• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.374-376
• /
• 2021

For the diagnosis of cancer-related diseases in clinical practice, pathological examination using biopsy is essential after basic diagnosis using imaging equipment. In order to proceed with such a biopsy, the assistance of an oncologist, clinical pathologist, etc. with specialized knowledge and the minimum required time are essential for confirmation. In recent years, research related to the establishment of a system capable of automatic classification of cancer cells using artificial intelligence is being actively conducted. However, previous studies show limitations in the type and accuracy of cells based on a limited algorithm. In this study, we propose a method to identify a total of 4 cancer cells through a convolutional neural network, a kind of deep learning. The optical images obtained through cell culture were learned through EfficientNet after performing pre-processing such as identification of the location of cells and image segmentation using OpenCV. The model used various hyper parameters based on EfficientNet, and trained InceptionV3 to compare and analyze the performance. As a result, cells were classified with a high accuracy of 96.8%, and this analysis method is expected to be helpful in confirming cancer.

• The KIPS Transactions:PartB
• /
• v.19B no.2
• /
• pp.123-126
• /
• 2012

This paper proposed a method to separate a liver into left and right liver lobes for simple and exact volumetry of the river graft at abdominal MDCT(Multi-Detector Computed Tomography) image before the living donor liver transplantation. A medical team can evaluate an accurate river graft with minimized interaction between the team and a system using this algorithm for ensuring donor's and recipient's safe. On the image of segmented liver, 2 points(PMHV: a point in Middle Hepatic Vein and PPV: a point at the beginning of right branch of Portal Vein) are selected to separate a liver into left and right liver lobes. Middle hepatic vein is automatically segmented using PMHV, and the cutting line is decided on the basis of segmented Middle Hepatic Vein. A liver is separated on connecting the cutting line and PPV. The volume and ratio of the river graft are estimated. The volume estimated using 2 points are compared with a manual volume that diagnostic radiologist processed and estimated and the weight measured during surgery to support proof of exact volume. The mean ${\pm}$ standard deviation of the differences between the actual weights and the estimated volumes was $162.38cm^3{\pm}124.39$ in the case of manual segmentation and $107.69cm^3{\pm}97.24$ in the case of 2 points method. The correlation coefficient between the actual weight and the manually estimated volume is 0.79, and the correlation coefficient between the actual weight and the volume estimated using 2 points is 0.87. After selection the 2 points, the time involved in separation a liver into left and right river lobe and volumetry of them is measured for confirmation that the algorithm can be used on real time during surgery. The mean ${\pm}$ standard deviation of the process time is $57.28sec{\pm}32.81$ per 1 data set ($149.17pages{\pm}55.92$).

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.12
• /
• pp.431-438
• /
• 2020

In this paper, the method to extract lip contour by multiple threshold is proposed. Spyridonos et. el. proposed a method to extract lip contour. First step is get Q image from transform of RGB into YIQ. Second step is to find lip corner points by change point detection and split Q image into upper and lower part by corner points. The candidate lip contour can be obtained by apply threshold to Q image. From the candidate contour, feature variance is calculated and the contour with maximum variance is adopted as final contour. The feature variance 'D' is based on the absolute difference near the contour points. The conventional method has 3 problems. The first one is related to lip corner point. Calculation of variance depends on much skin pixels and therefore the accuracy decreases and have effect on the split for Q image. Second, there is no analysis for color systems except YIQ. YIQ is a good however, other color systems such as HVS, CIELUV, YCrCb would be considered. Final problem is related to selection of optimal contour. In selection process, they used maximum of average feature variance for the pixels near the contour points. The maximum of variance causes reduction of extracted contour compared to ground contours. To solve the first problem, the proposed method excludes some of skin pixels and got 30% performance increase. For the second problem, HSV, CIELUV, YCrCb coordinate systems are tested and found there is no relation between the conventional method and dependency to color systems. For the final problem, maximum of total sum for the feature variance is adopted rather than the maximum of average feature variance and got 46% performance increase. By combine all the solutions, the proposed method gives 2 times in accuracy and stability than conventional method.

• Journal of Korea Water Resources Association
• /
• v.54 no.8
• /
• pp.553-566
• /
• 2021

River confluence is often characterized by shear layer and the associated strong mixing. In natural rivers, the main channel and its tributary can be separated by the shear layer using contrasting colors. The shear layer can be easily observed using aerial images from satellite or unmanned aerial vehicles. This study proposes a low-cost identification method extracting geographic features of the shear layer using RGB aerial image. The method consists of three stages. At first, in order to identify the shear layer, it performs image segmentation using a Gaussian mixture model and extracts the water bodies of the main channel and tributary. Next, the self-organizing map simplifies the flow line of the water bodies into the 1-dimensional curve grid. After that, the curvilinear coordinate transformation is performed using the water body pixels and the curve grid. As a result, the shear layer identification method was successfully applied to the confluence between Nakdong River and Nam River to extract geometric shear layer features (confluence angle, upstream- and downstream- channel widths, shear layer length, maximum shear layer thickness).

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.43 no.1 s.307
• /
• pp.9-16
• /
• 2006

This Paper proposes an adaptive error diffusioThis paper proposes an adaptive error diffusion algorithm for text enhancement followed by an efficient text segmentation that uses the maximum gradient difference (MGD). The gradients are calculated along with scan lines, and the MGD values are filled within a local window to merge the potential text segments. Isolated segments are then eliminated in the non-text region filtering process. After the left segmentation, a conventional error diffusion method is applied to the background, while the edge enhancement error diffusion is used for the text. Since it is inevitable that visually objectionable artifacts are generated when using two different halftoning algorithms, the gradual dilation is proposed to minimize the boundary artifacts in the segmented text blocks before halftoning. Sharpening based on the gradually dilated text region (GDTR) prevents the printing of successive dots around the text region boundaries. The error diffusion algorithm with edge enhancement is extended to halftone color images to sharpen the tort regions. The proposed adaptive error diffusion algorithm involves color halftoning that controls the amount of edge enhancement using a general error filter. The multiplicative edge enhancement parameters are selected based on the amount of edge sharpening and color difference. Plus, the additional error factor is introduced to reduce the dot elimination artifact generated by the edge enhancement error diffusion. By using the proposed algorithm, the text of a scanned image is sharper than that with a conventional error diffusion without changing background.

• Journal of radiological science and technology
• /
• v.35 no.2
• /
• pp.133-139
• /
• 2012

For the examination of nuclear medicine, myocardial scan is a good method to evaluate a hemodynamic importance of coronary heart disease. but, the automatized qualitative measurement is additionally necessary to improve the decoding efficiency. we suggests the creation of cardiac three-dimensional model and model of three-dimensional cardiac thickness as a new measurement. For the experiment, cardiac reduced cross section was obtained from SPECT. Next, the pre-process was performed and image segmentation was fulfilled by level set. for the modeling of left cardiac thickness, it was realized by applying difference equation of two-dimensional laplace equation. As the result of experiment, it was successful to measure internal wall and external wall and three-dimensional modeling was realized by coordinate. and, with laplace formula, it was successful to develop the thickness of cardiac wall. through the three-dimensional model, defects were observed easily and position of lesion was grasped rapidly by the revolution of model. The model which was developed as the support index of decoding will provide decoding information to doctor additionally and reduce the rate of false diagnosis as well as play a great role for diagnosing IHD early.

• 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.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.23 no.5
• /
• pp.261-273
• /
• 2024

Automobiles are one of the major sources of air pollution, and analyzing data on air pollutants, where vehicles are the primary pollutants, can help elucidate the correlation between factors like electric vehicles, traffic volume, and actual air pollution. Ensuring the reliability of air pollutant data is crucial for such analyses. This paper proposes a method for detecting sections of data exhibiting 'baseline anomalies' measured at air pollutant monitoring stations across the country by combining deep learning models with algorithms such as dynamic time warping and change point detection. While previous studies have focused on detecting data with unprecedented patterns and defined them as anomalies, this approach was not suitable for detecting baseline anomalies. In this study, we modify the U-Net model, typically used for image segmentation, to be more suitable for time-series data and apply dynamic time warping and change point detection algorithms to compare with nearby monitoring stations, thereby minimizing false detections.

• Journal of KIISE:Software and Applications
• /
• v.31 no.3
• /
• pp.379-386
• /
• 2004

Video segmentation is a fundamental task in video indexing and it includes two kinds of shot change detections such as the abrupt transition and the gradual transition. The abrupt shot boundaries are detected by computing the image-based distance between adjacent frames and comparing this distance with a pre-determined threshold value. However, the gradual shot boundaries are difficult to detect with this approach. To overcome this difficulty, we propose the method that detects gradual transition in the MPEG compressed video using the HMM (Hidden Markov Model). We take two different HMMs such as a discrete HMM and a continuous HMM with a Gaussian mixture model. As image features for HMM's observations, we use two distinct features such as the difference of histogram of DC images between two adjacent frames and the difference of each individual macroblock's deviations at the corresponding macroblock's between two adjacent frames, where deviation means an arithmetic difference of each macroblock's DC value from the mean of DC values in the given frame. Furthermore, we obtain the DC sequences of P and B frame by the first order approximation for a fast and effective computation. Experiment results show that we obtain the best detection and classification performance of gradual transitions when a continuous HMM with one Gaussian model is taken and two image features are used together.

• Journal of Korean Society of Forest Science
• /
• v.96 no.1
• /
• pp.107-114
• /
• 2007

Establishments of the seedlings inside the natural forest from adjacent artificial forests would be an important factor in forest stand dynamics. This study was conducted to see the invasion of Korean pine (Pinus koraiensis) seedlings which is not native in this region, into the natural deciduous broad-leaved forest in Gwangneung, Korea. There is no mother tree at the I ha study site while the number of naturally regenerated P. koraiensis seedlings was 345 trees and 56% of them were clumped with more than two seedlings at each point. Applying the image segmentation method to IKONOS satellite image of January, 2003, the distance from the center of 1 ha study site to the nearest mother tree and plantation of Korean pine were 200 m and 270 m, respectively. The average height and root-collar diameter of the seedlings were 34 em and 7 mm, respectively and the age of 207 seedlings (60%) were below 5 years old. Most abundant range of soil moisture gradient and LAl (leaf area index) were from 16 to 20% and those of LAI were from 3.1 to 3.5. To understand the dynamics and seed dispersal pattern of Korean pine in the Gwangneung natural deciduous broad-leaved forests, additional studies not only long-term monitoring of growth and mortality of naturally regenerated Korean pine seedlings but also application of stable isotope analysis and molecular genetic techniques was recommended.