• Korean Journal of Optics and Photonics
• /
• v.26 no.2
• /
• pp.88-97
• /
• 2015

An omnidirectional optical system can be described as a special optical system that images in real time a panoramic image with an azimuthal angle of $360^{\circ}$ and the altitude angle corresponding to the upper and lower fields of view from the horizon line. In this paper, for easy fabrication and compact size, we designed and fabricated a catadioptric omnidirectional optical system consisting of the mirror part of a spherical mirror with a central hole (that is, obscuration), a plane mirror, the imaging lens part of 3 single spherical lenses, and a spherical doublet in the visible light spectrum. We evaluated its image performance by measuring the cut-off spatial frequency using automobile license plates, and the vertical field of view using an ISO 12233 chart. We achieved a catadioptric omnidirectional optical system with vertical field of view from $+53^{\circ}$ to $-17^{\circ}$ and an azimuthal angle of $360^{\circ}$. This optical system cleaniy imaged letters on a car's front license plate at the object distance of 3 meters, which corresponds to a cut-off spatial frequency of 135 lp/mm.

• Korean Journal of Remote Sensing
• /
• v.37 no.5_1
• /
• pp.1149-1161
• /
• 2021

With a trend of the utilization of computer vision for satellite images, cloud detection using deep learning also attracts attention recently. In this study, we conducted a U-Net cloud detection modeling using SPARCS (Spatial Procedures for Automated Removal of Cloud and Shadow) Cloud Dataset with the image data augmentation and carried out 10-fold cross-validation for an objective assessment of the model. Asthe result of the blind test for 1800 datasets with 512 by 512 pixels, relatively high performance with the accuracy of 0.821, the precision of 0.847, the recall of 0.821, the F1-score of 0.831, and the IoU (Intersection over Union) of 0.723. Although 14.5% of actual cloud shadows were misclassified as land, and 19.7% of actual clouds were misidentified as land, this can be overcome by increasing the quality and quantity of label datasets. Moreover, a state-of-the-art DeepLab V3+ model and the NAS (Neural Architecture Search) optimization technique can help the cloud detection for CAS500 (Compact Advanced Satellite 500) in South Korea.

• Journal of Broadcast Engineering
• /
• v.26 no.2
• /
• pp.125-131
• /
• 2021

In recent years, Convolutional Neural Networks (CNNs) have achieved outstanding performance in the fields of computer vision such as image classification, object detection, visual quality enhancement, etc. However, as huge amount of computation and memory are required in CNN models, there is a limitation in the application of CNN to low-power environments such as mobile or IoT devices. Therefore, the need for neural network compression to reduce the model size while keeping the task performance as much as possible has been emerging. In this paper, we propose a method to compress CNN models by combining matrix decomposition methods of LR (Low-Rank) approximation and CP (Canonical Polyadic) decomposition. Unlike conventional methods that apply one matrix decomposition method to CNN models, we selectively apply two decomposition methods depending on the layer types of CNN to enhance the compression performance. To evaluate the performance of the proposed method, we use the models for image classification such as VGG-16, RestNet50 and MobileNetV2 models. The experimental results show that the proposed method gives improved classification performance at the same range of 1.5 to 12.1 times compression ratio than the existing method that applies only the LR approximation.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.1
• /
• pp.161-170
• /
• 2019

Multi-product mass production small and medium enterprise factories have a wide variety of products and a large number of products, wasting manpower and expenses for inventory management. In addition, there is no way to check the status of inventory in real time, and it is suffering economic damage due to excess inventory and shortage of stock. There are many ways to build a real-time data collection environment, but most of them are difficult to afford for small and medium-sized companies. Therefore, smart factories of small and medium enterprises are faced with difficult reality and it is hard to find appropriate countermeasures. In this paper, we implemented the contents of extension of existing inventory management method through character extraction on label with barcode and QR code, which are widely adopted as current product management technology, and evaluated the effect. Technically, through preprocessing using OpenCV for automatic recognition and classification of stock labels and barcodes, which is a method for managing input and output of existing products through computer image processing, and OCR (Optical Character Recognition) function of Google vision API. And it is designed to recognize the barcode through Zbar. We propose a method to manage inventory by real-time image recognition through Raspberry Pi without using expensive equipment.

• Journal of Cadastre & Land InformatiX
• /
• v.51 no.2
• /
• pp.169-184
• /
• 2021

Recently, land cover spectral libraries have been widely used in studies to classify land cover based on hyperspectral images. Overseas, various institutions have built and provided land cover spectral libraries, but in Korea, the building and provision of land cover spectral libraries is insufficient. Against this background, the purpose of this study is to suggest the possibility of using domestic and foreign spectral libraries in the classification studies of domestic land cover. Band matching is required for comparative analysis of the spectral libraries and land cover classification using the spectral libraries, and in this study, an automation logic to automatically perform this is presented. In addition, the directly constructed domestic land cover spectral library and the existing overseas land cover spectral library were comparatively analyzed. As a result, the directly constructed land cover spectral library had the highest correlation coefficient of 0.974. Finally, for the accuracy evaluation, aerial hyperspectral images of the study area were supervised and classified using the domestic and foreign land cover spectral libraries using the SAM technique. As a result of the accuracy evaluation, it is judged that Soils, Artificial Materials, and Coatings among the classification items of the foreign land cover spectral library can be sufficiently applied to classify the cover in Korea.

• Journal of Internet Computing and Services
• /
• v.23 no.3
• /
• pp.35-45
• /
• 2022

Even in recent years, treatment of first-aid patients is still often delayed due to a shortage of medical resources in marginalized areas. Research on automating the analysis of medical data to solve the problems of inaccessibility for medical services and shortage of medical personnel is ongoing. Computer vision-based medical inspection automation requires a lot of cost in data collection and labeling for training purposes. These problems stand out in the works of classifying lesion that are rare, or pathological features and pathogenesis that are difficult to clearly define visually. Anomaly detection is attracting as a method that can significantly reduce the cost of data collection by adopting an unsupervised learning strategy. In this paper, we propose methods for detecting abnormal images on chest X-RAY images as follows based on existing anomaly detection techniques. (1) Normalize the brightness range of medical images resampled as optimal resolution. (2) Some feature vectors with high representative power are selected in set of patch features extracted as intermediate-level from lesion-free images. (3) Measure the difference from the feature vectors of lesion-free data selected based on the nearest neighbor search algorithm. The proposed system can simultaneously perform anomaly classification and localization for each image. In this paper, the anomaly detection performance of the proposed system for chest X-RAY images of PA projection is measured and presented by detailed conditions. We demonstrate effect of anomaly detection for medical images by showing 0.705 classification AUROC for random subset extracted from the PadChest dataset. The proposed system can be usefully used to improve the clinical diagnosis workflow of medical institutions, and can effectively support early diagnosis in medically poor area.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.50 no.6
• /
• pp.42-57
• /
• 2022

This study aims to create a basic model for classifying the activity photos that urban park users shared on social media using Deep Learning through Artificial Intelligence. Regarding the social media data, photos related to urban parks were collected through a Naver search, were collected, and used for the classification model. Based on the indicators of Naturalness, Potential Attraction, and Activity, which can be used to evaluate the characteristics of urban parks, 21 classification categories were created. Urban park photos shared on Naver were collected by category, and annotated datasets were created. A custom CNN model and a transfer learning model utilizing a CNN pre-trained on the collected photo datasets were designed and subsequently analyzed. As a result of the study, the Xception transfer learning model, which demonstrated the best performance, was selected as the urban park user activity image classification model and evaluated through several evaluation indicators. This study is meaningful in that it has built AI as an index that can evaluate the characteristics of urban parks by using user-shared photos on social media. The classification model using Deep Learning mitigates the limitations of manual classification, and it can efficiently classify large amounts of urban park photos. So, it can be said to be a useful method that can be used for the monitoring and management of city parks in the future.

• Journal of Practical Agriculture & Fisheries Research
• /
• v.21 no.2
• /
• pp.103-114
• /
• 2019

In this study we examined the preferences of college life factors for students at Korea National College of Agriculture and Fisheries(KNCAF). Analytical techniques of unstructured data used opinion mining and text mining techniques, and the results of text mining were visualized as word cloud. And those results were used for statistical analysis of the students' willingness to farm after graduation. The items of the favorable survey consisted of 10 items in 5 areas including university image, self-capacity, dormitory, education system, and future vision. After classifying the emotions of positive and negative in the collected questionnaire, a dictionary of positive and negative was created to evaluate the preference. The items of 'college image' at the time of university support, 'self after 10 years' after graduation, 'self-capacity' and 'present KNCAF' showed high positive emotion. On the other hand, positive emotion was low in the items of 'college dormitory', 'educational course', 'long-term field practice' and 'future of Korean agriculture'. In the cross-analysis of the difference in the will to farming according to gender, farming base, and entrance motivation, the will to farm according to gender and entrance motivation showed statistically significant results, but it was not significant in farming base. Also in binary logistic regression analysis on the will to farming, the statistically significant variable was found to be 'motivation for admission'

• Journal of The Korean Association For Science Education
• /
• v.26 no.6
• /
• pp.675-690
• /
• 2006

This study investigated students' views in relation to a career in science as a first step towards developing science career education programs and materials. The instrument was developed through preliminary surveys. The questionnaire was sent to, administered and returned by 3608 students including 1036 primary, 1137 middle, and 1435 high school students, who were selected through stratified cluster sampling from all over the country. The results were analyzed using statistical package program. The students' image view of science, preference for science and science learning, perception of science achievement, career hope for oneself, degree of hope in getting science-related job, perception about science-related career, important factors of science career choice, and their hope for science/non-science career and the reason for their choice in the past and present were investigated. The results show that "science as an experimentation activity" was the most frequent image view of science. The preference for science and science learning was more positive than the perception of science achievement. The percentage of total students who want to have science career was 21%, and this percentage of middle school students was especially low. The most frequent answers for advantages of science career were 'useful for national development', 'possible to get new knowledge', and those for disadvantages were 'dangerous', and 'too much to study'. The most frequent reason for students' choice of a science career was 'interest in science and science learning'. Judging from these results, the basic direction for science career education should be in raising the students' preference for and interest in science learning. The positive awareness for a career in science and vision for a science job in the future should be given through a science career education and science career education proper to the students' developmental stage and characteristics is more important.

• Journal of Advanced Navigation Technology
• /
• v.28 no.4
• /
• pp.559-565
• /
• 2024

Research on autonomous vehicles is being actively conducted. As autonomous vehicles emerge, there will be a transitional period in which traditional and autonomous vehicles coexist, potentially leading to a higher accident rate. Currently, when a traffic accident occurs, the fault ratio is determined according to the criteria set by the General Insurance Association of Korea. However, the time required to investigate the type of accident is substantial. Additionally, there is an increasing trend in fault ratio disputes, with requests for reconsideration even after the fault ratio has been determined. To reduce these temporal and material costs, we propose a deep learning model that automatically determines fault ratios. In this study, we aimed to determine fault ratios based on accident video through a image classification model based on ResNet-18 and video action recognition using TSN. If this model commercialized, could significantly reduce the time required to measure fault ratios. Moreover, it provides an objective metric for fault ratios that can be offered to the parties involved, potentially alleviating fault ratio disputes.