• Korean Journal of Radiology
• /
• 제23권10호
• /
• pp.1009-1018
• /
• 2022

Objective: This study aimed to investigate the feasibility of using artificial intelligence (AI) to identify normal chest radiography (CXR) from the worklist of radiologists in a health-screening environment. Materials and Methods: This retrospective simulation study was conducted using the CXRs of 5887 adults (mean age ± standard deviation, 55.4 ± 11.8 years; male, 4329) from three health screening centers in South Korea using a commercial AI (Lunit INSIGHT CXR3, version 3.5.8.8). Three board-certified thoracic radiologists reviewed CXR images for referable thoracic abnormalities and grouped the images into those with visible referable abnormalities (identified as abnormal by at least one reader) and those with clearly visible referable abnormalities (identified as abnormal by at least two readers). With AI-based simulated exclusion of normal CXR images, the percentages of normal images sorted and abnormal images erroneously removed were analyzed. Additionally, in a random subsample of 480 patients, the ability to identify visible referable abnormalities was compared among AI-unassisted reading (i.e., all images read by human readers without AI), AI-assisted reading (i.e., all images read by human readers with AI assistance as concurrent readers), and reading with AI triage (i.e., human reading of only those rendered abnormal by AI). Results: Of 5887 CXR images, 405 (6.9%) and 227 (3.9%) contained visible and clearly visible abnormalities, respectively. With AI-based triage, 42.9% (2354/5482) of normal CXR images were removed at the cost of erroneous removal of 3.5% (14/405) and 1.8% (4/227) of CXR images with visible and clearly visible abnormalities, respectively. In the diagnostic performance study, AI triage removed 41.6% (188/452) of normal images from the worklist without missing visible abnormalities and increased the specificity for some readers without decreasing sensitivity. Conclusion: This study suggests the feasibility of sorting and removing normal CXRs using AI with a tailored cut-off to increase efficiency and reduce the workload of radiologists.

• 복식
• /
• 제61권2호
• /
• pp.131-151
• /
• 2011

The study alms to identify how the symbols and images of Greek Roman myths in contemporary fashion have been reflected in respects of meaning and forms, and to find out the organic course from meaning to forms by using Durand's classification. The results define the significance of Greek Roman myths in contemporary fashion, and systematically suggest a direction of imagination for more creative design. In the diurne regime, the symbolism of purity, heroism and fear appeared. In the nocturne regime, the symbolism of maternity and sensuality appeared. In the dramatic regime, the symbolism of androgyny appeared. The characteristics of designs contained in each symbolism are common. In this regard, it is possible to trace organic relationships in the creation of images through the verbal scheme. In addition, the verbal scheme creates archetypal images that lead to images and symbols in the socio-cultural context, so it is possible to analyze the relationships between archetypal images and the format of garments. The study examined how the archetypal images that appeared in the mythical images were expressed in garments through the verbal system.

• 치과방사선
• /
• 제23권2호
• /
• pp.335-344
• /
• 1993

The purpose of this study was to clarify the spatial relationship of temporomandibular joint and to an aid in the diagnosis of temporomandibular disorder. For this study, three-dimensional images of normal temporomandibular joints were reconstructed by computer image analysis system and three-dimensional reconstructive program integrated in computed tomography. The obtained results were as follows: 1. Two-dimensional computed tomograms had the better resolution than three dimensional computed tomograms in the evaluation of bone structure and the disk of TMJ. 2. Direct sagittal computed tomograms and coronal computed tomograms had the better resolution in the evaluation of the disk of TMJ. 3. The positional relationship of the disk could be visualized, but the configuration of the disk could not be clearly visualized on three-dimensional reconstructive CT images. 4. Three-dimensional reconstructive CT images had the smoother margin than three-dimensional images reconstructed by computer image analysis system, but the images of the latter had the better perspective. 5. Three-dimensional reconstructive images had the better spatial relationship of the TMJ articulation, and the joint spaces were more clearly visualized on dissection images.

• 한국멀티미디어학회논문지
• /
• 제24권12호
• /
• pp.1624-1631
• /
• 2021

With the continued development of image processing technology, we live in a time when it is difficult to visually discriminate processed (or tampered) images from real images. However, as the risk of fake images being misused for crime increases, the importance of image forensic science for identifying fake images is emerging. Currently, various deep learning-based identifiers have been studied, but there are still many problems to be used in real situations. Due to the inherent characteristics of deep learning that strongly relies on given training data, it is very vulnerable to evaluating data that has never been viewed. Therefore, we try to find a way to improve generalization ability of deep learning-based fake image identifiers. First, images with various contents were added to the training dataset to resolve the over-fitting problem that the identifier can only classify real and fake images with specific contents but fails for those with other contents. Next, color spaces other than RGB were exploited. That is, fake image identification was attempted on color spaces not considered when creating fake images, such as HSV and YCbCr. Finally, dropout, which is commonly used for generalization of neural networks, was used. Through experimental results, it has been confirmed that the color space conversion to HSV is the best solution and its combination with the approach of increasing the training dataset significantly can greatly improve the accuracy and generalization ability of deep learning-based identifiers in identifying fake images that have never been seen before.

• 대한원격탐사학회지
• /
• 제39권4호
• /
• pp.395-407
• /
• 2023

Super-resolution (SR) has great significance in image processing because it enables downstream vision tasks with high spatial resolution. Recently, SR studies have adopted deep learning networks and achieved remarkable SR performance compared to conventional example-based methods. Deep-learning-based SR models generally require low-resolution (LR) images and the corresponding high-resolution (HR) images as training dataset. Due to the difficulties in obtaining real-world LR-HR datasets, most SR models have used only HR images and generated LR images with predefined degradation such as bicubic downsampling. However, SR models trained on simple image degradation do not reflect the properties of the images and often result in deteriorated SR qualities when applied to real-world images. In this study, we propose an image degradation model for HR satellite images based on the modulation transfer function (MTF) of an imaging sensor. Because the proposed method determines the image degradation based on the sensor properties, it is more suitable for training SR models on remote sensing images. Experimental results on HR satellite image datasets demonstrated the effectiveness of applying MTF-based filters to construct a more realistic LR-HR training dataset.

• 복식
• /
• 제57권9호
• /
• pp.76-87
• /
• 2007

Since the emergence of postmodernism, as interest in human has increased, human face image is being highlighted as one of the themes that are addressed the most. Making the images of faces and individuals shown in costumes the subject matter, this thesis examines the internal values immanent in the images in more depth and understands them based on the Panofsky's iconological interpretation scheme. This study aims to identify designer's purposes and even their unconscious intention through iconological interpretation of faces shown in the fashion and images shown in human image, and further to present basic materials in the fashion design. This research used literature reviews and case studies, and used Panofsky's iconological interpretation theory as the scheme in order to interpret the symbolic significance implied in the images. The images of faces and individuals shown in costumes were classified into six types through historical reviews, and based on the types the images of faces and individuals shown in the fashion since the 20th century were examined. The iconological analysis of the images of faces and individuals shown in costumes based on the classification of types according to historical reviews showed parodies, cultural identity, commercial use, eroticism, respect for heros and its fiction. This study has found that all such things finally return to humanism that humans should be valued and loved the most.

• Imaging Science in Dentistry
• /
• 제34권4호
• /
• pp.185-190
• /
• 2004

Purpose: To compare the diagnostic accuracy for the detection of root fractures in CMOS-based digital periapical images with conventional film-based periapical images. Materials and Methods: Sixty extracted single-root human teeth with closed apices were prepared endodontically and divided into two groups; artificially induced vertical root fracture group and control group. All radiographs were obtained using the paralleling technique. The radiographs were examined by 4 observers three times within a 4 week interval. Receiver operating characteristic (ROC) analysis was carried out using data obtained from four observers. Intra- and inter-examiner agreements were computed using kappa analysis. Results: The area under the ROC curve (Az) was used as an indicator of the diagnostic accuracy of the imaging system. Az values were as follows: direct-digital images; 0.93, film-based images; 0.92, and inverted digital images; 0.91. There was no significant difference between imaging modalities (P<0.05). The kappa value of inter-observer agreement was 0.42 (range: 0.28-0.60) and intra-observer agreement was 0.57 (range: 0.44-0.75). Conclusion : There is no statistical difference in diagnostic accuracy for the detection of vertical root fractures between digital periapical images and conventional periapical images. The results indicate that the CMOS sensor is a good image detector for the evaluation of vertical root fractures.

• Journal of Animal Science and Technology
• /
• 제65권3호
• /
• pp.638-651
• /
• 2023

The objective of this study was to quantitatively estimate the level of grazing area damage in outdoor free-range pig production using a Unmanned Aerial Vehicles (UAV) with an RGB image sensor. Ten corn field images were captured by a UAV over approximately two weeks, during which gestating sows were allowed to graze freely on the corn field measuring 100 × 50 m². The images were corrected to a bird's-eye view, and then divided into 32 segments and sequentially inputted into the YOLOv4 detector to detect the corn images according to their condition. The 43 raw training images selected randomly out of 320 segmented images were flipped to create 86 images, and then these images were further augmented by rotating them in 5-degree increments to create a total of 6,192 images. The increased 6,192 images are further augmented by applying three random color transformations to each image, resulting in 24,768 datasets. The occupancy rate of corn in the field was estimated efficiently using You Only Look Once (YOLO). As of the first day of observation (day 2), it was evident that almost all the corn had disappeared by the ninth day. When grazing 20 sows in a 50 × 100 m² cornfield (250 m²/sow), it appears that the animals should be rotated to other grazing areas to protect the cover crop after at least five days. In agricultural technology, most of the research using machine and deep learning is related to the detection of fruits and pests, and research on other application fields is needed. In addition, large-scale image data collected by experts in the field are required as training data to apply deep learning. If the data required for deep learning is insufficient, a large number of data augmentation is required.

• 한국멀티미디어학회논문지
• /
• 제7권12호
• /
• pp.1657-1664
• /
• 2004

We propose a method that classifies images into two object types man-made and natural objects. A central object is extracted from each image by using central object extraction method[1] before classification. A central object in an images defined as a set of regions that lies around center of the image and has significant color distribution against its surrounding. We define three measures to classify the object images. The first measure is energy of edge direction histogram. The energy is calculated based on the direction of only non-circular edges. The second measure is an energy difference along directions in Gabor filter dictionary. Maximum and minimum energy along directions in Gabor filter dictionary are selected and the energy difference is computed as the ratio of the maximum to the minimum value. The last one is a shape of an object, which is also represented by Gabor filter dictionary. Gabor filter dictionary for the shape of an object differs from the one for the texture in an object in which the former is computed from a binarized object image. Each measure is combined by using majority rule tin which decisions are made by the majority. A test with 600 images shows a classification accuracy of 86%.

• 치과방사선
• /
• 제24권2호
• /
• pp.305-316
• /
• 1994

The comparative detectability of the artificial defects among conventional radiographs, digital images and digital subtraction images was evaluated. The artificial defects were made within spogy bone of 24 unilateral mandibles of adult dogs. The results were as follows: 1. With normal exposure time, the detectability of digital subtraction radiographs was 90.3% which was statistically significant superior to those of conventional radiographs(78.0%) and digital images(75.9%) (p<0.05). 2. With half-exposure time, the detectability of conventional radiographs, digital images and digital subtraction radiographs was 68.4%, 67.3% and 69.9% respectively. There was no statistical significant difference among the detectability of these methods(p>0.05). 3. All radiographic images with normal exposure time showed statistically significant superior detectability to those with half-exposure time(p<0.05). 4. The detectability of digital subtraction radiographs was not linearly related to the standard deviation of the grey levels of reference line(p<0.05).