• Smart Media Journal
• /
• v.1 no.4
• /
• pp.59-71
• /
• 2012

Since mobile phones are used as common communication devices, their applications are increasingly important to human's life. Using smart-phones camera to collect daily life environment's information is one of targets for many applications such as text recognition, object recognition or context awareness. Studies have been conducted to provide important information through the recognition of texts, which are artificially or naturally included in images and movies acquired from mobile phones. In this study, a character segmentation method that improves character-recognition accuracy in images obtained from mobile phone cameras is proposed. The proposed method first classifies texts in a given image to printed letters and handwritten letters since segmentation approaches for them are different. For printed letters, rough segmentation process is conducted, then the segmented regions are integrated, deleted, and re-segmented. Segmentation for the handwritten letters is performed after skews are corrected and the characters are classified by integrating them. The experimental result shows our method achieves a successful performance for both printed and handwritten letters as 95.9% and 84.7%, respectively.

• International Journal of CAD/CAM
• /
• v.7 no.1
• /
• pp.31-40
• /
• 2007

Reverse engineering of reliefs aims to turn an existing relief superimposed on an underlying surface into a geometric model which may be applied to a different base surface. Steps in this process include segmenting the relief from the background, and describing it as an offset height field relative to the underlying surface. We have previously considered relief segmentation using a geometric snake. Here, we show how to use this initial segmentation to estimate the background surface lying under the relief, which can be used (i) to refine the segmentation and (ii) to express the relief as an offset field. Our approach fits a B-spline surface patch to the measured background data surrounding the relief, while tension terms ensure this background surface smoothly continues underneath the relief where there are no measured background data points to fit. After making an initial estimate of relief offset height everywhere within the patch, we use a support vector machine to refine the segmentation. Tests demonstrate that this approach can accurately model the background surface where it underlies the relief, providing more accurate segmentation, as well as relief height field estimation. In particular, this approach provides significant improvements for relief concavities with narrow mouths and can segment reliefs with small internal holes.

• Journal of Biomedical Engineering Research
• /
• v.42 no.5
• /
• pp.201-210
• /
• 2021

In Korean medicine, tongue diagnosis is one of the important diagnostic methods for diagnosing abnormalities in the body. Representative features that are used in the tongue diagnosis include color, shape, texture, cracks, and tooth marks. When diagnosing a patient through these features, the diagnosis criteria may be different for each oriental medical doctor, and even the same person may have different diagnosis results depending on time and work environment. In order to overcome this problem, recent studies to automate and standardize tongue diagnosis using machine learning are continuing and the basic process of such a machine learning-based tongue diagnosis system is tongue segmentation. In this paper, image data is augmented based on the main tongue features, and backbones of various famous deep learning architecture models are used for automatic tongue segmentation. The experimental results show that the proposed augmentation technique improves the accuracy of tongue segmentation, and that automatic tongue segmentation can be performed with a high accuracy of 99.12%.

• ETRI Journal
• /
• v.45 no.5
• /
• pp.822-835
• /
• 2023

Recent semantic segmentation frameworks usually combine low-level and high-level context information to achieve improved performance. In addition, postlevel context information is also considered. In this study, we present a Context ReFinement Network (CRFNet) and its training method to improve the semantic predictions of segmentation models of the encoder-decoder structure. Our study is based on postprocessing, which directly considers the relationship between spatially neighboring pixels of a label map, such as Markov and conditional random fields. CRFNet comprises two modules: a refiner and a combiner that, respectively, refine the context information from the output features of the conventional semantic segmentation network model and combine the refined features with the intermediate features from the decoding process of the segmentation model to produce the final output. To train CRFNet to refine the semantic predictions more accurately, we proposed a sequential training scheme. Using various backbone networks (ENet, ERFNet, and HyperSeg), we extensively evaluated our model on three large-scale, real-world datasets to demonstrate the effectiveness of our approach.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.8
• /
• pp.225-231
• /
• 2013

Medical image segmentation is an image processing technology prior to performing various medical image processing. Therefore, a variety of methods have been researched for fast and accurate medical image segmentation. Accurate judgment of segmentation region is needed to segment the interest region in which patient requested in medical image that various organs exist. However, an case that scanned a part of organs is small occurs. In this case, information to determine the segmentation region is lack. consequently, a removal of segmentation region occurs during the segmentation process. In this paper, we improved segmentation results in a small region using volume data and linear equation. In order to verify the performance of the proposed method, we segmented the lung region of chest CT images. As a result of experiments, we confirmed that image segmentation accuracy rose from 0.978 to 0.981 and standard deviation also improved from 0.281 to 0.187.

• The KIPS Transactions:PartB
• /
• v.12B no.3 s.99
• /
• pp.239-246
• /
• 2005

Biologist must have to do 2DGE biological experiment for Protein Search and Analysis. This experiment coming into being 2 dimensional image. 2DGE (2D Gel Electrophoresis : two dimensional gel electrophoresis) image is the most widely used method for isolating of the objective protein by comparative analysis of the protein spot pattern in the gel plane. The process of protein spot analysis, firstly segment protein spots that are spread in 2D gel plane by image processing and can find important protein spots through comparative analysis with protein pattern of contrast group. In the algorithm which detect protein spots, previous 2DGE image analysis is applies gaussian fitting, however recently Watersheds region based segmentation algorithm, which is based on morphological segmentation is applied. Watersheds has the benefit that segment rapidly needed field in big sized image, however has under-segmentation and over-segmentation of spot area when gray level is continuous. The drawback was somewhat solved by marker point institution, but needs the split and merge process. This paper introduces a novel marker search of protein spots by watersheds-based hierarchical threshold, which can resolve the problem of marker-driven watersheds.

• Journal of Sensor Science and Technology
• /
• v.6 no.2
• /
• pp.123-130
• /
• 1997

In this paper, we present an original stereo vision system which comprises two process: 1. An image segmentation algorithm based on new concept called declivity and using automatic thresholds. 2. A new stereo matching algorithm based on an optimal path search. This path is obtained by dynamic programming method which uses the threshold values calculated during the segmentation process. At present, a complete depth map of indoor scene only needs about 3 s on a Sun workstation IPX, and this time will be reduced to a few tenth of second on a specialised architecture based on several DSPs which is currently under consideration.

• Nuclear Engineering and Technology
• /
• v.56 no.5
• /
• pp.1733-1737
• /
• 2024

In order to specify the location of the scintillation pixel that interacted with gamma rays in the positron emission tomography (PET) detector, conventionally, after acquiring a flood image, the location of interaction between the scintillation pixel and gamma ray could be specified through a pixel-segmentation process. In this study, the experimentally acquired signal was specified as the location of the scintillation pixel directly, without any conversion process, through the simulation data and the deep learning algorithm. To evaluate the accuracy of the specification of the scintillation pixel location through deep learning, a comparative analysis with experimental data through pixel segmentation was performed. In the same way as in the experiment, a detector was configured on the simulation, a model was built using the acquired data through deep learning, and the location was specified by applying the experimental data to the built model. Accuracy was calculated through comparative analysis between the specified location and the location obtained through the segmentation process. As a result, it showed excellent accuracy of about 85 %. When this method is applied to a PET detector, the position of the scintillation pixel of the detector can be specified simply and conveniently, without additional work.

• Journal of Korea Multimedia Society
• /
• v.23 no.5
• /
• pp.633-640
• /
• 2020

Defect region segmentation of TFT-LCD images is performed by combining defect pixels detected by a defect detection method into defect region, or by using morphological operations to segment defect region. Therefore, the result of segmentation of the defect region is highly dependent on the defect detection result. In this paper, we propose a method which segments defect regions sequentially according to the possibility of being included in defect regions in TFT-LCD images. The proposed method repeats the process of detecting a seed using the median value and the median absolute deviation of the image, and segments the defect region using the seeded region growing method. We confirmed the superiority of the proposed method to segment defect regions using pseudo-images and real TFT-LCD images.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.22 no.11
• /
• pp.2531-2542
• /
• 1997

This paper describes a method for region segmentation and string extractionin documents which are mixed with text, graphic and picture images by the use of the structural characteristic of connceted components. In segmentation of non-text regionas, with connection-characteristic functions which are made by structural characteristic of connected components, segmentation process is progressed. In the string extraction, first we organize basic-unit-region of which vertical and horizontal length are 1/4 of average length of connection components. Second, by merging the basic-unit-regions one other that have smaller values than a given connection intensity threshold. Third, by linking the word blocks with similar block anagles, initial strings are cresed. Finally the whold strings are generated by merging remaining word blocks whose angles are not decided, if their height and prosition are similar to the initial strings. This method can extract strings that are neither horizontal nor of various character sizes. Through computer exteriments with different style documents, we have shown that the feasibility of our method successes.