• Journal of Digital Convergence
• /
• v.19 no.5
• /
• pp.307-313
• /
• 2021

In this paper, a lightweight network with fewer parameters compared to the existing object detection method is proposed. In the case of the currently used detection model, the network complexity has been greatly increased to improve accuracy. Therefore, the proposed network uses EfficientNet as a feature extraction network, and the subsequent layers are formed in a pyramid structure to utilize low-level detailed features and high-level semantic features. An attention process was applied between pyramid structures to suppress unnecessary noise for prediction. All computational processes of the network are replaced by depth-wise and point-wise convolutions to minimize the amount of computation. The proposed network was trained and evaluated using the PASCAL VOC dataset. The features fused through the experiment showed robust properties for various objects through a refinement process. Compared with the CNN-based detection model, detection accuracy is improved with a small amount of computation. It is considered necessary to adjust the anchor ratio according to the size of the object as a future study.

• International Journal of Precision Engineering and Manufacturing
• /
• v.5 no.1
• /
• pp.27-35
• /
• 2004

In the field of machine vision using a single camera mounted on a mobile robot, although the detection and tracking of moving objects from a moving observer, is complex and computationally demanding task. In this paper, we propose a new scheme for a mobile robot to track and capture a moving object using images of a camera. The system consists of the following modules: data acquisition, feature extraction and visual tracking, and trajectory generation. And a single camera is used as visual sensors to capture image sequences of a moving object. The moving object is assumed to be a point-object and projected onto an image plane to form a geometrical constraint equation that provides position data of the object based on the kinematics of the active camera. Uncertainties in the position estimation caused by the point-object assumption are compensated using the Kalman filter. To generate the shortest time trajectory to capture the moving object, the linear and angular velocities are estimated and utilized. The experimental results of tracking and capturing of the target object with the mobile robot are presented.

• Journal of the Korean Society for Precision Engineering
• /
• v.20 no.6
• /
• pp.20-26
• /
• 2003

This study proposes the construction of attractor quadrant method for high-precision cutting characteristics evaluation of non-ferrous metals. Also this paper aims to find the optimal cutting conditions of diamond turning machine by measuring surface form and roughness to perform the cutting experiment of non-ferrous metals, which are aluminum, with diamond tool. As well, according to change cutting conditions such as feed rate, using diamond turning machine to Perform cutting Processing, by measuring cutting force and surface roughness and according to cutting conditions the aluminum about cutting properties. Trajectory changes in the attractor indicated a substantial difference in fractal characteristics and attractor quadrant characteristics. In quantitative quadrant feature extraction, 1,309 point in the case of A17075 (one quadrant) and 1,406 point (one quadrant) in the case of brass were proposed on the basis of attractor reconstruction. Proposed attractor quadrant method can be used for high-precision cutting characteristics evaluation of non-ferrous metals.

• Journal of Korea Multimedia Society
• /
• v.17 no.9
• /
• pp.1041-1051
• /
• 2014

In this paper, we propose a new image alignment technique based on CUDA SURF in order to solve the initial image alignment problem that frequently occurs in machine vision applications. Machine vision systems using multi-spectral images have recently become more common for solving various decision problems that cannot be performed by the human vision system. These machine vision systems mostly use markers for the initial image alignment. However, there are some applications where the markers cannot be used and the alignment techniques have to be changed whenever their markers are changed. In order to solve these problems, we propose a new image alignment method for multi-spectral machine vision applications based on SURF extracting image features without depending on markers. In this paper, we propose an image alignment method that obtains a sufficient number of feature points from multi-spectral images using SURF and removes outlier iteratively based on a least squares method. We further propose an effective preliminary scheme for removing mismatched feature point pairs that may affect the overall performance of the alignment. In addition, we reduce the execution time by implementing the proposed method using CUDA based on GPGPU in order to guarantee real-time operation. Simulation results show that the proposed method is able to align images effectively in applications where markers cannot be used.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.9
• /
• pp.1-10
• /
• 2013

In this paper, we implement the SIFT(Scale-Invariant Feature Transform) algorithm for feature point extraction using a many-core processor, and analyze the performance, area efficiency, and system area efficiency of the many-core processor. In addition, we demonstrate the potential of the proposed many-core processor by comparing the performance of the many-core processor with that of high-performance CPU and GPU(Graphics Processing Unit). Experimental results indicate that the accuracy result of the SIFT algorithm using the many-core processor was same as that of OpenCV. In addition, the many-core processor outperforms CPU and GPU in terms of execution time. Moreover, this paper proposed an optimal model of the SIFT algorithm on the many-core processor by analyzing energy efficiency and area efficiency for different octave sizes.

• The Journal of the Korea Contents Association
• /
• v.6 no.2
• /
• pp.44-50
• /
• 2006

This study has proposed a method of content-based image retrieval using object region in order to overcome disadvantages of existing color histogram methods. The existing color histogram methods have a weak point of reducing accuracy, because these have both a quantization error and an absence of spatial information. In order to overcome this problem, we convert a color information to a HSV space, quantize hue factor being pure color information, and calculate histogram. And then we use hue for retrieval feature that is robust in brightness, movement, and rotation. To solve the problem of the absence of spatial information, we select object region in terms of color feature and region correlation. And we use both the edge and the DC in the selected region for retrieving. As a result of experiment with 1,000 natural color images, the proposed method shows better precision and recall than the existing methods.

• The KIPS Transactions:PartB
• /
• v.12B no.5 s.101
• /
• pp.521-526
• /
• 2005

In this Paper, we propose a keyword spotting system as an alternative to searching system for poor quality Korean document images and compare the Proposed system with an OCR-based document retrieval system. The system is composed of character segmentation, feature extraction for the query keyword, and word-to-word matching. In the character segmentation step, we propose an effective method to remove the connectivity between adjacent characters and a character segmentation method by making the variance of character widths minimum. In the query creation step, feature vector for the query is constructed by a combination of a character model by typeface. In the matching step, word-to-word matching is applied base on a character-to-character matching. We demonstrated that the proposed keyword spotting system is more efficient than the OCR-based one to search a keyword on the Korean document images, especially when the quality of documents is quite poor and point size is small.

• Journal of Korea Multimedia Society
• /
• v.16 no.1
• /
• pp.108-118
• /
• 2013

Recently the 3D modeling techniques were developed rapidly due to rapid development of computer vision, computer graphics with the excellent performance of hardware. With the advent of a variety of 3D contents, 3D modeling technology becomes more in demand and it's quality is increased. 3D face models can be applied widely to such contents with high usability. In this paper, a 3D face modeling is attempted from a given single 2D frontal face image. To achieve the goal, we thereafter the feature points using AAM are extracted from the input frontal face image. With the extracted feature points we deform the 3D general model by 2-pass mesh warping, and also the depth extraction based on intensity values is attempted to. Throughout those processes, a universal 3D face modeling method with less expense and less restrictions to application environment was implemented and it's validity was shown through experiments.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.2
• /
• pp.427-436
• /
• 2016

In general, QRS duration represent a distance of Q start and S end point. However, since criteria of QRS duration are vague and Q, S point is not detected accurately, arrhythmia classification performance can be reduced. In this paper, we propose extraction of Q, S start and end point RS feature based on phase transition tracking method after we detected R wave that is large peak of electrocardiogram(ECG) signal. For this purpose, we detected R wave, from noise-free ECG signal through the preprocessing method. Also, we classified QRS pattern through differentiation value of ECG signal and extracted Q, S start and end point by tracking direction and count of phase based on R wave. The performance of R wave detection is evaluated by using 48 record of MIT-BIH arrhythmia database. The achieved scores indicate the average detection rate of 99.60%. PVC classification is evaluated by using 9 record of MIT-BIH arrhythmia database that included over 30 premature ventricular contraction(PVC). The achieved scores indicate the average detection rate of 94.12% in PVC.

• Analytical Science and Technology
• /
• v.25 no.5
• /
• pp.292-299
• /
• 2012

GC/MS has been utilized for many applications due to great resolution and reproducibility, which made it possible to build up the database of mass spectrum, while HS-SPME has the advantage of solventfree extraction of volatile compounds. The combination of these two methods, HS-SPME GC/MS, enabled many scientific applications with various possibilities. In this study, the analysis of trail pheromone excreted from live Camponotus japonicus with the feature of solvent-free extraction was carried out and the optimization for this analysis was performed. The major compounds detected were n-decane, n-undecane, and n-tridecane. Optimization for the best detection of these hydrocarbons was processed in the point of SPME parameter (selection of fiber, extraction temperature, extraction time, etc.). The advantage of the analysis of live sample is to analyze phenomenon right after it is excreted by ants. But the experimental process has restriction of extraction temperature and time because of the analysis of live ants. Establishing the process of HS-SPME GC/MS applied to live samples shown in this study can be a breakthrough for the ecofriendly and ethical research of live things.