• Journal of the Institute of Convergence Signal Processing
• /
• v.2 no.1
• /
• pp.53-62
• /
• 2001

In this paper, the hybrid function by local scale control is proposed to detect the optimal edges from blurred images. In the case of image capturing, some blurring is occurred by the characteristics of the illumination and the reflected light. During processing the blurred image, it is difficult to detect perfect edges. This algorithm proposed a new hybrid function which is merged Gaussian function and the second derivative of Gaussian function. And it detects the optimal edges applying directional edge detection by Canny algorithm as the scale factor of $\sigma$ in the given local mask has been changed after convolving the hybrid function for input image. In the result, the performance is confirmed that this algorithm is better than Sobel, Robert and Canny edge detector by analyzing the some test images. And the results is obtained 0.2 ㏈ ~ 14 ㏈ of PSNR than those conventional method.

• Journal of Information Processing Systems
• /
• v.17 no.5
• /
• pp.947-959
• /
• 2021

Aiming at the problem of low accuracy in the water boundary automatic extraction of islands from GF-2 remote sensing image with high resolution in three bands, new water edges automatic extraction method in island based on GF-2 remote sensing images, genetic algorithm (GA) method, is proposed in this paper. Firstly, the GA-OTSU threshold segmentation algorithm based on the combination of GA and the maximal inter-class variance method (OTSU) was used to segment the island in GF-2 remote sensing image after pre-processing. Then, the morphological closed operation was used to fill in the holes in the segmented binary image, and the boundary was extracted by the Sobel edge detection operator to obtain the water edge. The experimental results showed that the proposed method was better than the contrast methods in both the segmentation performance and the accuracy of water boundary extraction in island from GF-2 remote sensing images.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.9
• /
• pp.4572-4586
• /
• 2019

Mobile Edge Computing (MEC) and Information-Centric Networking (ICN) are essential network architectures for the future Internet. The advantages of MEC and ICN such as computation and storage capabilities at the edge of the network, in-network caching and named-data communication paradigm can greatly improve the quality of video streaming applications. However, the packet loss in wireless network environments still affects the video streaming performance and the existing loss recovery approaches in ICN does not exploit the capabilities of MEC. This paper proposes a Deep Learning based Loss Recovery Mechanism (DL-LRM) for video streaming over MEC based ICN. Different with existing approaches, the Forward Error Correction (FEC) packets are generated at the edge of the network, which dramatically reduces the workload of core network and backhaul. By monitoring network states, our proposed DL-LRM controls the FEC request rate by deep reinforcement learning algorithm. Considering the characteristics of video streaming and MEC, in this paper we develop content caching detection and fast retransmission algorithm to effectively utilize resources of MEC. Experimental results demonstrate that the DL-LRM is able to adaptively adjust and control the FEC request rate and achieve better video quality than the existing approaches.

• Transactions of Materials Processing
• /
• v.18 no.5
• /
• pp.377-384
• /
• 2009

The shape of edge cracking in rolling process generally occurred "V" shape. This cracking is successively generated at width edge of strip. The edge cracking is developed to center of strip during rolling process. In the results, the strip is occurred fracture, and the productivity is gone down because of the extensive production time. Accordingly, we need to control crack propagation during rolling process. But, the control of cracking is very difficult in rolling process. Previously the studies of edge cracking were mainly performed on hot rolling process. In this paper, the shape of the edge cracking in rolling was estimated according to process conditions such as initial edge crack size, reduction ratio and tension using FE-simulation and the simplicity experiments on cold rolling process.

• Transactions of Materials Processing
• /
• v.9 no.1
• /
• pp.66-71
• /
• 2000

The present study tackles the metallurgical subjects involving the thin slab-direct hot rolling process, i.e. mini-mill process. In order to clarify the effect of chemical composition of steel and MnS precipitation behaviors on the development of edge cracking during hot rolling, the content of manganese and sulfur in low carbon steel was varied and the isothermal treatment prior to roughing was applied. Edge cracking during roughing in the hot-rolling process of mini-mill was effectively prevented by means of the isothermal treatment at 115$0^{\circ}C$ for 5 minutes in the 0.4% manganese steel containing sulfur lower than 0.013%. With the increase in manganese content in low carbon steel, coarser MnS developed. The edge cracking index which denotes the total length of edge crack per unit edge-length of rolled specimens was proposed in this paper. It was found that the edge cracking index linearly decreased with the increase in the ratio of MnS.

• Proceedings of the KIEE Conference
• /
• 1998.11b
• /
• pp.569-571
• /
• 1998

In this paper, we propose a method of restoring blurred images by an edge-sensitive adaptive filter. The direction of the edge is estimated using the properties of 2-D block FFT. Reduction of blurring due to the added noise during image transfer and the focus of lens caused by shooting a fast moving object is very important. To remove this phenomenon effectively, we can use the edge information obtained by processing the blurred images. The proposed algorithm estimates both the existence and the direction of the edge. On the basis of the acquired edge direction information, we choose the appropriate edge-sensitive adaptive filter, which enables us to get better images than images obtained by methods not considering the direction of the edge. The performance of the proposed algorithm is shown in the simulation result.

• Transactions of Materials Processing
• /
• v.28 no.3
• /
• pp.154-158
• /
• 2019

The hole expansion tests using conical punch, flat punch or hemispherical punch are widely used for stretch flangeability verification of HSS. In this study, we investigate the strain distribution on the shear edges of the hole expansion test using grid marking and a projector. A small crack at the edge is distributed, resulting in a large gap between the HER and the crack strain. The strain distribution at the edges is irregular due to anisotropy of sheet metal. While an edge perpendicular to the rolling direction indicate a lower strain level compared to an edge parallel to the rolling direction, edge cracks occur at the edge perpendicular to the rolling direction. To predict the manifestation of edge cracks in FE analysis, the result of the hole expansion test with a crack strain measurement may well be a better tool than FLD. In this case, the level of strain and the direction of the edge relative to the rolling direction should be well considered.

• Proceedings of The KACE
• /
• 2017.08a
• /
• pp.61-62
• /
• 2017

Edge computing 환경은 자원이 제한된 IoT의 디바이스가 증가하면서 데이터가 급등하는 환경을 극복하기 위한 패러다임으로 주목받고 있다. 그러나 자원이 제한된 IoT 디바이스의 특성상 실시간성 데이터의 손실, 지연과 같은 문제가 존재한다. 본 논문에서는 이를 해결하기 위해서 폭발적으로 데이터가 증가하는 Edge computing 환경에 적합한 프레임워크에 대하여 서술한다.

• Journal of Industrial Technology
• /
• v.27 no.B
• /
• pp.97-101
• /
• 2007

We have developed a Edge Position Controller (EPC) using a ultrasonic sensor and applied to a fabric machine as a web guide system. Hardware devices composed of a ultrasonic transmitter-receiver sensor module and microprocessor-based sensor signal processing system are developed to realize the proposed system. We evaluated the control characteristics of the EPC and the performance of the system was good enough to apply the actual system.

• Journal of Information Processing Systems
• /
• v.20 no.3
• /
• pp.375-390
• /
• 2024

Edge computing architecture has effectively alleviated the computing pressure on cloud platforms, reduced network bandwidth consumption, and improved the quality of service for user experience; however, it has also introduced new security issues. Existing anomaly detection methods in big data scenarios with cloud-edge computing collaboration face several challenges, such as sample imbalance, difficulty in dealing with complex network traffic attacks, and difficulty in effectively training large-scale data or overly complex deep-learning network models. A lightweight deep-learning model was proposed to address these challenges. First, normalization on the user side was used to preprocess the traffic data. On the edge side, a trained Wasserstein generative adversarial network (WGAN) was used to supplement the data samples, which effectively alleviates the imbalance issue of a few types of samples while occupying a small amount of edge-computing resources. Finally, a trained lightweight deep learning network model is deployed on the edge side, and the preprocessed and expanded local data are used to fine-tune the trained model. This ensures that the data of each edge node are more consistent with the local characteristics, effectively improving the system's detection ability. In the designed lightweight deep learning network model, two sets of convolutional pooling layers of convolutional neural networks (CNN) were used to extract spatial features. The bidirectional long short-term memory network (BiLSTM) was used to collect time sequence features, and the weight of traffic features was adjusted through the attention mechanism, improving the model's ability to identify abnormal traffic features. The proposed model was experimentally demonstrated using the NSL-KDD, UNSW-NB15, and CIC-ISD2018 datasets. The accuracies of the proposed model on the three datasets were as high as 0.974, 0.925, and 0.953, respectively, showing superior accuracy to other comparative models. The proposed lightweight deep learning network model has good application prospects for anomaly traffic detection in cloud-edge collaborative computing architectures.