• Journal of Information Processing Systems
• /
• v.17 no.5
• /
• pp.892-904
• /
• 2021

An image semantic segmentation model is proposed based on improved ENet network in order to achieve the low accuracy of image semantic segmentation in complex environment. Firstly, this paper performs pruning and convolution optimization operations on the ENet network. That is, the network structure is reasonably adjusted for better results in image segmentation by reducing the convolution operation in the decoder and proposing the bottleneck convolution structure. Squeeze-and-excitation (SE) module is then integrated into the optimized ENet network. Small-scale targets see improvement in segmentation accuracy via automatic learning of the importance of each feature channel. Finally, the experiment was verified on the public dataset. This method outperforms the existing comparison methods in mean pixel accuracy (MPA) and mean intersection over union (MIOU) values. And in a short running time, the accuracy of the segmentation and the efficiency of the operation are guaranteed.

• Journal of information and communication convergence engineering
• /
• v.19 no.3
• /
• pp.148-154
• /
• 2021

With the advent of context-aware computing, many attempts were made to understand emotions. Among these various attempts, Speech Emotion Recognition (SER) is a method of recognizing the speaker's emotions through speech information. The SER is successful in selecting distinctive 'features' and 'classifying' them in an appropriate way. In this paper, the performances of SER using neural network models (e.g., fully connected network (FCN), convolutional neural network (CNN)) with Mel-Frequency Cepstral Coefficients (MFCC) are examined in terms of the accuracy and distribution of emotion recognition. For Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS) dataset, by tuning model parameters, a two-dimensional Convolutional Neural Network (2D-CNN) model with MFCC showed the best performance with an average accuracy of 88.54% for 5 emotions, anger, happiness, calm, fear, and sadness, of men and women. In addition, by examining the distribution of emotion recognition accuracies for neural network models, the 2D-CNN with MFCC can expect an overall accuracy of 75% or more.

• Smart Media Journal
• /
• v.7 no.4
• /
• pp.79-89
• /
• 2018

In recent years, falls among elderly people have gained serious attention as a major cause of injuries. Falls often lead to fatal consequences due to lack of prompt response and rescue. Therefore, a more accurate fall detection system and an effective feature extraction technique are required to prevent and reduce the risk of such incidents. In this paper, we proposed an efficient feature extraction technique based on multiple sliding windows and validated it through a series of experiments using supervised learning algorithms. The experiments were conducted using the public datasets obtained from tri-axial accelerometers. The results depicted that extraction of the feature from adjacent sliding windows led to high accuracy in supervised machine learning-based fall detection. Also, the experiments conducted in this study suggested that the best accuracy can be achieved by keeping the window size as small as 2 seconds. With the kNN classifier and dataset from wearable sensors, the experiments achieved accuracy rates of 94%.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.68 no.1
• /
• pp.189-198
• /
• 2019

In recent years, visual-inertial odometry(VIO) algorithms have been extensively studied for the indoor/urban environments because it is more robust to dynamic scenes and environment changes. In this paper, we propose loosely coupled(LC) VIO algorithm that utilizes the velocity vectors from both visual odometry(VO) and inertial measurement unit(IMU) as a filter measurement of Extended Kalman filter. Our approach improves the estimation performance of a filter without adding extra sensors while maintaining simple integration framework, which treats VO as a black box. For the VO algorithm, we employed a fundamental part of the ORB-SLAM, which uses ORB features. We performed an outdoor experiment using an RGB-D camera to evaluate the accuracy of the presented algorithm. Also, we evaluated our algorithm with the public dataset to compare with other visual navigation systems.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.5
• /
• pp.521-528
• /
• 2019

Epilepsy is one of the most prevalent neurological diseases. Electroencephalogram (EEG) signals are widely used for monitoring and diagnosis tool for epileptic seizure. Typically, a huge amount of EEG signals is needed, where they are visually examined by experienced clinicians. In this study, we propose a simple automatic seizure detection framework using intracranial EEG signals. We suggest a sparse approximation based classification (SAC) scheme by solving overdetermined system. L1-norm minimization algorithms are utilized for efficient sparse signal recovery. For evaluation of the proposed scheme, the public EEG dataset obtained by five healthy subjects and five epileptic patients is utilized. The results show that the proposed fast L1-norm minimization based SAC methods achieve the 99.5% classification accuracy which is 1% improved result than the conventional L2 norm based method with negligibly increased execution time (42msec).

• International Journal of Computer Science & Network Security
• /
• v.21 no.6
• /
• pp.101-106
• /
• 2021

Breast cancer is the number one cause of deaths from cancer in women, knowing the type of breast cancer in the early stages can help us to prevent the dangers of the next stage. The performance of the deep learning depends on large number of labeled data, this paper presented convolutional neural network for classification breast cancer from images to benign or malignant. our network contains 11 layers and ends with softmax for the output, the experiments result using public BreakHis dataset, and the proposed methods outperformed the state-of-the-art methods.

• Journal of Korea Multimedia Society
• /
• v.25 no.10
• /
• pp.1433-1447
• /
• 2022

Recently, as the demand for physical security technology to prevent leakage of technical and business information of companies and public institutions increases, the high tech companies are operating X-ray security checkpoints at building entrances to protect their intellectual property and technology. X-ray security checkpoints are operated to detect cameras and storage media that may store or leak important technologies in the bags of people entering and leaving the building. In this study, we propose an X-ray security checkpoint system that automatically detects a storage medium in an X-ray image using a deep learning based object detection method. The proposed system consists of an edge computing unit and a cloud-computing unit. We employ the RetinaNet for automatic storage media detection in the X-ray security checkpoint images. The proposed approach achieved mAP of 95.92% on private dataset.

• The Journal of Korea Robotics Society
• /
• v.18 no.3
• /
• pp.330-336
• /
• 2023

In this paper, a novel line-based simultaneous localization and mapping (SLAM) using a loss function of vanishing point measurements is proposed. In general, the Huber norm is used as a loss function for point and line features in feature-based SLAM. The proposed loss function of vanishing point measurements is based on the unit sphere model. Because the point and line feature measurements define the reprojection error in the image plane as a residual, linear loss functions such as the Huber norm is used. However, the typical loss functions are not suitable for vanishing point measurements with unbounded problems. To tackle this problem, we propose a loss function for vanishing point measurements. The proposed loss function is based on unit sphere model. Finally, we prove the validity of the loss function for vanishing point through experiments on a public dataset.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.11a
• /
• pp.171-174
• /
• 2022

Weakly-supervised learning is a widely adopted approach in video anomaly detection whereby only video labels are utilized instead of expensive frame-level annotations. Since the success of multi-instance learning (MIL), almost all recent approaches are based on maximizing the margin between the set of abnormal video snippets and those of normal video snippets. In this work, we present a simple contrastive approach for weakly supervised video anomaly detection (WS-VAD) with aims to enhance the performance of existing models. The method is generic in nature and introduces a loss function to encourage attraction of output features from the same video class and repel those from different video classes. Experimental results demonstrate our method can be applied to existing algorithms to improve detection accuracy in public video anomaly dataset.

• The Journal of Korea Robotics Society
• /
• v.18 no.1
• /
• pp.66-71
• /
• 2023

As cameras have become primary sensors for mobile robots, vision based Simultaneous Localization and Mapping (SLAM) has achieved impressive results with the recent development of computer vision and deep learning. However, vision information has a disadvantage in that a lot of information disappears in a low-light environment. To overcome the problem, we propose an image enhancement method to perform visual SLAM in a low-light environment. Using the deep generative adversarial models and modified gamma correction, the quality of low-light images were improved. The proposed method is less sharp than the existing method, but it can be applied to ORB-SLAM in real time by dramatically reducing the amount of computation. The experimental results were able to prove the validity of the proposed method by applying to public Dataset TUM and VIVID++.