• Journal of the Korea Society of Computer and Information
• /
• v.21 no.12
• /
• pp.125-130
• /
• 2016

Automatic quality inspection by computer vision can be applied and give a solution to the pharmaceutical industry field. Pharmaceutical capsule can be easily affected by flaws like dents, cracks, holes, etc. In order to solve the quality inspection problem, it is required computationally efficient image processing technique like thresholding, boundary edge detection and segmentation and some automated systems are available but they are very expensive to use. In this paper, we have developed a dented capsule image processing technique using edge-based image segmentation, TLS(Total Least Squares) curve fitting technique and adopted low cost camera module for capsule image capturing. We have tested and evaluated the accuracy, training and testing time of the classification recognition algorithms like PCA(Principal Component Analysis), ICA(Independent Component Analysis) and SVM(Support Vector Machine) to show the performance. With the result, PCA, ICA has low accuracy, but SVM has good accuracy to use for classifying the dented capsule.

• Journal of Veterinary Clinics
• /
• v.23 no.4
• /
• pp.416-420
• /
• 2006

Compare to testing sera individually, pooled-serum testing has considered as a cost-effective method, particularly on a large population-based seroprevalence studies. This study was to determine the relationship between individual sera and pooled sera titers for detection of avian infectious bronchitis virus (IBV) and to evaluate suitability of pooled sera by comparing prevalences estimated from both samples. A total of 5,000 individual samples were collected from 500 flocks in Chungcheong, Gyunsgi, and Kangwon provinces between January 2005 and February 2006. Ten samples were randomly selected from each flock. Five-hundred pooled sera were prepared by mixing equal amount of each 10 individual serum from the original samples. IBV antibody titers were measured by hemagglutination inhibition (HI) test. The least squares regression analysis was performed to construct equation between pooled and mean individual titers. To determine whether the flock is infected 4 arbitrary criteria were used: detection of at least 1 chicken with HI titer ${\ge}$ 9 (criterion 1), detection of at least 2 samples with HI titer ${\ge}$9 (criterion 2), detection of at least 1 sample with HI titer ${\ge}$ 10 (criterion 3), and filially detection of at least 1 sample with HI titer ${\ge}$ 11 (criterion 4). The receiver operating characteristic (ROC) curve was used to examine the cut-off points of pooled titers showing optimal diagnostic accuracy. The area under the curve (AUC), sensitivities (Se), specificities (Sp), and positive (PPV) and negative (NPV) predictive values were calculated. The regression equation between pooled titers (pool) and mean individual titers (mean) was: $pool= 1.2498+0.8952{\times}mean$, with coefficient of determination of 87% (p< 0.0001). The optimal cut-off points of pooled titers were titer 8 for criterion 1 (AUC=0.975, Se=0.883, Sp=0.959, PPV=0.985, NPV=0.728), titer 8 for criterion 2 (AUC=0.969, Se=0.954, Sp=0.855, PPV=0.926, NPV=0.907), titer 9 for criterion 3 (AUC=0.970, Se=0.836, Sp=0.967, PPV=0.978, NPV=0.772), and titer 9 for criterion 4 (AUC= 0.946, Se=0.928, Sp=0.843, PPV=0.857, NPV=0.921). The difference of 'prevalence estimated by individual and pooled sample showed a minimum of 2% for criteria 2 and a maximum of 9.1:% for criteria 3. These results indicate that the use of pooled sera in HI test for screening IBV infection in laying hen flocks is considered as a cost-effective method of testing large numbers of samples with high diagnostic accuracy.

• Smart Media Journal
• /
• v.6 no.4
• /
• pp.32-40
• /
• 2017

In this paper, a robust text detection method based on edge enhanced contrasting extremal region (CER) is proposed using stroke width transform (SWT) and tensor voting. First, the edge enhanced CER extracts a number of covariant regions, which is a stable connected component from input images. Next, SWT is created by the distance map, which is used to eliminate non-text regions. Then, these candidate text regions are verified based on tensor voting, which uses the input center point in the previous step to compute curve salience values. Finally, the connected component grouping is applied to a cluster closed to characters. The proposed method is evaluated with the ICDAR2003 and ICDAR2013 text detection competition datasets and the experiment results show high accuracy compared to previous methods.

• Proceedings of the IEEK Conference
• /
• 2009.05a
• /
• pp.34-36
• /
• 2009

This paper proposes an effective lane detection algorithm using hugh transform and lane model. The proposed lane detection algorithm includes two major components, i.e., lane marks segmentation and an exact lane extraction using a novel postprocessing technique. The first step is to segment lane marks from background images using HSV color model. Then, a novel postprocessing is used to detect an exact lane using Hugh transform and lane models(linear and curved lane models). The postprocessing consists of three parts, i.e, thinning process, Hugh Transform and filtering process. We divide input image into three regions of interests(ROIs). Based on lane curve function(LCF), we can detect an exact lane from various extracted lane lines. The lane models(linear and curved lane mode]) are used in order to judge whether each lane segment is fit or not in each ROIs. Experimental results show that the proposed scheme is very effective in lane detection.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.59 no.2
• /
• pp.429-435
• /
• 2010

Three dimensional (3-D) object detection and pose estimation from a single view query image has been an important issue in various fields such as medical applications, robot vision, and manufacturing automation. However, most of the existing methods are not appropriate in a real time environment since object detection and pose estimation requires extensive information and computation. In this paper, we present a fast 3-D object detection and pose estimation scheme based on surrounding camera view-changed images of objects. Our scheme has two parts. First, we detect images similar to the query image from the database based on the shape feature, and calculate candidate poses. Second, we perform accurate pose estimation for the candidate poses using the scale invariant feature transform (SIFT) method. We earned out extensive experiments on our prototype system and achieved excellent performance, and we report some of the results.

• Journal of Broadcast Engineering
• /
• v.17 no.6
• /
• pp.1069-1079
• /
• 2012

Lane detection with the use of a camera is an essential task required for the development of advanced driving assistance system. In this paper, edges of the lane are generated by applying Canny's method. The edge detection usually makes different results for several environmental conditions depending on the clearness of lane quality, so that it sometimes causes wrong lane detection. Therefore, we propose robust algorithm to environmental changes that automatically adjusts parameter for edge detection and generates edges more stably. Based on the acquired edges, we finally generate the spline curve of lane by using Catmull Rom spline.

• Journal of Institute of Control, Robotics and Systems
• /
• v.22 no.1
• /
• pp.1-7
• /
• 2016

ADAS (Advanced Driver Assistance Systems) requires not only real-time robust lane detection, both straight and curved, but also predicting upcoming steering direction by detecting the curvature of lanes. In this paper, a curvature lane detection algorithm is proposed to enhance the accuracy and detection rate based on using inverse perspective images and Gaussian Mixture Model (GMM) to segment the lanes from the background under various illumination condition. To increase the speed and accuracy of the lane detection, this paper used template matching, RANSAC and proposed post processing method. Through experiments, it is validated that the proposed algorithm can detect both straight and curved lanes as well as predicting the upcoming direction with 92.95% of detection accuracy and 50fps speed.

• Journal of Nuclear Fuel Cycle and Waste Technology(JNFCWT)
• /
• v.9 no.1
• /
• pp.41-48
• /
• 2011

A mobile laser-induced breakdown detection (LIBD) system was developed for the field measurement of the size and concentration of aquatic colloidal nanoparticles sampled from Korea Atomic Energy Research Institute Underground Research Tunnel (KURT). The established LIBD apparatus is based on the optical detection of a laser-induced plasma by means of a two-dimensional optical imaging method for determining the size of nanoparticle. Calibration curve for determining the size of nanoparticle was obtained from the polystyrene reference particles of a well-defined size. The first direct application was made at KURT for investigating the particle sizes in groundwater. By comparing the size of particles in groundwater with the sizes of reference particles, the mean particle size of approximately $108{\pm}26$ nm with the concentration lower than 50 ppb was determined.

• Journal of Biomedical Engineering Research
• /
• v.38 no.1
• /
• pp.25-31
• /
• 2017

Testing TB in chest X-ray images is a typical method to diagnose presence and magnitude of PTB lesion. However, the method has limitation due to inter-reader variability. Therefore, it is essential to overcome this drawback with automatic interpretation. In this study, we propose a novel method for detection of PTB using SegNet, which is a deep learning architecture for semantic pixel wise image labelling. SegNet is composed of a stack of encoders followed by a corresponding decoder stack which feeds into a soft-max classification layer. We modified parameters of SegNet to change the number of classes from 12 to 2 (TB or none-TB) and applied the architecture to automatically interpret chest radiographs. 552 chest X-ray images, provided by The Korean Institute of Tuberculosis, used for training and test and we constructed a receiver operating characteristic (ROC) curve. As a consequence, the area under the curve (AUC) was 90.4% (95% CI:[85.1, 95.7]) with a classification accuracy of 84.3%. A sensitivity was 85.7% and specificity was 82.8% on 431 training images (TB 172, none-TB 259) and 121 test images (TB 63, none-TB 58). This results show that detecting PTB using SegNet is comparable to other PTB detection methods.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.2
• /
• pp.47-52
• /
• 2016

The used digital images in the smart device and small displayer has been a downscaled image. In this paper, the detection of the downscaling image forgery is proposed using the feature vector according to the pixel value's gradients. In the proposed algorithm, AR (Autoregressive) coefficients are computed from pixel value's gradients of the image. These coefficients as the feature vectors are used in the learning of a SVM (Support Vector Machine) classification for the downscaling image forgery detector. On the performance of the proposed algorithm, it is excellent at the downscaling 90% image forgery compare to MFR (Median Filter Residual) scheme that had the same 10-Dim. feature vectors and 686-Dim. SPAM (Subtractive Pixel Adjacency Matrix) scheme. In averaging filtering ($3{\times}3$) and median filtering ($3{\times}3$) images, it has a higher detection ratio. Especially, the measured performances of all items in averaging and median filtering ($3{\times}3$), AUC (Area Under Curve) by the sensitivity and 1-specificity is approached to 1. Thus, it is confirmed that the grade evaluation of the proposed algorithm is 'Excellent (A)'.