• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.6
• /
• pp.1818-1832
• /
• 2022

Violence in the Internet era poses a new challenge to the current counter-riot work, and according to research and analysis, most of the violent incidents occurring are related to the dissemination of violence images. The use of the popular deep learning neural network to automatically analyze the massive amount of images on the Internet has become one of the important tools in the current counter-violence work. This paper focuses on the use of transfer learning techniques and the introduction of an attention mechanism to the residual network (ResNet) model for the classification and identification of violence images. Firstly, the feature elements of the violence images are identified and a targeted dataset is constructed; secondly, due to the small number of positive samples of violence images, pre-training and attention mechanisms are introduced to suggest improvements to the traditional residual network; finally, the improved model is trained and tested on the constructed dedicated dataset. The research results show that the improved network model can quickly and accurately identify violence images with an average accuracy rate of 92.20%, thus effectively reducing the cost of manual identification and providing decision support for combating rebel organization activities.

• Applied Microscopy
• /
• v.50
• /
• pp.25.1-25.11
• /
• 2020

Scanning acoustic microscopy (SAM) or Acoustic Micro Imaging (AMI) is a powerful, non-destructive technique that can detect hidden defects in elastic and biological samples as well as non-transparent hard materials. By monitoring the internal features of a sample in three-dimensional integration, this technique can efficiently find physical defects such as cracks, voids, and delamination with high sensitivity. In recent years, advanced techniques such as ultrasound impedance microscopy, ultrasound speed microscopy, and scanning acoustic gigahertz microscopy have been developed for applications in industries and in the medical field to provide additional information on the internal stress, viscoelastic, and anisotropic, or nonlinear properties. X-ray, magnetic resonance, and infrared techniques are the other competitive and widely used methods. However, they have their own advantages and limitations owing to their inherent properties such as different light sources and sensors. This paper provides an overview of the principle of SAM and presents a few results to demonstrate the applications of modern acoustic imaging technology. A variety of inspection modes, such as vertical, horizontal, and diagonal cross-sections have been presented by employing the focus pathway and image reconstruction algorithm. Images have been reconstructed from the reflected echoes resulting from the change in the acoustic impedance at the interface of the material layers or defects. The results described in this paper indicate that the novel acoustic technology can expand the scope of SAM as a versatile diagnostic tool requiring less time and having a high efficiency.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.97-99
• /
• 2022

Recently, the amount of data is increasing with the development of industries and technologies. Research on the processing and transmission of large amounts of data is attracting attention. Therefore, in this paper, compressed sensing was used to reduce the amount of data and its effect on Mask R-CNN algorithm was analyzed. We confirmed that as the compressed sensing rate increases, the amount of data in the image and the resolution decreases. However, it was confirmed that there was no significant degradation in the performance of object detection.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2022.11a
• /
• pp.47-48
• /
• 2022

The positioning technology that measures the position of a person or object is a key technology to deal with the location of the real coordinate system or converge the real and virtual worlds, such as digital twins, augmented reality, virtual reality, and autonomous driving. In estimating the location of a person or object at an indoor construction site, there are restrictions that it is impossible to receive location information from the outside, the communication infrastructure is insufficient, and it is difficult to install additional devices. Therefore, this study tested the direct sparse odometry algorithm, one of the visual Simultaneous Localization and Mapping (vSLAM) that estimate the current location and surrounding map using only image information, at an indoor construction site and analyzed its applicability as an indoor positioning technology. As a result, it was found that it is possible to properly estimate the surrounding map and the current location even in the indoor construction site, which has relatively few feature points. The results of this study can be used as reference data for researchers related to indoor positioning technology for construction sites in the future.

• Smart Structures and Systems
• /
• v.30 no.5
• /
• pp.501-511
• /
• 2022

Three-dimensional (3D) models have become crucial for improving civil infrastructure analysis, and they can be used for various purposes such as damage detection, risk estimation, resolving potential safety issues, alarm detection, and structural health monitoring. 3D point cloud data is used not only to make visual models but also to analyze the states of structures and to monitor them using semantic data. This study proposes automating the generation of high-quality 3D point cloud data and removing noise using deep learning algorithms. In this study, large-format aerial images of civilian infrastructure, such as cut slopes and dams, which were captured by drones, were used to develop a workflow for automatically generating a 3D point cloud model. Through image cropping, downscaling/upscaling, semantic segmentation, generation of segmentation masks, and implementation of region extraction algorithms, the generation of the point cloud was automated. Compared with the method wherein the point cloud model is generated from raw images, our method could effectively improve the quality of the model, remove noise, and reduce the processing time. The results showed that the size of the 3D point cloud model created using the proposed method was significantly reduced; the number of points was reduced by 20-50%, and distant points were recognized as noise. This method can be applied to the automatic generation of high-quality 3D point cloud models of civil infrastructures using aerial imagery.

• Nonlinear Functional Analysis and Applications
• /
• v.29 no.1
• /
• pp.131-164
• /
• 2024

In this research, we study a modified relaxed Tseng method with a single projection approach for solving common solution to a fixed point problem involving finite family of τ-demimetric operators and a quasi-monotone variational inequalities in real Hilbert spaces with alternating inertial extrapolation steps and adaptive non-monotonic step sizes. Under some appropriate conditions that are imposed on the parameters, the weak and linear convergence results of the proposed iterative scheme are established. Furthermore, we present some numerical examples and application of our proposed methods in comparison with other existing iterative methods. In order to show the practical applicability of our method to real word problems, we show that our algorithm has better restoration efficiency than many well known methods in image restoration problem. Our proposed iterative method generalizes and extends many existing methods in the literature.

• Journal of Advanced Navigation Technology
• /
• v.22 no.1
• /
• pp.20-26
• /
• 2018

In this paper, we propose a moving object detection and tracking algorithm based on multi-frame feature point tracking information to reduce false positives. However, there are problems of detection error and tracking speed in existing studies. In order to compensate for this, we first calculate the corner feature points and the optical flow of multiple frames for camera movement compensation and object tracking. Next, the tracking error of the optical flow is reduced by the multi-frame forward-backward tracking, and the traced feature points are divided into the background and the moving object candidate based on homography and RANSAC algorithm for camera movement compensation. Among the transformed corner feature points, the outlier points removed by the RANSAC are clustered and the outlier cluster of a certain size is classified as the moving object candidate. Objects classified as moving object candidates are tracked according to label tracking based data association analysis. In this paper, we prove that the proposed algorithm improves both precision and recall compared with existing algorithms by using quadrotor image - based detection and tracking performance experiments.

• Journal of the Society of Disaster Information
• /
• v.20 no.1
• /
• pp.199-205
• /
• 2024

Purpose: This study aims to improve the recognition rate of Auto People Counting (APC) in accurately identifying and providing information on remaining evacuees in disaster-vulnerable facilities such as nursing homes to firefighting and other response agencies in the event of a disaster. Methods: In this study, a baseline model was established using CNN (Convolutional Neural Network) models to improve the algorithm for recognizing images of incoming and outgoing individuals through cameras installed in actual disaster-vulnerable facilities operating APC systems. Various algorithms were analyzed, and the top seven candidates were selected. The research was conducted by utilizing transfer learning models to select the optimal algorithm with the best performance. Results: Experiment results confirmed the precision and recall of Densenet201 and Resnet152v2 models, which exhibited the best performance in terms of time and accuracy. It was observed that both models demonstrated 100% accuracy for all labels, with Densenet201 model showing superior performance. Conclusion: The optimal algorithm applicable to APC among various artificial intelligence algorithms was selected. Further research on algorithm analysis and learning is required to accurately identify the incoming and outgoing individuals in disaster-vulnerable facilities in various disaster situations such as emergencies in the future.

• Journal of the Institute of Convergence Signal Processing
• /
• v.3 no.3
• /
• pp.47-53
• /
• 2002

In this paper, we proposed a content-based video search method using the eigen value of key frame and intensity component. We divided the video stream into shot units to extract key frame representing each shot, and get the intensity distribution of the shot from the database generated by using ECA(Eigen Component Analysis). The generated codebook, their index value for each key frame, and the intensity values were used for database. The query image is utilized to find video stream that has the most similar frame by using the euclidean distance measure among the codewords in the codebook. The experimental results showed that the proposed algorithm is superior to any other methols in the search outcome since it makes use of eigen value and intensity elements, and reduces the processing time etc.

• Journal of Korea Multimedia Society
• /
• v.17 no.3
• /
• pp.342-352
• /
• 2014

This paper proposes an automatic segmentation method of cortical bone and trabecular bone and describes an implementation of structural analysis method of trabecular bone in micro-CT images. The proposed segmentation method extract bone region with binarization using a threshold value. Next, it finds adjacent contour lines from outer boundary line into inward direction and sets candidate regions of cortical bone. Next it remove cortical bone region by finding the candidate cortical region of which the average pixel value is maximum. We implemented the method which computes four structural indicators BV/TV, Tb.Th, Tb.Sp, Tb.N by using VTK(Visualization ToolKit) and sphere fitting algorithm. We applied the implemented method to twenty proximal femur of mouses and compared with the manual segmentation method. Experimental result shows that the average error rates between the proposed segmentation method and the manual segmentation method are less than 3% for the four structural indicatiors. This result means that the proposed method can be used instead of the combersome and time consuming manual segmentation method.