• Clinical Endoscopy
• /
• v.55 no.3
• /
• pp.390-400
• /
• 2022

Background/Aims: Previous artificial intelligence (AI) models attempting to segment gastric intestinal metaplasia (GIM) areas have failed to be deployed in real-time endoscopy due to their slow inference speeds. Here, we propose a new GIM segmentation AI model with inference speeds faster than 25 frames per second that maintains a high level of accuracy. Methods: Investigators from Chulalongkorn University obtained 802 histological-proven GIM images for AI model training. Four strategies were proposed to improve the model accuracy. First, transfer learning was employed to the public colon datasets. Second, an image preprocessing technique contrast-limited adaptive histogram equalization was employed to produce clearer GIM areas. Third, data augmentation was applied for a more robust model. Lastly, the bilateral segmentation network model was applied to segment GIM areas in real time. The results were analyzed using different validity values. Results: From the internal test, our AI model achieved an inference speed of 31.53 frames per second. GIM detection showed sensitivity, specificity, positive predictive, negative predictive, accuracy, and mean intersection over union in GIM segmentation values of 93%, 80%, 82%, 92%, 87%, and 57%, respectively. Conclusions: The bilateral segmentation network combined with transfer learning, contrast-limited adaptive histogram equalization, and data augmentation can provide high sensitivity and good accuracy for GIM detection and segmentation.

• Journal of Biomedical Engineering Research
• /
• v.28 no.3
• /
• pp.332-337
• /
• 2007

In the early twenty-first century, minimally invasive surgery is the mainstay of various kinds of surgical fields. Surgeons gave percutaneously surgical treatment of the screw directly using a fluoroscopic view in the past. The latest date, they began to operate the fractured carpal bone surgery using Computerized Tomography (CT). Carpal bones composed of wrist joint consist of eight small bones which have hexahedron and sponge shape. Because of these shape, it is difficult to grasp the shape of carpal bones using only CT image data. Although several image segmentation studies have been conducted with carpal bone CT image data, more studies about carpal bone using CT data are still required. Especially, to apply the software implemented from the studies to clinical fIeld, the outcomes should be user friendly and very accurate. To satisfy those conditions, we propose modified multi-seed region growing segmentation method which uses simple threshold and the canny edge detector for finding edge information more accurately. This method is able to use very easily and gives us high accuracy and high speed for extracting the edge information of carpal bones. Especially, using multi-seed points, multi-bone objects of the carpal bone are extracted simultaneously.

• The KIPS Transactions:PartB
• /
• v.8B no.5
• /
• pp.493-500
• /
• 2001

Video scene segmentation is fundamental role for content based video analysis. In this paper, we propose a new region based video scene segmentation method using continuity test for each object region which is segmented by the watershed algorithm for all frames in video data. For this purpose, we first classify video data segments into classes that are the dynamic and static sections according to the object movement rate by comparing the spatial and shape similarity of each region. And then, try to segment each sections by grouping each sections by comparing the neighbor section sections by comparing the neighbor section similarity. Because, this method uses the region which represented on object as a similarity measure, it can segment video scenes efficiently without undesirable fault alarms by illumination and partial changes.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.221-235
• /
• 2022

Manual inspection of steel box girders on long span bridges is time-consuming and labor-intensive. The quality of inspection relies on the subjective judgements of the inspectors. This study proposes an automated approach to detect and segment cracks in high-resolution images. An end-to-end cascaded framework is proposed to first detect the existence of cracks using a deep convolutional neural network (CNN) and then segment the crack using a modified U-Net encoder-decoder architecture. A Naïve Bayes data fusion scheme is proposed to reduce the false positives and false negatives effectively. To generate the binary crack mask, first, the original images are divided into 448 × 448 overlapping image patches where these image patches are classified as cracks versus non-cracks using a deep CNN. Next, a modified U-Net is trained from scratch using only the crack patches for segmentation. A customized loss function that consists of binary cross entropy loss and the Dice loss is introduced to enhance the segmentation performance. Additionally, a Naïve Bayes fusion strategy is employed to integrate the crack score maps from different overlapping crack patches and to decide whether a pixel is crack or not. Comprehensive experiments have demonstrated that the proposed approach achieves an 81.71% mean intersection over union (mIoU) score across 5 different training/test splits, which is 7.29% higher than the baseline reference implemented with the original U-Net.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.43 no.1
• /
• pp.16-25
• /
• 2020

Vegetation segmentation in a field color image is a process of distinguishing vegetation objects of interests like crops and weeds from a background of soil and/or other residues. The performance of the process is crucial in automatic precision agriculture which includes weed control and crop status monitoring. To facilitate the segmentation, color indices have predominantly been used to transform the color image into its gray-scale image. A thresholding technique like the Otsu method is then applied to distinguish vegetation parts from the background. An obvious demerit of the thresholding based segmentation will be that classification of each pixel into vegetation or background is carried out solely by using the color feature of the pixel itself without taking into account color features of its neighboring pixels. This paper presents a new pixel-based segmentation method which employs a multi-layer perceptron neural network to classify the gray-scale image into vegetation and nonvegetation pixels. The input data of the neural network for each pixel are 2-dimensional gray-level values surrounding the pixel. To generate a gray-scale image from a raw RGB color image, a well-known color index called Excess Green minus Excess Red Index was used. Experimental results using 80 field images of 4 vegetation species demonstrate the superiority of the neural network to existing threshold-based segmentation methods in terms of accuracy, precision, recall, and harmonic mean.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.11
• /
• pp.585-592
• /
• 2016

A topic segmentation module is to divide statements or conversations into certain topic units. Until now, topic segmentation has progressed in the direction of finding an optimized set of segments for a whole document, considering it all together. However, some applications need topic segmentation for a part of document which is not finished yet. In this paper, we propose a model to perform topic segmentation during the progress of the statement with a supervised learning model that uses a convolution neural network. In order to show the effectiveness of our model, we perform experiments of topic segmentation both on-line status and off-line status using C99 algorithm. We can see that our model achieves 17.8 and 11.95 of Pk score, respectively.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.30 no.6_2
• /
• pp.643-651
• /
• 2012

Object recognition belongs to high-level processing that is one of the difficult and challenging tasks in computer vision. Digital photogrammetry based on the computer vision paradigm has begun to emerge in the middle of 1980s. However, the ultimate goal of digital photogrammetry - intelligent and autonomous processing of surface reconstruction - is not achieved yet. Object recognition requires a robust shape description about objects. However, most of the shape descriptors aim to apply 2D space for image data. Therefore, such descriptors have to be extended to deal with 3D data such as LiDAR(Light Detection and Ranging) data obtained from ALS(Airborne Laser Scanner) system. This paper introduces extension of chain code to 3D object space with hierarchical approach for segmenting point cloud data. The experiment demonstrates effectiveness and robustness of the proposed method for shape description and point cloud data segmentation. Geometric characteristics of various roof types are well described that will be eventually base for the object modeling. Segmentation accuracy of the simulated data was evaluated by measuring coordinates of the corners on the segmented patch boundaries. The overall RMSE(Root Mean Square Error) is equivalent to the average distance between points, i.e., GSD(Ground Sampling Distance).

• Journal of the Korea Convergence Society
• /
• v.6 no.1
• /
• pp.7-16
• /
• 2015

A segmentation process extracts an interesting area of barcode in an image and gives a crucial impart on the performance of barcode verifier. Previous segmentation methods occurs some issues as follows. First, it is very hard to determine a threshold of length in Hough Line transform because it is sensitive. Second, Morphology transform delays the process when you conduct dilation and erosion operations during the image extraction. Therefore, we proposes a novel Converged Harris Corner detection-based segmentation method to detect an interesting area of barcode in Data Matrix. In order to evaluate the performance of proposed method, we conduct experiments by a dataset of barcode in accordance with size and location in an image. In result, our method solves the problems of delay and surrounding environments, threshold setting, and extracts the barcode area 100% from test images.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.18 no.2
• /
• pp.183-203
• /
• 1993

One of the most difficult and time-consuming stages in the development of the knowledge-based system is a knowledge acquisition. A splitting algorithm is developed to infer a rule-tree which can be converted to a rule-typed knowledge. A market segmentation may be performed in order to establish market strategy suitable to each market segment. As the sales data of a product market is probabilistic and noisy, it becomes necessary to prune the rule-tree-at an acceptable level while generating a rule-tree. A splitting algorithm is developed using the pruning measure based on a total amount of information gain and the measure of existing algorithms. A user can easily adjust the size of the resulting rule-tree according to his(her) preferences and problem domains. The algorithm is applied to a market segmentation problem of a medium-large computer market. The algorithm is illustrated step by step with a sales data of a computer market and is analyzed.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.26 no.4
• /
• pp.333-342
• /
• 2022

In this paper, we perform a direct comparison study of real experimental data for domain rearrangement and the Cahn-Hilliard (CH) equation on the dynamics of morphological evolution. To validate a mathematical model for physical phenomena, we take initial conditions from experimental images by using an image segmentation technique. The image segmentation algorithm is based on the Mumford-Shah functional and the Allen-Cahn (AC) equation. The segmented phase-field profile is similar to the solution of the CH equation, that is, it has hyperbolic tangent profile across interfacial transition region. We use unconditionally stable schemes to solve the governing equations. As a test problem, we take domain rearrangement of lipid bilayers. Numerical results demonstrate that comparison of the evolutions with experimental data is a good benchmark test for validating a mathematical model.