• The Journal of the Acoustical Society of Korea
• /
• v.18 no.6
• /
• pp.57-64
• /
• 1999

In this paper, we investigated the robust endpoint detection algorithm in noisy environment. A new feature parameter based on a discrete wavelet transform is proposed for word boundary detection of isolated utterances. The sum of standard deviation of wavelet coefficients in the third coarse and weighted first detailed scale is defined as a new feature parameter for endpoint detection. We then developed a new and robust endpoint detection algorithm using the feature found in the wavelet domain. For the performance evaluation, we evaluated the detection accuracy and the average recognition error rate due to endpoint detection in an HMM-based recognition system across several signal-to-noise ratios and noise conditions.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.05a
• /
• pp.358-361
• /
• 2021

In order to improve the accuracy of small target detection more effectively, this paper proposes an improved single shot detector (SSD) target detection and recognition method based on cspdarknet53, which introduces lightweight ECA attention mechanism and Feature Pyramid Network (FPN). First, the original SSD backbone network is replaced with cspdarknet53 to enhance the learning ability of the network. Then, a lightweight ECA attention mechanism is added to the basic convolution block to optimize the network. Finally, FPN is used to gradually fuse the multi-scale feature maps used for detection in the SSD from the deep to the shallow layers of the network to improve the positioning accuracy and classification accuracy of the network. Experiments show that the proposed target detection algorithm has better detection accuracy, and it improves the detection accuracy especially for small targets.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2023.05a
• /
• pp.251-252
• /
• 2023

As technology improves with the 4th industrial revolution, smart construction is becoming a key part of safety management in the architecture and civil engineering. By using object detection technology with CCTV data, construction sites can be managed efficiently. In this study, super resolution technology based on deep learning is proposed to improve the accuracy of object detection in construction sites. As the resolution of a train set data and test set data get higher, the accuracy of object detection model gets better. Therefore, according to the scale of construction sites, different object detection models can be considered.

• Tunnel and Underground Space
• /
• v.19 no.1
• /
• pp.52-63
• /
• 2009

This is a study on large-scale rock joint roughness measurements using LIDAR (light detection and ranging) and the Split-FX point cloud processing software. The large-scale rock Joint Roughness Coefficient (JRC) is calculated using the maximum amplitude of joint asperities over the profile length on large-scale Joint surfaces of rock. As the profile length increases, JRC decreases due to scale-effects of rock specimens and is non-stationary. Also JRC shows anisotropy depending on the profile direction. The profile direction is measured relative to either dip or strike of the large-scale joint.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39C no.8
• /
• pp.731-737
• /
• 2014

This paper presents abnormal behavior detection in crowd within surveillance video. We have defined below two cases as a abnormal behavior; first as a sporadically spread phenomenon and second as a sudden running in same direction. In order to detect these two abnormal behaviors, we first extract the motion vector and propose a new descriptor which is combined MHOF(Multi-scale Histogram of Optical Flow) and DCHOF(Directional Change Histogram of Optical Flow). Also, binary classifier SVM(Support Vector Machine) is used for detection. The accuracy of the proposed algorithm is evaluated by both UMN and PETS 2009 dataset and comparisons with the state-of-the-art method validate the advantages of our algorithm.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.11
• /
• pp.5624-5638
• /
• 2016

A novel method based on Surf points is proposed to detect and lock-track single ground target in aerial videos. Videos captured by moving cameras contain complex motions, which bring difficulty in moving object detection. Our approach contains three parts: moving target template detection, search area estimation and target tracking. Global motion estimation and compensation are first made by grids-sampling Surf points selecting and matching. And then, the single ground target is detected by joint spatial-temporal information processing. The temporal process is made by calculating difference between compensated reference and current image and the spatial process is implementing morphological operations and adaptive binarization. The second part improves KALMAN filter with surf points scale information to predict target position and search area adaptively. Lastly, the local Surf points of target template are matched in this search region to realize target tracking. The long-term tracking is updated following target scaling, occlusion and large deformation. Experimental results show that the algorithm can correctly detect small moving target in dynamic scenes with complex motions. It is robust to vehicle dithering and target scale changing, rotation, especially partial occlusion or temporal complete occlusion. Comparing with traditional algorithms, our method enables real time operation, processing $520{\times}390$ frames at around 15fps.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.1
• /
• pp.828-831
• /
• 2005

In this paper we propose a method to detect human faces from color image using pupil-template matching. Face detection is done by three stages. (i)separating skin regions from non-skin regions; (ii)generating a face regions by application of the best-fit ellipse; (iii)detecting face by pupil-template. Detecting skin regions is based on a skin color model. we generate a gray scale image from original image by the skin model. The gray scale image is segmented to separated skin regions from non-skin regions. Face region is generated by application of the best-fit ellipse is computed on the base of moments. Generated face regions are matched by pupil-template. And we detection face.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.19 no.2
• /
• pp.73-81
• /
• 2024

Recently, mobile manipulators have been utilized in agriculture industry for weed removal and harvest automation. This paper proposes a domain adaptive fruit detection method for harvest automation, by utilizing OWL-ViT model which is an open-vocabulary object detection model. The vision-language model can detect objects based on text prompt, and therefore, it can be extended to detect objects of undefined categories. In the development of deep learning models for real-world problems, constructing a large-scale labeled dataset is a time-consuming task and heavily relies on human effort. To reduce the labor-intensive workload, we utilized a large-scale public dataset as a source domain data and employed a domain adaptation method. Adversarial learning was conducted between a domain discriminator and feature extractor to reduce the gap between the distribution of feature vectors from the source domain and our target domain data. We collected a target domain dataset in a real-like environment and conducted experiments to demonstrate the effectiveness of the proposed method. In experiments, the domain adaptation method improved the AP50 metric from 38.88% to 78.59% for detecting objects within the range of 2m, and we achieved 81.7% of manipulation success rate.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.3
• /
• pp.127-136
• /
• 2024

This study explores the correlation between system anomalies and large-scale logs within the Spark cluster environment. While research on anomaly detection using logs is growing, there remains a limitation in adequately leveraging logs from various components of the cluster and considering the relationship between anomalies and the system. Therefore, this paper analyzes the distribution of normal and abnormal logs and explores the potential for anomaly detection based on the occurrence of log templates. By employing Hadoop and Spark, normal and abnormal log data are generated, and through t-SNE and K-means clustering, templates of abnormal logs in anomalous situations are identified to comprehend anomalies. Ultimately, unique log templates occurring only during abnormal situations are identified, thereby presenting the potential for anomaly detection.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.10 no.1
• /
• pp.38-45
• /
• 2017

Attacking of computer and network is increasing as information processing technology heavily depends on computer and network. To prevent the attack of system and network, host and network based intrusion detection system has developed. But previous rule based system has a lot of difficulties. For this reason demand for developing a intrusion detection system which detects and cope with the attack of system and network resource in real time. In this paper we develop a real time intrusion detection system which is combination of APEX and LS-SVM classifier. Proposed system is for nonlinear data and guarantees convergence. While real time processing system has its advantages, such as memory efficiency and allowing a new training data, it also has its disadvantages of inaccuracy compared to batch way. Therefore proposed real time intrusion detection system shows similar performance in accuracy compared to batch way intrusion detection system, it can be deployed on a commercial scale.