• Journal of Broadcast Engineering
• /
• v.22 no.1
• /
• pp.62-69
• /
• 2017

Accurate detection of traffic lights is very important for the advanced driver assistance system (ADAS). There have been many research developments in this area. However, conventional of image processing methods are usually sensitive to varying illumination conditions. This paper proposes a traffic light detection algorithm to overcome this situation. The proposed algorithm first detects the candidates of traffic light using the proposed color map and hue-saturation-value (HSV) Traffic lights are then detected using the conventional histogram of oriented gradients (HOG) descriptor and support vector machine (SVM). Finally, the proposed Multilayer HOG descriptor is used to determine the direction information indicated by traffic lights. The proposed algorithm shows a high detection rate in real-time.

• The Journal of the Korea Contents Association
• /
• v.9 no.6
• /
• pp.266-274
• /
• 2009

An abdominal aortic aneurysm occurs most commonly in older individuals (between 65 and 75), and more in men and smokers. The most important complication of an abdominal aortic aneurysm is rupture, which is most often a fatal event. An abdominal aortic aneurysm weakens the walls of the blood vessel, leaving it vulnerable to bursting open, or rupturing, and spilling large amounts of blood into the abdominal cavity. surface modeling is very useful to surgery for quantitative analysis of abdominal aortic aneurysm. the 3D representation and surface modeling an abdominal aortic aneurysm structure taken from Multi Detector Computed Tomography. The construction of the 3D model is generally carried out by staking the contours obtained from 2D segmentation of each CT slice, so the quality of the 3D model strongly defends on the precision of segmentation process. In this work we present deformable model algorithm. deformable model is an energy-minimizing spline guided by external constraint force. External force which we call Gradient Vector Flow, is computed as a diffusion of a gradient vectors of gray level or binary edge map derived from the image. Finally, we have used snakes successfully for abdominal aortic aneurysm segmentation the performance of snake was visually and quantitatively validated by experts.

• Korean Journal of Remote Sensing
• /
• v.23 no.4
• /
• pp.273-285
• /
• 2007

Most stereo matching methods use intensity values in small image patches to measure the correspondence between two points. If the noisy pixels are used in computing the corresponding point, the matching performance becomes low. For this reason, the noise plays a critical role in determining the matching performance. In this paper, we propose a method for combining intensity and edge filters robust to the noise in order to improve the performance of stereo matching using high resolution satellite imagery. We used intensity filters such as Mean, Median, Midpoint and Gaussian filter and edge filters such as Gradient, Roberts, Prewitt, Sobel and Laplacian filter. To evaluate the performance of intensity and edge filters, experiments were carried out on both synthetic images and satellite images with uniform or gaussian noise. Then each filter was ranked based on its performance. Among the intensity and edge filters, Median and Sobel filter showed best performance while Midpoint and Laplacian filter showed worst result. We used Ikonos satellite stereo imagery in the experiments and the matching method using Median and Sobel filter showed better matching results than other filter combinations.

• Journal of the Society of Disaster Information
• /
• v.18 no.2
• /
• pp.364-373
• /
• 2022

Purpose: The purpose of this paper is to develope smoke detection using AI model for detecting the initial fire in underground utility tunnels using CCTV Method: To improve detection performance of smoke which is high irregular, a deep learning model for fire detection was trained to optimize smoke detection. Also, several approaches such as dataset cleansing and gradient exploding release were applied to enhance model, and compared with results of those. Result: Results show the proposed approaches can improve the model performance, and the final model has good prediction capability according to several indexes such as mAP. However, the final model has low false negative but high false positive capacities. Conclusion: The present model can apply to smoke detection in underground utility tunnel, fixing the defect by linking between the model and the utility tunnel control system.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.7
• /
• pp.33-40
• /
• 2009

Edge detection is very significant in low level image processing. However, majority edge detection methods are not only effective enough cause of the noise points' influence, even not flexible enough to different input images. In order to sort these problems, in this paper an algorithm is presented that has an extra noise reduction stage at first, and then automatically selects the both thresholds depending on gradient amplitude histogram and intra class minimum variance. Using this algorithm, can fade out almost all of the sensitive noise points, and calculate the propose thresholds for different images without setting up the practical parameters artificially, and then choose edge pixels by fuzzy algorithm. In finally, get the better result than the former Canny algorithm.

• Journal of radiological science and technology
• /
• v.36 no.4
• /
• pp.291-297
• /
• 2013

This study is aimed to evaluate the effect of $T_{1/2}$ upon count rates in the analysis of dynamic scan using NaI (Tl) scintillation camera, and suggest a new quality control method with this effects. We producted a point source with $^{99m}TcO_4^-$ of 18.5 to 185 MBq in the 2 mL syringes, and acquired 30 frames of dynamic images with 10 to 60 seconds each using Infinia gamma camera (GE, USA). In the second experiment, 90 frames of dynamic images were acquired from 74 MBq point source by 5 gamma cameras (Infinia 2, Forte 2, Argus 1). There were not significant differences in average count rates of the sources with 18.5 to 92.5 MBq in the analysis of 10 to 60 seconds/frame with 10 seconds interval in the first experiment (p>0.05). But there were significantly low average count rates with the sources over 111 MBq activity at 60 seconds/frame (p<0.01). According to the second analysis results of linear regression by count rates of 5 gamma cameras those were acquired during 90 minutes, counting efficiency of fourth gamma camera was most low as 0.0064%, and gradient and coefficient of variation was high as 0.0042 and 0.229 each. We could not find abnormal fluctuation in $x^2$ test with count rates (p>0.02), and we could find the homogeneity of variance in Levene's F-test among the gamma cameras (p>0.05). At the correlation analysis, there was only correlation between counting efficiency and gradient as significant negative correlation (r=-0.90, p<0.05). Lastly, according to the results of calculation of $T_{1/2}$ error from change of gradient with -0.25% to +0.25%, if $T_{1/2}$ is relatively long, or gradient is high, the error increase relationally. When estimate the value of 4th camera which has highest gradient from the above mentioned result, we could not see $T_{1/2}$ error within 60 minutes at that value. In conclusion, it is necessary for the scintillation gamma camera in medical field to manage hard for the quality of radiation measurement. Especially, we found a tendency that count rate changes over time at this study, and we proved that it can effect $T_{1/2}$. And also, there is need of appropriate phantoms and the method of quality management like this study, because there are not any advice or limitation degrees for domestic medical purpose scintillation camera.

• Journal of KIISE:Computer Systems and Theory
• /
• v.36 no.5
• /
• pp.404-411
• /
• 2009

We introduce in this paper a new method for smooth foldover-free warping of images, based on the vector field deformation technique proposed by Von Funck et al. It allows users to specify the constraints in two different ways: positional constraints to constrain the position of a point in the image and gradient constraints to constrain the orientation and scaling of some parts of the image. From the user-specified constraints, it computes in the image domain a C1-continuous velocity vector field, along which each pixel progressively moves from its original position to the target. The target positions of the pixels are obtained by solving a set of partial derivative equations with the 4th order Runge-Kutta method. We show how our method can be useful for texture mapping with hard constraints. We start with an unconstrained planar embedding of a target mesh using a previously known method (Least Squares Conformal Map). Then, in order to obtain a texture map that satisfies the given constraints, we use the proposed warping method to align the features of the texture image with those on the unconstrained embedding. Compared to previous work, our method generates a smoother texture mapping, offers higher level of control for defining the constraints, and is simpler to implement.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_2
• /
• pp.1691-1707
• /
• 2022

Cheonji Lake, the caldera of Baekdu Mountain, located on the border of the Korean Peninsula and China, alternates between melting and freezing seasonally. There is a magma chamber beneath Cheonji, and variations in the magma chamber cause volcanic antecedents such as changes in the temperature and water pressure of hot spring water. Consequently, there is an abnormal region in Cheonji where ice melts quicker than in other areas, freezes late even during the freezing period, and has a high-temperature water surface. The abnormal area is a discharge region for hot spring water, and its ice gradient may be used to monitor volcanic activity. However, due to geographical, political and spatial issues, periodic observation of abnormal regions of Cheonji is limited. In this study, the degree of ice change in the optimal region was quantified using a Landsat -5/-7/-8 optical satellite image and a Modified U-Net regression model. From January 22, 1985 to December 8, 2020, the Visible and Near Infrared (VNIR) band of 83 Landsat images including anomalous regions was utilized. Using the relative spectral reflectance of water and ice in the VNIR band, unique data were generated for quantitative ice variability monitoring. To preserve as much information as possible from the visible and near-infrared bands, ice gradient was noticed by applying it to U-Net with two encoders, achieving good prediction accuracy with a Root Mean Square Error (RMSE) of 140 and a correlation value of 0.9968. Since the ice change value can be seen with high precision from Landsat images using Modified U-Net in the future may be utilized as one of the methods to monitor Baekdu Mountain's volcanic activity, and a more specific volcano monitoring system can be built.

• Progress in Medical Physics
• /
• v.24 no.2
• /
• pp.99-107
• /
• 2013

The surgical resection was occurred mainly in liver metastasis before the development of radiation therapy techniques. Recently, Radiation therapy is increased gradually due to the development of radiation dose delivery techniques. 18F-FDG PET image showed better sensitivity and specificity in liver metastasis detection. This image modality is important in the radiation treatment with planning CT for tumor delineation. In this study, we applied automatic image segmentation methods on PET image of liver metastasis and examined the impact of image factors on these methods. We selected the patients who were received the radiation therapy and 18F-FDG PET/CT in Korea Cancer Center Hospital from 2009 to 2012. Then, three kinds of image segmentation methods had been applied; The relative threshold method, the Gradient method and the region growing method. Based on these results, we performed statistical analysis in two directions. 1. comparison of GTV and image segmentation results. 2. performance of regression analysis for relation between image factor affecting image segmentation techniques. The mean volume of GTV was $60.9{\pm}65.9$ cc and the $GTV_{40%}$ was $22.43{\pm}35.27$ cc, and the $GTV_{50%}$ was $10.11{\pm}17.92$ cc, the $GTV_{RG}$ was $32.89{\pm}36.8$4 cc, the $GTV_{GD}$ was $30.34{\pm}35.77$ cc, respectively. The most similar segmentation method with the GTV result was the region growing method. For the quantitative analysis of the image factors which influenced on the region growing method, we used the standardized coefficient ${\beta}$, factors affecting the region growing method show GTV, $TumorSUV_{MAX/MIN}$, $SUV_{max}$, TBR in order. The result of the region growing (automatic segmentation) method showed the most similar result with the CT based GTV and the region growing method was affected by image factors. If we define the tumor volume by the auto image segmentation method which reflect the PET image parameters, more accurate and consistent tumor contouring can be done. And we can irradiate the optimized radiation dose to the cancer, ultimately.

• The Korean Journal of Applied Statistics
• /
• v.24 no.2
• /
• pp.359-371
• /
• 2011

Edge detection in images is an important step in image segmentation and object recognition as preprocessing for image processing. This paper presents a new edge detection using local histogram analysis based on wavelet transform. In this work, the wavelet transform uses three components (horizontal, vertical and diagonal) to find the magnitude of the gradient vector, instead of the conventional approach in which tw components are used. We compare the magnitude of the gradient vector with the threshold that is obtained from a local histogram analysis to conclude that an edge is present or not. Some experimental results for our edge detector with a Sobel, Canny, Scale Multiplication, and Mallat edge detectors on sample images are given and the performances of these edge detectors are compared in terms of quantitative and qualitative measures. Our detector performs better than the other wavelet-based detectors such as Scale Multiplication and Mallat detectors. Our edge detector also preserves a good performance even if the Sobel and Canny detector are sharply low when the images are highly corrupted.