• Journal of Korea Multimedia Society
• /
• v.25 no.8
• /
• pp.1212-1223
• /
• 2022

Skin lesions are common diseases that range from skin rashes to skin cancer, which can lead to death. Note that early diagnosis of skin diseases can be important because early diagnosis of skin diseases considerably can reduce the course of treatment and the harmful effect of the disease. Recently, the development of computer-aided diagnosis (CAD) systems based on artificial intelligence has been actively made for the early diagnosis of skin diseases. In a typical CAD system, the accurate classification of skin lesion types is of great importance for improving the diagnosis performance. Motivated by this, we propose a novel deep ensemble classification with multi-scale attention networks. The proposed deep ensemble networks are jointly trained using a single loss function in an end-to-end manner. In addition, the proposed deep ensemble network is equipped with a multi-scale attention mechanism and segmentation information of the original skin input image, which improves the classification performance. To demonstrate our method, the publicly available human skin disease dataset (HAM 10000) and the private animal skin lesion dataset were used for the evaluation. Experiment results showed that the proposed methods can achieve 97.8% and 81% accuracy on each HAM10000 and animal skin lesion dataset. This research work would be useful for developing a more reliable CAD system which helps doctors early diagnose skin diseases.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.5 s.305
• /
• pp.29-40
• /
• 2005

According as base of digital hologram has been magnified, discussion of compression technology is expected as a international standard which defines the compression technique of 3D image and video has been progressed in form of 3DAV which is a part of MPEG. As we can identify in case of 3DAV, the coding technique has high possibility to be formed into the hybrid type which is a merged, refined, or mixid with the various previous technique. Therefore, we wish to present the relationship between various image/video coding techniques and digital hologram In this paper, we propose an efficient coding method of digital hologram using standard compression tools for video and image. At first, we convert fringe patterns into video data using a principle of CGH(Computer Generated Hologram), and then encode it. In this research, we propose a compression algorithm is made up of various method such as pre-processing for transform, local segmentation with global information of object image, frequency transform for coding, scanning to make fringe to video stream, classification of coefficients, and hybrid video coding. Finally the proposed hybrid compression algorithm is all of these methods. The tool for still image coding is JPEG2000, and the toots for video coding include various international compression algorithm such as MPEG-2, MPEG-4, and H.264 and various lossless compression algorithm. The proposed algorithm illustrated that it have better properties for reconstruction than the previous researches on far greater compression rate above from four times to eight times as much. Therefore we expect that the proposed technique for digital hologram coding is to be a good preceding research.

• Journal of Korean Society for Geospatial Information Science
• /
• v.23 no.4
• /
• pp.25-33
• /
• 2015

The study of ecosystem assessment(ES) is based on land cover information, and primarily it is performed at the global scale. However, these results as data for decision making have a limitation at the aspects of range and scale to solve the regional issue. Although the Ministry of Environment provides available land cover data at the regional scale, it is also restricted in use due to the intrinsic limitation of on screen digitizing method and temporal and spatial difference. This study of objective is to generate UAV land cover map. In order to classify the imagery, we have performed resampling at 5m resolution using UAV imagery. The results of object-based image segmentation showed that scale 20 and merge 34 were the optimum weight values for UAV imagery. In the case of RapidEye imagery;we found that the weight values;scale 30 and merge 30 were the most appropriate at the level of land cover classes for sub-category. We generated land cover imagery using example-based classification method and analyzed the accuracy using stratified random sampling. The results show that the overall accuracies of RapidEye and UAV classification imagery are each 90% and 91%.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.18 no.1
• /
• pp.19-25
• /
• 2014

Purpose: Integrated PET/MRI has been developed recently has become a lot of help to the point oncologic, neological, cardiological nuclear medicine. By using this PET/MRI, a ${\mu}-map$ is created some special MRI sequence which may be divided parts of the body for attenuation correction. However, because an MRI contrast agent is necessary in order to obtain an more MRI information, we will evaluate to see an effect of SUV on PET image that corrected attenuation by MRI with contrast agent. Materials and Methods: As PET/MRI machine, Biograph mMR (Siemens, Germany) was used. For phantom test, 1mCi $^{18}F-FDG$ was injected in cylinderical uniformity phantom, and then acquire PET data about 10 minutes with VIBE-DIXON, UTE MRI sequence image for attenuation correction. T1 weighted contrast media, 4 cc DOTAREM (GUERBET, FRANCE) was injected in a same phatnom, and then PET data, MRI data were acquired by same methodes. Using this PET, non-contrast MRI and contrast MRI, it was reconstructed attenuation correction PET image, in which we evanuated the difference of SUVs. Additionally, for let a high desity of contrast media, 500 cc 2 plastic bottles were used. We injected $^{18}F-FDG$ with 5 cc DOTAREM in first bottle. At second bottle, only $^{18}F-FDG$ was injected. and then we evaluated a SUVs reconstructed by same methods. For clinical patient study, rectal caner-pancreas cancer patients were selected. we evaluated SUVs of PET image corrected attenuastion by contrast weighted MRI and non-contrast MRI. Results: For a phantom study, although VIBE DIXON MRI signal with contrast media is 433% higher than non-contrast media MRI, the signals intensity of ${\mu}-map$, attenuation corrected PET are same together. In case of high contrast media density, image distortion is appeared on ${\mu}-map$ and PET images. For clinical a patient study, VIBE DIXON MRI signal on lesion portion is increased in 495% by using DOTAREM. But there are no significant differences at ${\mu}-map$, non AC PET, AC-PET image whether using contrast media or not. In case of whole body PET/MRI study, %diff between contras and non contrast MRAC at lung, liver, renal cortex, femoral head, myocardium, bladder, muscle are -4.32%, -2.48%, -8.05%, -3.14%, 2.30%, 1.53%, 6.49% at each other. Conclusion: In integrated PET/MRI, a segmentation ${\mu}-map$ method is used for correcting attenuation of PET signal. although MRI signal for attenuation correciton change by using contrast media, ${\mu}-map$ will not change, and then MRAC PET signal will not change too. Therefore, MRI contrast media dose not affect for attenuation correction PET. As well, not only When we make a flow of PET/MRI protocol, order of PET and MRI sequence dose not matter, but It's possible to compare PET images before and after contrast agent injection.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.20 no.1
• /
• pp.1-14
• /
• 2017

A land cover map provides basic information to help understand the current state of a region, but its utilization in the ecological research field has deteriorated due to limited temporal and spatial resolutions. The purpose of this study was to investigate the possibility of using a land cover map with data based on high resolution images acquired by UAV. Using the UAV, 10.5 cm orthoimages were obtained from the $2.5km^2$ study area, and land cover maps were obtained from object-based and pixel-based classification for comparison and analysis. From accuracy verification, classification accuracy was shown to be high, with a Kappa of 0.77 for the pixel-based classification and a Kappa of 0.82 for the object-based classification. The overall area ratios were similar, and good classification results were found in grasslands and wetlands. The optimal image segmentation weights for object-based classification were Scale=150, Shape=0.5, Compactness=0.5, and Color=1. Scale was the most influential factor in the weight selection process. Compared with the pixel-based classification, the object-based classification provides results that are easy to read because there is a clear boundary between objects. Compared with the land cover map from the Ministry of Environment (subdivision), it was effective for natural areas (forests, grasslands, wetlands, etc.) but not developed areas (roads, buildings, etc.). The application of an object-based classification method for land cover using UAV images can contribute to the field of ecological research with its advantages of rapidly updated data, good accuracy, and economical efficiency.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.24 no.6
• /
• pp.583-598
• /
• 2022

As human-based tunnel inspections are affected by the subjective judgment of the inspector, making continuous history management difficult. There is a lot of deep learning-based automatic crack detection research recently. However, the large public crack datasets used in most studies differ significantly from those in tunnels. Also, additional work is required to build sophisticated crack labels in current tunnel evaluation. Therefore, we present a method to improve crack detection performance by inputting existing datasets into a deep learning model. We evaluate and compare the performance of deep learning models trained by combining existing tunnel datasets, high-quality tunnel datasets, and public crack datasets. As a result, DeepLabv3+ with Cross-Entropy loss function performed best when trained on both public datasets, patchwise classification, and oversampled tunnel datasets. In the future, we expect to contribute to establishing a plan to efficiently utilize the tunnel image acquisition system's data for deep learning model learning.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.25 no.4
• /
• pp.63-80
• /
• 2022

As a technology to analyze and predict an issue has been developed by constructing real space into virtual space, it is becoming more important to acquire precise spatial information in complex cities. In this study, images were acquired using an unmanned aerial vehicle for urban area with complex landscapes, and land cover classification was performed object-based image analysis and semantic segmentation techniques, which were image classification technique suitable for high-resolution imagery. In addition, based on the imagery collected at the same time, the replicability of land cover classification of each artificial intelligence (AI) model was examined for areas that AI model did not learn. When the AI models are trained on the training site, the land cover classification accuracy is analyzed to be 89.3% for OBIA-RF, 85.0% for OBIA-DNN, and 95.3% for U-Net. When the AI models are applied to the replicability assessment site to evaluate replicability, the accuracy of OBIA-RF decreased by 7%, OBIA-DNN by 2.1% and U-Net by 2.3%. It is found that U-Net, which considers both morphological and spectroscopic characteristics, performs well in land cover classification accuracy and replicability evaluation. As precise spatial information becomes important, the results of this study are expected to contribute to urban environment research as a basic data generation method.

• Journal of Korea Multimedia Society
• /
• v.9 no.7
• /
• pp.808-817
• /
• 2006

Although the generally used the velocity index of doppler effect is a very significant factor in the functional evaluation of the left ventricle, it depends on the subjective evaluation of an inspector. The objective data of the motility can be obtained from the gated myocardial SPECT images by quantitative analysis. However, it is difficult to image visual of the velocity of the motion. The aim of our study is to develop a new method for the imaging velocity using the gated myocardial SPECT images and use it as an evaluation index for analyzing motility. First we visualized left ventricle into 3 dimensions using the coordinates of the points which were obtained through a segmentation of myocardium. Each point was represented by the different colors, according to the velocity of each point. We performed a validation study using 7 normal subjects and 15 myocardial infarction patients. To analyze motility, we used the average of the moved distance and the velocity. In normal cases, the average of the moved distance was 4.3mm and the average of the velocity was 11.9mm. In patient cases, the average of the moved distance was 3.9mm and the average of the velocity was 10.5mm. These results show that the motility of normal subjects is higher than the abnormal subjects. We expect that our proposed method could become a way to improve the accuracy and reproducibility for the functional evaluation of myocardial wall.

• Journal of Korean Society for Geospatial Information Science
• /
• v.4 no.1 s.6
• /
• pp.93-102
• /
• 1996

For an atomatic generation of DEM(Digital Elevation Model) by computer, it is a time-consumed work to determine adquate matches from stereo images. Correlation and evenly distributed area-based method is generally used for matching operation. In this paper, we propose a new approach that computes matches efficiantly by changing the size of mask window and search area according to the given terrain information. For image segmentation, at first edge-preserving smoothing filter is used for preprocessing, and then region growing algorithm is applied for the filterd images. The segmented regions are classifed into mountain, plain and water area by using MRF(Markov Random Filed) model. Maching is composed of predicting parallex and fine matching. Predicted parallex determines the location of search area in fine matching stage. The size of search area and mask window is determined by terrain information for each pixel. The execution time of matching is reduced by lessening the size of search area in the case of plain and water. For the experiments, four images which are covered $10km{\times}10km(1024{\times}1024\;pixel)$ of Taejeon-Kumsan in each are studied. The result of this study shows that the computing time of the proposed method using terrain information for matching operation can be reduced from 25% to 35%.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.8
• /
• pp.132-139
• /
• 2015

Visual SLAM(Simultaneous Localization And Mapping) has been used widely to estimate a mobile robot's location. Visual SLAM estimates relative motions with static visual features over image sequence. Because visual SLAM methods assume generally static features in the environment, we cannot obtain precise results in dynamic situation including many moving objects: cars and human beings. This paper presents a stereo vision based SLAM method in dynamic environment. First, we extract disparity map with stereo vision and compute optical flow. We then compute disparity change that is the estimated flow field between stereo views. After examining the disparity change value, we detect ROIs(Region Of Interest) in disparity space to determine dynamic scene objects. In indoor environment, many structural planes like walls may be determined as false dynamic elements. To solve this problem, we segment the scene into planar structure. More specifically, disparity values by the stereo vision are projected to X-Z plane and we employ Hough transform to determine planes. In final step, we remove ROIs nearby the walls and discriminate static scene elements in indoor environment. The experimental results show that the proposed method can obtain stable performance in dynamic environment.