• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.10
• /
• pp.3729-3749
• /
• 2021

At present, deep convolution network-based salient object detection (SOD) has achieved impressive performance. However, it is still a challenging problem to make full use of the multi-scale information of the extracted features and which appropriate feature fusion method is adopted to process feature mapping. In this paper, we propose a new adjacency auxiliary network (AANet) based on multi-scale feature fusion for SOD. Firstly, we design the parallel connection feature enhancement module (PFEM) for each layer of feature extraction, which improves the feature density by connecting different dilated convolution branches in parallel, and add channel attention flow to fully extract the context information of features. Then the adjacent layer features with close degree of abstraction but different characteristic properties are fused through the adjacent auxiliary module (AAM) to eliminate the ambiguity and noise of the features. Besides, in order to refine the features effectively to get more accurate object boundaries, we design adjacency decoder (AAM_D) based on adjacency auxiliary module (AAM), which concatenates the features of adjacent layers, extracts their spatial attention, and then combines them with the output of AAM. The outputs of AAM_D features with semantic information and spatial detail obtained from each feature are used as salient prediction maps for multi-level feature joint supervising. Experiment results on six benchmark SOD datasets demonstrate that the proposed method outperforms similar previous methods.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.11
• /
• pp.5580-5593
• /
• 2019

A malicious Uniform Resource Locator (URL) recognition and detection method based on the combination of Attention mechanism with Convolutional Neural Network and Long Short-Term Memory Network (Attention-Based CNN-LSTM), is proposed. Firstly, the WHOIS check method is used to extract and filter features, including the URL texture information, the URL string statistical information of attributes and the WHOIS information, and the features are subsequently encoded and pre-processed followed by inputting them to the constructed Convolutional Neural Network (CNN) convolution layer to extract local features. Secondly, in accordance with the weights from the Attention mechanism, the generated local features are input into the Long-Short Term Memory (LSTM) model, and subsequently pooled to calculate the global features of the URLs. Finally, the URLs are detected and classified by the SoftMax function using global features. The results demonstrate that compared with the existing methods, the Attention-based CNN-LSTM mechanism has higher accuracy for malicious URL detection.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.2
• /
• pp.235-241
• /
• 2022

Several studies using Convolutional Neural Network (CNN) have been conducted to detect and classify the sounds of marine mammals in underwater acoustic data collected through passive acoustic monitoring. In this study, the possibility of automatic classification of bearded seal sounds was confirmed using a CNN model based on the underwater acoustic spectrogram images collected from August 2017 to August 2018 in East Siberian Sea. When only the clear seal sound was used as training dataset, overfitting due to memorization was occurred. By evaluating the entire training data by replacing some training data with data containing noise, it was confirmed that overfitting was prevented as the model was generalized more than before with accuracy (0.9743), precision (0.9783), recall (0.9520). As a result, the performance of the classification model for bearded seals signal has improved when the noise was included in the training data.

• International Journal of Computer Science & Network Security
• /
• v.23 no.3
• /
• pp.57-66
• /
• 2023

Rice is an important food crop for most of the population in the world and it is largely cultivated in Pakistan. It not only fulfills food demand in the country but also contributes to the wealth of Pakistan. But its production can be affected by climate change. The irregularities in the climate can cause several diseases such as brown spots, bacterial blight, tungro and leaf blasts, etc. Detection of these diseases is necessary for suitable treatment. These diseases can be effectively detected using deep learning such as Convolution Neural networks. Due to the small dataset, transfer learning models such as vgg16 model can effectively detect the diseases. In this paper, vgg16, inception and xception models are used. Vgg16, inception and xception models have achieved 99.22%, 88.48% and 93.92% validation accuracies when the epoch value is set to 10. Evaluation of models has also been done using accuracy, recall, precision, and confusion matrix.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.5
• /
• pp.1778-1797
• /
• 2021

Document image binarization is an important pre-processing step in document analysis and archiving. The state-of-the-art models for document image binarization are variants of encoder-decoder architectures, such as FCN (fully convolutional network) and U-Net. Despite their success, they still suffer from three limitations: (1) reduced feature map resolution due to consecutive strided pooling or convolutions, (2) multiple scales of target objects, and (3) reduced localization accuracy due to the built-in invariance of deep convolutional neural networks (DCNNs). To overcome these three challenges, we propose an improved semantic segmentation model, referred to as DP-LinkNet, which adopts the D-LinkNet architecture as its backbone, with the proposed hybrid dilated convolution (HDC) and spatial pyramid pooling (SPP) modules between the encoder and the decoder. Extensive experiments are conducted on recent document image binarization competition (DIBCO) and handwritten document image binarization competition (H-DIBCO) benchmark datasets. Results show that our proposed DP-LinkNet outperforms other state-of-the-art techniques by a large margin. Our implementation and the pre-trained models are available at https://github.com/beargolden/DP-LinkNet.

• Journal of Broadcast Engineering
• /
• v.26 no.3
• /
• pp.296-307
• /
• 2021

A digital hologram (DH) is an ultra-high value-added video content that includes 3D information in 2D data. Therefore, its intellectual property rights must be protected for its distribution. For this, this paper proposes a watermarking method of DH using a deep neural network. This method is a watermark (WM) invisibility, attack robustness, and blind watermarking method that does not use host information in WM extraction. The proposed network consists of four sub-networks: pre-processing for each of the host and WM, WM embedding watermark, and WM extracting watermark. This network expand the WM data to the host instead of shrinking host data to WM and concatenate it to the host to insert the WM by considering the characteristics of a DH having a strong high frequency component. In addition, in the training of this network, the difference in performance according to the data distribution property of DH is identified, and a method of selecting a training data set with the best performance in all types of DH is presented. The proposed method is tested for various types and strengths of attacks to show its performance. It also shows that this method has high practicality as it operates independently of the resolution of the host DH and WM data.

• Journal of the Semiconductor & Display Technology
• /
• v.17 no.2
• /
• pp.67-70
• /
• 2018

Object detection and tracking is an exciting and interesting research area in the field of computer vision, and its technologies have been widely used in various application systems such as surveillance, military, and augmented reality. This paper proposes and implements a novel and more robust object recognition and tracking system to localize and track multiple objects from input images, which estimates target state using the likelihoods obtained from multiple CNNs. As the experimental result, the proposed algorithm is effective to handle multi-modal target appearances and other exceptions.

• KIPS Transactions on Computer and Communication Systems
• /
• v.10 no.7
• /
• pp.191-198
• /
• 2021

Since the size of training dataset become large and the model is getting deeper to achieve high accuracy in deep learning, the deep neural network training requires a lot of computation and it takes too much time with a single node. Therefore, distributed deep learning is proposed to reduce the training time by distributing computation across multiple nodes. In this study, we propose hybrid allreduce strategy that considers the characteristics of each layer and communication and computational overlapping technique for synchronization of distributed deep learning. Since the convolution layer has fewer parameters than the fully-connected layer as well as it is located at the upper, only short overlapping time is allowed. Thus, butterfly allreduce is used to synchronize the convolution layer. On the other hand, fully-connecter layer is synchronized using ring all-reduce. The empirical experiment results on PyTorch with our proposed scheme shows that the proposed method reduced the training time by up to 33% compared to the baseline PyTorch.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.8
• /
• pp.999-1005
• /
• 2020

While search portals' 'Portal News' account for the largest portion of aggregated news outlet, its neutrality as an outlet is questionable. This is because news aggregation may lead to prejudiced information consumption by recommending biased news articles. In this paper we introduce a new method of measuring political bias of news articles by using deep learning. It can provide its readers with insights on critical thinking. For this method, we build the dataset for deep learning by analyzing articles' bias from keywords, sourced from the National Assembly proceedings, and assigning bias to said keywords. Based on these data, news article bias is calculated by applying deep learning with a combination of Convolution Neural Network and Recurrent Neural Network. Using this method, 95.6% of sentences are correctly distinguished as either conservative or progressive-biased; on the entire article, the accuracy is 46.0%. This enables analyzing any articles' bias between conservative and progressive unlike previous methods that were limited on article subjects.

• Smart Media Journal
• /
• v.12 no.7
• /
• pp.59-67
• /
• 2023

Recently, research on behavior analysis tracking human posture and movement has been actively conducted. In particular, OpenPose, an open-source software developed by CMU in 2017, is a representative method for estimating human appearance and behavior. OpenPose can detect and estimate various body parts of a person, such as height, face, and hands in real-time, making it applicable to various fields such as smart healthcare, exercise training, security systems, and medical fields. In this paper, we propose a method for classifying four exercise movements - Squat, Walk, Wave, and Fall-down - which are most commonly performed by users in the gym, using OpenPose-based deep learning models, DNN and CNN. The training data is collected by capturing the user's movements through recorded videos and real-time camera captures. The collected dataset undergoes preprocessing using OpenPose. The preprocessed dataset is then used to train the proposed DNN and CNN models for exercise movement classification. The performance errors of the proposed models are evaluated using MSE, RMSE, and MAE. The performance evaluation results showed that the proposed DNN model outperformed the proposed CNN model.