences each of which possesses its own category or label closely related to social problem-solving such as problems, purposes, solutions, effects and so on. Furthermore, we implemented four sentence classification models based on various neural language models and conducted a series of performance experiments using our dataset. As a result of the experiment, the model fine-tuned to the KLUE-BERT pre-trained language model showed the best performance with an accuracy of 75.853% and an F1 score of 63.503%.

• Structural Engineering and Mechanics
• /
• v.46 no.5
• /
• pp.673-693
• /
• 2013

This paper presents a systematic design and training procedure for the feed-forward back-propagation neural network (NN) modeling of both forward and inverse behavior of a rotary magnetorheological (MR) damper based on experimental data. For the forward damper model, with damper force as output, an optimization procedure demonstrates accurate training of the NN architecture with only current and velocity as input states. For the inverse damper model, with current as output, the absolute value of velocity and force are used as input states to avoid negative current spikes when tracking a desired damper force. The forward and inverse damper models are trained and validated experimentally, combining a limited number of harmonic displacement records, and constant and half-sinusoidal current records. In general the validation shows accurate results for both forward and inverse damper models, where the observed modeling errors for the inverse model can be related to knocking effects in the measured force due to the bearing plays between hydraulic piston and MR damper rod. Finally, the validated models are used to emulate pure viscous damping. Comparison of numerical and experimental results demonstrates good agreement in the post-yield region of the MR damper, while the main error of the inverse NN occurs in the pre-yield region where the inverse NN overestimates the current to track the desired viscous force.