• Journal of Information Display
• /
• v.8 no.4
• /
• pp.27-31
• /
• 2007

In this paper, we present an efficient tiled stereo display system for tangible meeting. For tangible meeting, it is important to provide immersive display with high resolution image to cover up the field of view and provide to the local user the same environment as that of remote site. To achieve these, a high resolution image needs to be transmitted for reconstruction of remote world, and it should be displayed using a tiled display. However, it is hard to transmit high resolution image in real time due to the limit of network bandwidth, and so we receive multiple images and reconstruct a remote world with received images in advance. Then, we update only a specific area where remote user exists by receiving low resolution image in realtime. We synthesize the transmitted image to the existing environmental map of remote world and display it as a stereo image. For this, we developed a new system which supports GPU based real time warping and blending, automatic feature extraction using machine vision technique.

• Journal of Korea Society of Industrial Information Systems
• /
• v.23 no.2
• /
• pp.1-9
• /
• 2018

In this paper, we propose a face recognition method which is not sensitive to illumination change and prevents false recognition of photographs. The proposed method uses infrared and depth images at the same time, solves sensitivity of illumination change by infrared image, and prevents false recognition of two - dimensional image such as photograph by depth image. Face detection method using infrared and depth images simultaneously and feature extraction and matching method for face recognition are realized. Simulation results show that accuracy of face recognition is increased compared to conventional methods.

• Smart Structures and Systems
• /
• v.6 no.3
• /
• pp.335-347
• /
• 2010

Recent advances in hardware and instrumentation technology have allowed the possibility of deploying very large sensor arrays on structures. Exploiting the huge amount of data that can result in order to perform vibration-based structural health monitoring (SHM) is not a trivial task and requires research into a number of specific problems. In terms of pressing problems of interest, this paper discusses: the design and optimisation of appropriate sensor networks, efficient data reduction techniques, efficient and automated feature extraction methods, reliable methods to deal with environmental and operational variability, efficient training of machine learning techniques and multi-scale approaches for dealing with very local damage. The paper is a result of the ESF-S3T Eurocores project "Smart Sensing For Structural Health Monitoring" (S3HM) in which a consortium of academic partners from across Europe are attempting to address issues in the design of automated vibration-based SHM systems for structures.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.20 no.5
• /
• pp.688-693
• /
• 2010

Vision based object recognition in mobile robots has many issues for image analysis problems with neighboring elements in dynamic environments. SURF(Speeded Up Robust Features) is the local feature extraction method of the image and its performance is constant even if disturbances, such as lighting, scale change and rotation, exist. However, it has a difficulty of real-time processing caused by representation of high dimensional vectors. To solve th problem, execution of SURF in GPU(Graphics Processing Unit) is proposed and implemented using CUDA of NVIDIA. Comparisons of recognition rates and processing time for SURF between CPU and GPU by variation of robot velocity and image sizes is experimented.

• ETRI Journal
• /
• v.36 no.6
• /
• pp.913-923
• /
• 2014

This paper proposes a global mapping algorithm for multiple robots from an omnidirectional-vision simultaneous localization and mapping (SLAM) approach based on an object extraction method using Lucas-Kanade optical flow motion detection and images obtained through fisheye lenses mounted on robots. The multi-robot mapping algorithm draws a global map by using map data obtained from all of the individual robots. Global mapping takes a long time to process because it exchanges map data from individual robots while searching all areas. An omnidirectional image sensor has many advantages for object detection and mapping because it can measure all information around a robot simultaneously. The process calculations of the correction algorithm are improved over existing methods by correcting only the object's feature points. The proposed algorithm has two steps: first, a local map is created based on an omnidirectional-vision SLAM approach for individual robots. Second, a global map is generated by merging individual maps from multiple robots. The reliability of the proposed mapping algorithm is verified through a comparison of maps based on the proposed algorithm and real maps.

• The KIPS Transactions:PartB
• /
• v.12B no.4 s.100
• /
• pp.403-412
• /
• 2005

In this paper, we propose an effective method for detecting and classifying shot transitions in video sequences. The proposed method detects and classifies shot transitions including cuts, fades and dissolves by compensating camera operations in video sequences, so that our method prevents false positives resulting from camera operations. Also, our method eliminates local moving objects in the process of compensating camera operations, so that our method prevents errors resulting from moving objects. In the experiments, we show that our shot transition approach can work as a promising solution by comparing the proposed method with previously known methods in terms of performance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.12
• /
• pp.6000-6017
• /
• 2018

This paper presents a new descriptor, named Histograms of Prominent Edge Directions (HPED), for the recognition of facial expressions in a person-independent environment. In this paper, we raise the issue of sampling error in generating the code-histogram from spatial regions of the face image, as observed in the existing descriptors. HPED describes facial appearance changes based on the statistical distribution of the top two prominent edge directions (i.e., primary and secondary direction) captured over small spatial regions of the face. Compared to existing descriptors, HPED uses a smaller number of code-bins to describe the spatial regions, which helps avoid sampling error despite having fewer samples while preserving the valuable spatial information. In contrast to the existing Histogram of Oriented Gradients (HOG) that uses the histogram of the primary edge direction (i.e., gradient orientation) only, we additionally consider the histogram of the secondary edge direction, which provides more meaningful shape information related to the local texture. Experiments on popular facial expression datasets demonstrate the superior performance of the proposed HPED against existing descriptors in a person-independent environment.

• International Journal of Computer Science & Network Security
• /
• v.21 no.8
• /
• pp.127-136
• /
• 2021

Real-time human recognition is a challenging task, as the images are captured in an unconstrained environment with different poses, makeups, and styles. This limitation is addressed by generating several facial images with poses, makeup, and styles with a single reference image of a person using Generative Adversarial Networks (GAN). In this paper, we propose deep learning-based human recognition using integration of GAN and Spatial Domain Techniques. A novel concept of human recognition based on face depiction approach by generating several dissimilar face images from single reference face image using Domain Transfer Generative Adversarial Networks (DT-GAN) combined with feature extraction techniques such as Local Binary Pattern (LBP) and Histogram is deliberated. The Euclidean Distance (ED) is used in the matching section for comparison of features to test the performance of the method. A database of millions of people with a single reference face image per person, instead of multiple reference face images, is created and saved on the centralized server, which helps to reduce memory load on the centralized server. It is noticed that the recognition accuracy is 100% for smaller size datasets and a little less accuracy for larger size datasets and also, results are compared with present methods to show the superiority of proposed method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.4
• /
• pp.1448-1463
• /
• 2021

With the continuous development of cloud storage, plenty of redundant data exists in cloud storage, especially multimedia data such as images and videos. Data deduplication is a data reduction technology that significantly reduces storage requirements and increases bandwidth efficiency. To ensure data security, users typically encrypt data before uploading it. However, there is a contradiction between data encryption and deduplication. Existing deduplication methods for regular files cannot be applied to image deduplication because images need to be detected based on visual content. In this paper, we propose a secure image deduplication scheme based on hashing and clustering, which combines a novel perceptual hash algorithm based on Local Binary Pattern. In this scheme, the hash value of the image is used as the fingerprint to perform deduplication, and the image is transmitted in an encrypted form. Images are clustered to reduce the time complexity of deduplication. The proposed scheme can ensure the security of images and improve deduplication accuracy. The comparison with other image deduplication schemes demonstrates that our scheme has somewhat better performance.

• ETRI Journal
• /
• v.44 no.6
• /
• pp.1004-1019
• /
• 2022

For autonomous driving without high-definition maps, we present a model capable of generating multiple plausible paths from egocentric images for autonomous vehicles. Our generative model comprises two neural networks: feature extraction network (FEN) and path generation network (PGN). The FEN extracts meaningful features from an egocentric image, whereas the PGN generates multiple paths from the features, given a driving intention and speed. To ensure that the paths generated are plausible and consistent with the intention, we introduce an attentive discriminator and train it with the PGN under a generative adversarial network framework. Furthermore, we devise an interaction model between the positions in the paths and the intentions hidden in the positions and design a novel PGN architecture that reflects the interaction model for improving the accuracy and diversity of the generated paths. Finally, we introduce ETRIDriving, a dataset for autonomous driving, in which the recorded sensor data are labeled with discrete high-level driving actions, and demonstrate the state-of-the-art performance of the proposed model on ETRIDriving in terms of accuracy and diversity.