• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.3
• /
• pp.701-720
• /
• 2023

Aerial objects are more challenging to segment than normal objects, which are usually smaller and have less textural detail. In the process of segmentation, target objects are easily omitted and misdetected, which is problematic. To alleviate these issues, we propose local aggregation feature pyramid networks (LAFPNs) and pyramid integrated context modules (PICMs) for aerial object segmentation. First, using an LAFPN, while strengthening the deep features, the extent to which low-level features interfere with high-level features is reduced, and numerous dense and small aerial targets are prevented from being mistakenly detected as a whole. Second, the PICM uses global information to guide local features, which enhances the network's comprehensive understanding of an entire image and reduces the missed detection of small aerial objects due to insufficient texture information. We evaluate our network with the MS COCO dataset using three categories: airplanes, birds, and kites. Compared with Mask R-CNN, our network achieves performance improvements of 1.7%, 4.9%, and 7.7% in terms of the AP metrics for the three categories. Without pretraining or any postprocessing, the segmentation performance of our network for aerial objects is superior to that of several recent methods based on classic algorithms.

• Journal of the Korean Society of Radiology
• /
• v.82 no.1
• /
• pp.29-48
• /
• 2021

Diffusion-weighted magnetic resonance imaging (DW MRI) is a fast unenhanced technique that shows promise as a stand-alone modality for cancer screening and characterization. Currently, DW MRI may have lower sensitivity than that of dynamic contrast-enhanced MRI as a standalone modality for breast cancer detection but superior to that of mammography, which may provide a useful alternative for supplemental screening. Standardized acquisition and interpretation of DW MRI can improve the image quality and reduce the variability of the results. Furthermore, high-resolution DW MRI, with advanced techniques and postprocessing, will facilitate better detection and characterization of subcentimeter cancers and reduce false-negatives and false-positives. Future results from ongoing prospective multicenter clinical trials using standardized and optimized protocols will facilitate the use of DW MRI as a stand-alone modality.

• Investigative Magnetic Resonance Imaging
• /
• v.8 no.2
• /
• pp.94-99
• /
• 2004

Purpose : To investigate the signal enhancement ratio by NOE effect on in vivo $^{31}P$ MRS in human heart muscle and liver. we also evaluated the enhancement ratios of different phosphorus metabolites, which are important in 31P MRS for each organ. Materials and Methods : Ten normal subjects (M:F = 8:2, age range = 24-32 yrs) were included for in vivo $^{31}P$ MRS measurements on a 1.5 T whole-body MRI/MRS system using $^1H-^{31}P$ dual tuned surface coil. Two-dimensional Chemical Shift Imaging (2D CSI) pulse sequence for $^{31}P$ MRS was employed in all $^{31}P$ MRS measurements. First, $^{31}P$ MRS performed without NOE effect and then the same 2D CSI data acquisitions were repeated with NOE effect. After postprocessing the MRS raw data in the time domain, the signal enhancements in percent were estimated from the major metabolites. Results : The calculated NOE enhancement for liver $^{31}P$ MRS were $\alpha-ATP\;(7\%),\;\beta-ATP\;(9\%),\;\gamma-ATP\;(17\%),\;Pi\;(1\%),\;PDE\;(19\%)$ and $PME\;(31\%)$. Because there is no creatine kinase activity in liver, PCr signal is absent. For cardiac $^{31}P$ MRS, whole body coil gave better scout images and thus better localization than surface coil. In $^{31}P$cardiac multi-voxel spectra, DPG signal increased from left to right according to the amount of blood included. The calculated enhancement for cardiac $^{31}P$ MRS were : $\alpha-ATP\;(12\%),\;\beta-ATP\;(19\%),\;\gamma-ATP\;(30\%),\;PCr\;(34\%),\;Pi\;(20\%),\;(PDE)\;(51\%),\;and\;DPG\;(72\%)$. Conclusion : Our results revealed that the NOE effect was more pronounced in heart muscle than in liver with different coupling to 1H spin system and thus different heteronuclear cross-relaxation.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.39 no.3
• /
• pp.58-65
• /
• 2002

In order to reduce blocking artifacts in inter-frame coded images, we propose a new image restoration algorithm, which directly processes differential images before reconstruction. We note that blocking artifact in inter-frame coded images is caused by both 8$\times$8 DCT and 16$\times$16 macroblock based motion compensation, while that of intra-coded images is caused by 8$\times$8 DCT only. According to the observation, we Propose a new degradation model for differential images and the corresponding restoration algorithm that utilizes additional constraints and convex sets for discontinuity inside blocks. The proposed restoration algorithm is a modified version of standard regularization that incorporate!; spatially adaptive lowpass filtering with consideration of edge directions by utilizing a part of DCT coefficients. Most of video coding standard adopt a hybrid structure of block-based motion compensation and block discrete cosine transform (BDCT). By this reason, blocking artifacts are occurred on both block boundary and block interior For more complete removal of both kinds of blocking artifacts, the restored differential image must satisfy two constraints, such as, directional discontinuities on block boundary and block interior Those constraints have been used for defining convex sets for restoring differential images.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.3 no.2 s.5
• /
• pp.19-28
• /
• 2004

This paper shows how to compensate the skew from the traffic sign included in the natural image and extract the text. The research deals with the Process related to the array image. Ail the process comprises four steps. In the first fart we Perform the preprocessing and Canny edge extraction for the edge in the natural image. In the second pan we perform preprocessing and postprocessing for Hough Transform in order to extract the skewed angle. In the third part we remove the noise images and the complex lines, and then extract the candidate region using the features of the text. In the last part after performing the local binarization in the extracted candidate region, we demonstrate the text extraction by using the differences of the features which appeared between the tett and the non-text in order to select the unnecessary non-text. After carrying out an experiment with the natural image of 100 Pieces that includes the traffic sign. The research indicates a 82.54 percent extraction of the text and a 79.69 percent accuracy of the extraction, and this improved more accurate text extraction in comparison with the existing works such as the method using RLS(Run Length Smoothing) or Fourier Transform. Also this research shows a 94.5 percent extraction in respect of the extraction on the skewed angle. That improved a 26 percent, compared with the way used only Hough Transform. The research is applied to giving the information of the location regarding the walking aid system for the blind or the operation of a driverless vehicle

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.37 no.2
• /
• pp.50-59
• /
• 2000

In this paper, a new preprocessing method is proposed to speedup the watershed-based image segmentation In the proposed method, the gradient correction values of ramp edges are calculated from the positions and width of the ramp edges using Laplacian operator, and then, unlike the conventional method in which the monoscale or multi scale gradient image is directly used as a reference iImage, the reference image is obtained by adding the threshold value to each position of the ramp edges in the monoscale gradient image And the marker image is reconstructed on the reference image by erosion By preprocessing the image for the watershed transformation in such a manner, we can reduce the oversegmentations far more than those of applying the conventional morphological filter to the simple monoscale or multiscale gradient-based reference image Thus, we can reduce the total image segmentation time by reducing the time of postprocessing of region merging, which consumes most of the processing time In the watershed-based image segmentation, Experimental results indicate that the proposed method can speedup the total image segmentation about twice than those of the conventional methods, without the loss of ramp edges and principal edges around the dense-edge region.

• The Journal of Korean Academy of Prosthodontics
• /
• v.31 no.4
• /
• pp.580-610
• /
• 1993

The purpose of this study was to analyse the deflection and stress distribution at the supporting bone and it's superstructure by the alteration of angulation between implant and it's implant abutment. For this study, the free-end saddle case of mandibular first and second molar missing would be planned to restore with fixed prosthesis. So the mandibular second premolar was prepared for abutment, and the cylinder type osseointegrated implant was placed at the site of mandibular second molar for abutment. The finite element stress analysis was applied for this study. 13 two-dimensional FEM models were created, a standard model at $0^{\circ}$ and 12 models created by changing the angulation between implant and implant abutment as increasing the angulation mesially and distally with $5^{\circ}$ unittill $30^{\circ}$. The preprocessing decording, solving and postprocessing procedures were done by using FEM analysis software PATRAN and SUN-SPARC2GX. The deflections and von Mises stresses were calculated under concentrated load (load 1) and distributed load(load 2) at the reference points. The results were as follows : 1. Observing at standard model, the amount of total deflection at the distobuccal cusp-tip of pontic under concentrated load was largest of all, and that at the apex of implant was least of all, and the amount of total deflection at the buccal cusp-tip of second premolar under distributed load was largest of all, and that at the apex of implant was least of all. 2. Increasing the angulation mesially or distally, the amounts of total deflection were increased or decreased according to the reference points. But the order according to the amount of total deflection was not changed except apex of second premolar and central fossa of implant abutment under concentrated load during distal inclination. 3. Observing at standard model, the von Mises stress at the distal joint of pontic under concentrated load was largest of all, and that at the apex of implant was least of all. The von Mises stress at the distal margin of second premolar under distributed load was largest of all, and that at the apex of Implant was least of ail. 4. Increasing the angulation of implant mesially, the von Mises stresses at the mesial crest of implant were increased under concentrated load and distributed load, but those were increased remarkably under distributed load and so that at $30^{\circ}$ mesial inclination was largest of all. 5. Increasing the angulation of implant distally, the von Mises stresses at the distal crest of implant were increased remarkably under concentrated load and distributed load, and so those at $30^{\circ}$ distal inclination were largest of all.

• Investigative Magnetic Resonance Imaging
• /
• v.3 no.1
• /
• pp.53-59
• /
• 1999

Purpose : To evaluate the usefulness of functional MR imaging (fMRI) for language mapping and determination of language lateralization. Materials and Methods : Functional maps of the language area were obtained during word generation tasks and decision task in ten volunteers (7 right handed, 3 left-handed). MR examinations were performed at 1.5T scanner with EPI BOLD technique. Each task consisted of three resting periods and two activation periods with each period of 30 seconds. Total acquisition time was 162 sec. SPM program was used for the postprocessing of images. Statistical comparisons were performed by using t-statistics on a pixel-by- pixel basis after global normalization by ANCOVA. Activation areas were topographically analyzed (p>0.001) and activated pixels in each hemisphere were compared quantitatively by lateralization index. Results : Significant activation signals were demonstrated in 9 of 10 volunteers. Activation signals were found in the premotor and motor cortices, the inferior frontal, inferior parietal, and mid-temporal lobes during stimulation tasks. In the right handed seven volunteers, activation of language areas was lateralized to the left side. Verb generation task produced stronger activation in the language areas and higher value of lateralization index than noun generation task or decision task. Conclusion : fMRI could be a useful non-invasive method for language mapping and determination of language dominance.

• Investigative Magnetic Resonance Imaging
• /
• v.3 no.1
• /
• pp.78-83
• /
• 1999

Purpose : To evaluate the language areas activated by fMRI during different reading tasks and to assess the difference of activated areas according to each reading task. Materials and Methods : Functional maps of the language area were obtained during three reading tasks(Korean consonant letter, pseudoword, and word) in nine right-handed volunteers(7 males 2 females). MR examinations were performed at 1.5T scanner with EPI BOLD technique(gradient echo shot EPI, TR/TE 3000/60, flip angle $90^{\circ}$, matrix $64{\times}64$, 5mm thickness, no slice gap). Each task consisted of three resting periods and two activation periods and each period lasted 30 seconds. We used SPM program for the postprocessing of images and signification level was set at p<0.01. Activated areas were topographically analyzed in each stimulus. Results : Significant activated signals were demonstrated in all volunteers. Activated signals were seen in the frontal, temporal, parietal and occipital lobes during reading tasks and they were lateralized to the left hemisphere except occipital lobe. Letter and pseudoword produced stronger activated signals than word, and the activated signals were more lateralized to the left hemisphere in pseudoword reading than in letter reading. Conclusion : Activated signals were induced in the language areas by reading task of letter or wordform. Greater activation of language areas was induced when letter or pseudowords were presented than familiar words.

• Journal of KIISE
• /
• v.42 no.2
• /
• pp.242-254
• /
• 2015

In this paper, we address the problem of improving the performance of multi-step k-NN search using multi-dimensional indexes. Due to information loss by lower-dimensional transformations, existing multi-step k-NN search solutions produce a large tolerance (i.e., a large search range), and thus, incur a large number of candidates, which are retrieved by a range query. Those many candidates lead to overwhelming I/O and CPU overheads in the postprocessing step. To overcome this problem, we propose two efficient solutions that improve the search performance by reducing the tolerance of a range query, and accordingly, reducing the number of candidates. First, we propose a tolerance reduction-based (approximate) solution that forcibly decreases the tolerance, which is determined by a k-NN query on the index, by the average ratio of high- and low-dimensional distances. Second, we propose a coefficient control-based (exact) solution that uses c k instead of k in a k-NN query to obtain a tigher tolerance and performs a range query using this tigher tolerance. Experimental results show that the proposed solutions significantly reduce the number of candidates, and accordingly, improve the search performance in comparison with the existing multi-step k-NN solution.