• Journal of the Korea Society of Computer and Information
• /
• v.26 no.4
• /
• pp.47-53
• /
• 2021

In this paper, we propose a quadtree-based optimization technique that enables fast Super-resolution(SR) computation by efficiently classifying and dividing physics-based simulation data required to calculate SR. The proposed method reduces the time required for quadtree computation by downscaling the smoke simulation data used as input data. By binarizing the density of the smoke in this process, a quadtree is constructed while mitigating the problem of numerical loss of density in the downscaling process. The data used for training is the COCO 2017 Dataset, and the artificial neural network uses a VGG19-based network. In order to prevent data loss when passing through the convolutional layer, similar to the residual method, the output value of the previous layer is added and learned. In the case of smoke, the proposed method achieved a speed improvement of about 15 to 18 times compared to the previous approach.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.11
• /
• pp.4065-4083
• /
• 2021

Super-resolution can improve the clarity of low-resolution (LR) images, which can increase the accuracy of high-level compute vision tasks. Portable devices have low computing power and storage performance. Large-scale neural network super-resolution methods are not suitable for portable devices. In order to save the computational cost and the number of parameters, Lightweight image processing method can improve the processing speed of portable devices. Therefore, we propose the Enhanced Information Multiple Distillation Network (EIMDN) to adapt lower delay and cost. The EIMDN takes feedback mechanism as the framework and obtains low level features through high level features. Further, we replace the feature extraction convolution operation in Information Multiple Distillation Block (IMDB), with Ghost module, and propose the Enhanced Information Multiple Distillation Block (EIMDB) to reduce the amount of calculation and the number of parameters. Finally, coordinate attention (CA) is used at the end of IMDB and EIMDB to enhance the important information extraction from Spaces and channels. Experimental results show that our proposed can achieve convergence faster with fewer parameters and computation, compared with other lightweight super-resolution methods. Under the condition of higher peak signal-to-noise ratio (PSNR) and higher structural similarity (SSIM), the performance of network reconstruction image texture and target contour is significantly improved.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.12
• /
• pp.4420-4438
• /
• 2021

Image dehazing is an ill-posed problem which is far from being solved. Traditional image dehazing methods often yield mediocre effects and possess substandard processing speed, while modern deep learning methods perform best only in certain datasets. The haze removal effect when processed by said methods is unsatisfactory, meaning the generalization performance fails to meet the requirements. Concurrently, due to the limited processing speed, most dehazing algorithms cannot be employed in the industry. To alleviate said problems, a lightweight fast dehazing network based on a multiple scale-patch framework (MSP) is proposed in the present paper. Firstly, the multi-scale structure is employed as the backbone network and the multi-patch structure as the supplementary network. Dehazing through a single network causes problems, such as loss of object details and color in some image areas, the multi-patch structure was employed for MSP as an information supplement. In the algorithm image processing module, the image is segmented up and down for processed separately. Secondly, MSP generates a clear dehazing effect and significant robustness when targeting real-world homogeneous and nonhomogeneous hazy maps and different datasets. Compared with existing dehazing methods, MSP demonstrated a fast inference speed and the feasibility of real-time processing. The overall size and model parameters of the entire dehazing model are 20.75M and 6.8M, and the processing time for the single image is 0.026s. Experiments on NTIRE 2018 and NTIRE 2020 demonstrate that MSP can achieve superior performance among the state-of-the-art methods, such as PSNR, SSIM, LPIPS, and individual subjective evaluation.

• International Journal of Computer Science & Network Security
• /
• v.23 no.3
• /
• pp.177-186
• /
• 2023

Hyperspectral imaging technology is one of the most efficient and fast-growing technologies in recent years. Hyperspectral image (HSI) comprises contiguous spectral bands for every pixel that is used to detect the object with significant accuracy and details. HSI contains high dimensionality of spectral information which is not easy to classify every pixel. To confront the problem, we propose a novel RGB channel Assimilation for classification methods. The color features are extracted by using chromaticity computation. Additionally, this work discusses the classification of hyperspectral image based on Domain Transform Interpolated Convolution Filter (DTICF) and 3D-CNN with Bi-directional-Long Short Term Memory (Bi-LSTM). There are three steps for the proposed techniques: First, HSI data is converted to RGB images with spatial features. Before using the DTICF, the RGB images of HSI and patch of the input image from raw HSI are integrated. Afterward, the pair features of spectral and spatial are excerpted using DTICF from integrated HSI. Those obtained spatial and spectral features are finally given into the designed 3D-CNN with Bi-LSTM framework. In the second step, the excerpted color features are classified by 2D-CNN. The probabilistic classification map of 3D-CNN-Bi-LSTM, and 2D-CNN are fused. In the last step, additionally, Markov Random Field (MRF) is utilized for improving the fused probabilistic classification map efficiently. Based on the experimental results, two different hyperspectral images prove that novel RGB channel assimilation of DTICF-3D-CNN-Bi-LSTM approach is more important and provides good classification results compared to other classification approaches.

• Journal of Korea Water Resources Association
• /
• v.56 no.10
• /
• pp.679-690
• /
• 2023

Leakage is one of the representative abnormal conditions in Water distribution systems (WDSs). Leakage can potentially occur and cause immediate economic and hydraulic damage upon occurrence. Therefore, leakage detection is essential, but WDSs are located underground, it is difficult. Moreover, when multiple leakage occurs, it is required to prioritize restoration according to the scale and location of the leakage, applying for an optimal restoration framework can be advantageous in terms of system resilience. In this study, various leakage scenarios were generated based on the WDSs hydraulic model, and leakage detection was carried out containing location and scale using a Deep learning-based model. Finally, the leakage location and scale obtained from the detection results were used as a factor for the priority of leakage restoration, and the results of the priority of leakage restoration were derived. The priority of leakage restoration considered not only hydraulic factors but also socio-economic factors (e.g., leakage scale, important facilities).

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.12
• /
• pp.57-66
• /
• 2023

In this study, we propose a novel research framework for the recommendation system that can estimate the user's emotional state and reflect it in the recommendation process by applying deep learning techniques and emotion AI (artificial intelligence). To this end, we build an emotion classification model that classifies each of the seven emotions of angry, disgust, fear, happy, sad, surprise, and neutral, respectively, and propose a model that can reflect this result in the recommendation process. However, in the general emotion classification data, the difference in distribution ratio between each label is large, so it may be difficult to expect generalized classification results. In this study, since the number of emotion data such as disgust in emotion image data is often insufficient, correction is made through augmentation. Lastly, we propose a method to reflect the emotion prediction model based on data through image augmentation in the recommendation systems.

• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.163-177
• /
• 2019

As smartphones are getting widely used, human activity recognition (HAR) tasks for recognizing personal activities of smartphone users with multimodal data have been actively studied recently. The research area is expanding from the recognition of the simple body movement of an individual user to the recognition of low-level behavior and high-level behavior. However, HAR tasks for recognizing interaction behavior with other people, such as whether the user is accompanying or communicating with someone else, have gotten less attention so far. And previous research for recognizing interaction behavior has usually depended on audio, Bluetooth, and Wi-Fi sensors, which are vulnerable to privacy issues and require much time to collect enough data. Whereas physical sensors including accelerometer, magnetic field and gyroscope sensors are less vulnerable to privacy issues and can collect a large amount of data within a short time. In this paper, a method for detecting accompanying status based on deep learning model by only using multimodal physical sensor data, such as an accelerometer, magnetic field and gyroscope, was proposed. The accompanying status was defined as a redefinition of a part of the user interaction behavior, including whether the user is accompanying with an acquaintance at a close distance and the user is actively communicating with the acquaintance. A framework based on convolutional neural networks (CNN) and long short-term memory (LSTM) recurrent networks for classifying accompanying and conversation was proposed. First, a data preprocessing method which consists of time synchronization of multimodal data from different physical sensors, data normalization and sequence data generation was introduced. We applied the nearest interpolation to synchronize the time of collected data from different sensors. Normalization was performed for each x, y, z axis value of the sensor data, and the sequence data was generated according to the sliding window method. Then, the sequence data became the input for CNN, where feature maps representing local dependencies of the original sequence are extracted. The CNN consisted of 3 convolutional layers and did not have a pooling layer to maintain the temporal information of the sequence data. Next, LSTM recurrent networks received the feature maps, learned long-term dependencies from them and extracted features. The LSTM recurrent networks consisted of two layers, each with 128 cells. Finally, the extracted features were used for classification by softmax classifier. The loss function of the model was cross entropy function and the weights of the model were randomly initialized on a normal distribution with an average of 0 and a standard deviation of 0.1. The model was trained using adaptive moment estimation (ADAM) optimization algorithm and the mini batch size was set to 128. We applied dropout to input values of the LSTM recurrent networks to prevent overfitting. The initial learning rate was set to 0.001, and it decreased exponentially by 0.99 at the end of each epoch training. An Android smartphone application was developed and released to collect data. We collected smartphone data for a total of 18 subjects. Using the data, the model classified accompanying and conversation by 98.74% and 98.83% accuracy each. Both the F1 score and accuracy of the model were higher than the F1 score and accuracy of the majority vote classifier, support vector machine, and deep recurrent neural network. In the future research, we will focus on more rigorous multimodal sensor data synchronization methods that minimize the time stamp differences. In addition, we will further study transfer learning method that enables transfer of trained models tailored to the training data to the evaluation data that follows a different distribution. It is expected that a model capable of exhibiting robust recognition performance against changes in data that is not considered in the model learning stage will be obtained.

• Journal of Intelligence and Information Systems
• /
• v.29 no.3
• /
• pp.37-55
• /
• 2023

With the development of electronic journals and the emergence of various interdisciplinary studies, the selection of journals for publication has become a new challenge for researchers. Even if a paper is of high quality, it may face rejection due to a mismatch between the paper's topic and the scope of the journal. While research on assisting researchers in journal selection has been actively conducted in English, the same cannot be said for Korean journals. In this study, we propose a system that recommends Korean journals for submission. Firstly, we utilize SBERT (Sentence BERT) to embed abstracts of previously published papers at the document level, compare the similarity between new documents and published papers, and recommend journals accordingly. Next, the order of recommended journals is determined by considering the similarity of abstracts, keywords, and title. Subsequently, journals that are similar to the top recommended journal from previous stage are added by using a dictionary of words constructed for each journal, thereby enhancing recommendation diversity. The recommendation system, built using this approach, achieved a Top-10 accuracy level of 76.6%, and the validity of the recommendation results was confirmed through user feedback. Furthermore, it was found that each step of the proposed framework contributes to improving recommendation accuracy. This study provides a new approach to recommending academic journals in the Korean language, which has not been actively studied before, and it has also practical implications as the proposed framework can be easily applied to services.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.7
• /
• pp.326-334
• /
• 2024

Extra-oral imaging techniques such as Panoramic X-rays (PXs) and Cone Beam Computed Tomography (CBCT) are the most preferred imaging modalities in dental clinics owing to its patient convenience during imaging as well as their ability to visualize entire teeth information. PXs are preferred for routine clinical treatments and CBCTs for complex surgeries and implant treatments. However, PXs are limited by the lack of third dimensional spatial information whereas CBCTs inflict high radiation exposure to patient. When a PX is already available, it is beneficial to reconstruct the 3D oral structure from the PX to avoid further expenses and radiation dose. In this paper, we propose 3DentAI - an U-Net based deep learning framework for 3D reconstruction of oral structure from a PX image. Our framework consists of three module - a reconstruction module based on attention U-Net for estimating depth from a PX image, a realignment module for aligning the predicted flattened volume to the shape of jaw using a predefined focal trough and ray data, and lastly a refinement module based on 3D U-Net for interpolating the missing information to obtain a smooth representation of oral cavity. Synthetic PXs obtained from CBCT by ray tracing and rendering were used to train the networks without the need of paired PX and CBCT datasets. Our method, trained and tested on a diverse datasets of 600 patients, achieved superior performance to GAN-based models even with low computational complexity.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.5
• /
• pp.414-423
• /
• 2020

In this paper, we propose a Dilated Convolution Gate Linear Unit (DCGLU) to mitigate the lack of sparsity and small receptive field problems caused by the segmentation map extraction process in sound event detection with weak labels. In the advent of deep learning framework, segmentation map extraction approaches have shown improved performance in noisy environments. However, these methods are forced to maintain the size of the feature map to extract the segmentation map as the model would be constructed without a pooling operation. As a result, the performance of these methods is deteriorated with a lack of sparsity and a small receptive field. To mitigate these problems, we utilize GLU to control the flow of information and Dilated Convolutional Neural Networks (DCNNs) to increase the receptive field without additional learning parameters. For the performance evaluation, we employ a URBAN-SED and self-organized bird sound dataset. The relevant experiments show that our proposed DCGLU model outperforms over other baselines. In particular, our method is shown to exhibit robustness against nature sound noises with three Signal to Noise Ratio (SNR) levels (20 dB, 10 dB and 0 dB).