• Journal of the Korean Society of Radiology
• /
• v.10 no.3
• /
• pp.215-222
• /
• 2016

X-ray mammography is the most effective method for the diagnosis of calcified lesions of various breast diseases. To reduce patient dose and to obtain optimal image required for diagnosis, the performance of the mammography system should be maintained continuously. Because the target (anode) angle of the X-ray tube is measured from the central X-ray, the effective angle can be slightly different in view of the position on the detector, which can result in degrading spatial resolution of the imaging within the field of view. In this study, we measured the MTF to examine spatial resolution for positions on the detector in the digital mammography system. For a tungsten wire of $50{\mu}m$ diameter, the highest spatial frequency was obtained. It meant that a wire diameter for measuring MTF through LSF should be small compared to the pixel size of the detector used in the mammography system. The spatial resolution showed slightly different performance according to positions on the detector. The center position gave the best spatial resolution and positions away from the center showed the degraded performance although the difference of the spatial resolution was small. The effective focal spot size of the full width at half maximum also showed similar result. It concluded that the slightly increase of the effective focal spot size gave the degradation of the spatial resolution for positions on the detector.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.33 no.4
• /
• pp.267-275
• /
• 2015

Those land cover maps have widely been used in various fields, such as environmental studies, military strategies as well as in decision-makings. This study proposes a method to extract training data, automatically and classify the cover using ingle satellite images and national land cover maps, provided by the Ministry of Environment. For this purpose, as the initial training data, those three were used; the unsupervised classification, the ISODATA, and the existing land cover maps. The class was classified and named automatically using the class information in the existing land cover maps to overcome the difficulty in selecting classification by each class and in naming class by the unsupervised classification; so as achieve difficulty in selecting the training data in supervised classification. The extracted initial training data were utilized as the training data of MLC for the land cover classification of target satellite images, which increase the accuracy of unsupervised classification. Finally, the land cover maps could be extracted from updated training data that has been applied by an iterative method. Also, in order to reduce salt and pepper occurring in the pixel classification method, the MRF was applied in each repeated phase to enhance the accuracy of classification. It was verified quantitatively and visually that the proposed method could effectively generate the land cover maps.

• Korean Journal of Remote Sensing
• /
• v.11 no.3
• /
• pp.49-60
• /
• 1995

Features of geological lineaments generally play an important role at the data interpretation concerned geological processes, mineral exploration or natural hazard risk estimation. However, there are intrinsically discordances between lineaments-related features extracted from surficial geological syrvey and those from satellite imagery;nevertheless, any data set contained those information should not be considred as less meaningful within their own task. For the purpose of effective utilization task of extracted lineaments, the mathematical scheme, based on fuzzy set theory, for practical integration of various types of rasterized data sets is studied. As a real application, the geological map named Homyeong sheet(1:50,000) and the Landset TM imageries covering same area were used, and then lineaments-related data sets such as lineaments on the geological map, lineaments extracted from a false-color image composite satellite, and major drainage pattern were utilized. For data fusion process, fuzzy membership functions of pixel values in each data set were experimentally assigned by percentile, and then fuzzy algebraic sum operator was tested. As a result, integrated lineaments by this well-known operator are regarded as newly-generated reasonable ones. Conclusively, it was thought that the implementation within available GISs, or the stand-alone module for general applications of this simple scheme can be utilized as an effective scheme can be utilized as an effective scheme for further studies for spatial integration task for providing decision-supporting information, or as a kind of spatial reasoning scheme.

• Korean Journal of Remote Sensing
• /
• v.22 no.6
• /
• pp.553-563
• /
• 2006

This study aims to produce land-cover maps of Korean peninsula using multi-temporal MODIS (Moderate Resolution Imaging Spectroradiometer) imagery. To solve the low spatial resolution of MODIS data and enhance classification accuracy, Linear Spectral Mixture Analysis (LSMA) was employed. LSMA allowed to determine the fraction of each surface type in a pixel and develop vegetation, soil and water fraction images. To eliminate clouds, MVC (Maximum Value Composite) was utilized for vegetation fraction and MinVC (Minimum Value Composite) for soil fraction image respectively. With these images, using ISODATA unsupervised classifier, southern part of Korean peninsula was classified to low and mid level land-cover classes. The results showed that vegetation and soil fraction images reflected phenological characteristics of Korean peninsula. Paddy fields and forest could be easily detected in spring and summer data of the entire peninsula and arable land in North Korea. Secondly, in low level land-cover classification, overall accuracy was 79.94% and Kappa value was 0.70. Classification accuracy of forest (88.12%) and paddy field (85.45%) was higher than that of barren land (60.71%) and grassland (57.14%). In midlevel classification, forest class was sub-divided into deciduous and conifers and field class was sub-divided into paddy and field classes. In mid level, overall accuracy was 82.02% and Kappa value was 0.6986. Classification accuracy of deciduous (86.96%) and paddy (85.38%) were higher than that of conifers (62.50%) and field (77.08%).

• Korean Journal of Remote Sensing
• /
• v.37 no.6_3
• /
• pp.1965-1974
• /
• 2021

Satellite remote sensing approach can be actively used for forest monitoring. Especially, it is much meaningful to utilize Korea multi-purpose satellites, an independently operated satellite in Korea, for forest monitoring of Korea, Recently, several studies have been performed to exploit meaningful information from satellite remote sensed data via machine learning approaches. The forest information produced through machine learning approaches can be used to support the efficiency of traditional forest monitoring methods, such as in-situ survey or qualitative analysis of aerial image. The performance of machine learning approaches is greatly depending on the characteristics of study area and data. Thus, it is very important to survey the best model among the various machine learning models. In this study, the performance of deep neural network to classify artificial or natural forests was analyzed in Samcheok, Korea. As a result, the pixel accuracy was about 0.857. F1 scores for natural and artificial forests were about 0.917 and 0.433 respectively. The F1 score of artificial forest was low. However, we can find that the artificial and natural forest classification performance improvement of about 0.06 and 0.10 in F1 scores, compared to the results from single layered sigmoid artificial neural network. Based on these results, it is necessary to find a more appropriate model for the forest type classification by applying additional models based on a convolutional neural network.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.25 no.4
• /
• pp.65-74
• /
• 2021

In order to efficiently inspect rapidly increasing old tunnels in many well-developed countries, many inspection methodologies have been proposed using imaging equipment and image processing. However, most of the existing methodologies evaluated their performance on a clean concrete surface with a limited area where other objects do not exist. Therefore, this paper proposes a 6-step framework for tunnel crack detection deep learning model development. The proposed method is mainly based on negative sample (non-crack object) training and Cascade Mask R-CNN. The proposed framework consists of six steps: searching for cracks in images captured from real tunnels, labeling cracks in pixel level, training a deep learning model, collecting non-crack objects, retraining the deep learning model with the collected non-crack objects, and constructing final training dataset. To implement the proposed framework, Cascade Mask R-CNN, an instance segmentation model, was trained with 1561 general crack images and 206 non-crack images. In order to examine the applicability of the trained model to the real-world tunnel crack detection, field testing is conducted on tunnel spans with a length of about 200m where electric wires and lights are prevalent. In the experimental result, the trained model showed 99% precision and 92% recall, which shows the excellent field applicability of the proposed framework.

• Korean Journal of Remote Sensing
• /
• v.35 no.2
• /
• pp.229-239
• /
• 2019

All vegetation colonies have layered structure. This layer is called 'forest vertical structure.' Nowadays it is considered as an important indicator to estimate forest's vital condition, diversity and environmental effect of forest. So forest vertical structure should be surveyed. However, vertical structure is a kind of inner structure, so forest surveys are generally conducted through field surveys, a traditional forest inventory method which costs plenty of time and budget. Therefore, in this study, we propose a useful method to classify the vertical structure of forests using remote sensing aerial photographs and machine learning capable of mass data mining in order to reduce time and budget for forest vertical structure investigation. We classified it as SVM (Support Vector Machine) using RGB airborne photos and LiDAR (Light Detection and Ranging) DSM (Digital Surface Model) DTM (Digital Terrain Model). Accuracy based on pixel count is 66.22% when compared to field survey results. It is concluded that classification accuracy of layer classification is relatively high for single-layer and multi-layer classification, but it was concluded that it is difficult in multi-layer classification. The results of this study are expected to further develop the field of machine learning research on vegetation structure by collecting various vegetation data and image data in the future.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_2
• /
• pp.1663-1676
• /
• 2022

Recently, a number of deep-learning based land cover segmentation studies have been introduced. Some studies denoted that the performance of land cover segmentation deteriorated due to insufficient training data. In this study, we verified the improvement of land cover segmentation performance through data augmentation. U-Net was implemented for the segmentation model. And 2020 satellite-derived landcover dataset was utilized for the study data. The pixel accuracies were 0.905 and 0.923 for U-Net trained by original and augmented data respectively. And the mean F1 scores of those models were 0.720 and 0.775 respectively, indicating the better performance of data augmentation. In addition, F1 scores for building, road, paddy field, upland field, forest, and unclassified area class were 0.770, 0.568, 0.433, 0.455, 0.964, and 0.830 for the U-Net trained by original data. It is verified that data augmentation is effective in that the F1 scores of every class were improved to 0.838, 0.660, 0.791, 0.530, 0.969, and 0.860 respectively. Although, we applied data augmentation without considering class balances, we find that data augmentation can mitigate biased segmentation performance caused by data imbalance problems from the comparisons between the performances of two models. It is expected that this study would help to prove the importance and effectiveness of data augmentation in various image processing fields.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_4
• /
• pp.1925-1934
• /
• 2022

Monitoring nuclear activity for inaccessible areas using remote sensing technology is essential for nuclear non-proliferation. In recent years, deep learning has been actively used to detect nuclear-activity-related small objects. However, high-resolution satellite imagery containing small objects can result in class imbalance. As a result, there is a performance degradation problem in detecting small objects. Therefore, this study aims to improve detection accuracy by analyzing the effect of the ratio of small objects related to nuclear activity in the input data for the performance of the deep learning model. To this end, six case datasets with different ratios of small object pixels were generated and a U-Net model was trained for each case. Following that, each trained model was evaluated quantitatively and qualitatively using a test dataset containing various types of small object classes. The results of this study confirm that when the ratio of object pixels in the input image is adjusted, small objects related to nuclear activity can be detected efficiently. This study suggests that the performance of deep learning can be improved by adjusting the object pixel ratio of input data in the training dataset.

• Journal of Platform Technology
• /
• v.10 no.3
• /
• pp.60-68
• /
• 2022

As AI-technology develops, interest in the safety of autonomous driving is increasing. Recently, autonomous vehicles have been increasing, but efforts to solve side effects have been sluggish. In particular, night autonomous vehicles have more problems. This is because the probability of accidents is higher in the night driving environment than in the day environment. There are more factors to consider for self-driving at night. Among these factors, reflection of light or reflected light of lighting may be a fundamental cause of night accidents. Therefore, this study proposes method to reduce accidents and improve safety by reducing reflected light generated by the headlights of opposite vehicles or various surrounding light that appear as an important problem in night autonomous vehicles. Therefore, first, in an image obtained by a sensor of a night autonomous vehicle, illumination reflected light is extracted using reflected light characteristic information, and a color of each pixel using a reflection coefficient is found to reduce a special area generated by geometric characteristics. In addition, we find a new area using only the brightness component of the specular area, define it as Illuminated Reflection Light (IRL), and finally present a method to reduce it. Although the illumination reflection light could not be completely reduce, generally satisfactory results could be obtained. Therefore, it is believed that the proposed study can reduce casualties by solving the problems of night autonomous driving and improving safety.