• Journal of Sensor Science and Technology
• /
• v.13 no.1
• /
• pp.56-65
• /
• 2004

In this paper, we proposed the detection algorithm of microcalcification based on region growing method with contrast and edge sharpness in digital X-ray mammographic images. We extracted the local maximum pixel and watershed regions by using watershed algorithm. Then, we used the mean slope between local maximum and neighborhood pixels to extract microcalcification candidate pixels among local maximum pixels. During increasing threshold value to grow microcalcification region, at the maximum threshold value of the contrast and edge sharpness, the microcalcification area is decided. The regions of which area of grown candidate microcalfication region is larger than that of watershed region are excluded from microcalcifications. We showed the diagnosis algorithm can be used to aid diagnostic-radiologist in the early detection breast cancer.

• The Journal of the Korea Contents Association
• /
• v.17 no.6
• /
• pp.81-85
• /
• 2017

The aim of this study was to evaluate the usability of applied Low dose Computed Tomography(LDCT) protocol in examining urinary calculus using computed tomography. The subjects of this study were urological patients who visited a medical institution located in Busan from June to December 2016 and the protocol used in this study was Adaptive Statistical Iterative Reconstruction: low-dose CT with 50% Adaptive Statistical Iterative Reconstruction (ASIR). As results of quantitative analysis, the mean pixel value and standard deviation within kidney region of image(ROI)of the axial image were $26.21{\pm}7.08$ in abdomen CT pre scan and $20.03{\pm}8.16$ in low-dose CT. Also the mean pixel value and standard deviation within kidney ROI of the coronal image were $22.07{\pm}7.35$ in abdomen CT pre scan and $21.67{\pm}6.11$ in low dose CT. The results of qualitative analysis showed that four raters' mean values of observed kidney artifacts were $19.14{\pm}0.36$ when using abdomen CT protocol and $19.17{\pm}0.43$ in low-dose CT, and the mean value of resolution and contrast was $19.35{\pm}0.70$ when using abdomen CT protocol and $19.29{\pm}0.58$ in low-dose CT. Also the results of a exposure dose analysis showed that the mean values of CTDIvol and DLP in abdomen CT pre scan were 18.02 mGy and $887.51mGy{\cdot}cm$ respectively and the mean values of CTDIvol and DLP when using low-dose CT protocol were 7.412 mGy and $361.22mGy{\cdot}cm$ respectively. The resulting dose reduction rate was 58.82% and 59.29%, respectively.

• Journal of the Korea Society of Computer and Information
• /
• v.4 no.2
• /
• pp.120-134
• /
• 1999

Image processing steps consist of image acquisition, pre-processing, region segmentation and recognition, and the images are easily corrupted by noise during the data transmission, data capture, and data processing. Impulse noise and gaussian noise are major noises, which can occur during the process. Many filters such as mean filter, median filter, weighted median filter, Cheikh filter, and Kyu-cheol Lee filter were proposed as spatial noise reduction filters so far. Many researches have been focused on the reduction of impulse noise, but comparatively the research in the reduction of gaussian noise has been neglected. For the reduction of gaussian noise, subset averaged median filter, using median information and subset average information of pixels in a window. was proposed. At this time, consider of the window size as 3$^{*}$3 pixel. The window is divided to 4 subsets consisted of 4 pixels. First of all, we calculate the average value of each subset, and then find the median value by sorting the average values and center pixel's value. In this paper, a better reduction of gaussian noise was proved. The proposed algorithms were implemented by ANSI C language on a Sun Ultra 2 for testing purposes and the effects and results of the filter in the various levels of noise and images were proposed by comparing the values of PSNR, MSE, and RMSE with the value of the other existing filtering methods.thods.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.5
• /
• pp.617-622
• /
• 2004

In this work, we suggested the method which improves the efficiency of the face detection algorithm using Pulse-Coupled Neural Network. Face detection algorithm which uses the color information is independent on pose, size and obstruction of a face. But the use of color information encounters some problems arising from skin-tone color in the background, intensity variation within faces, and presence of random noise, and so on. Depending on these conditions, we obtained the mean and variance of the skin-tone colors by experiments. Then we introduce a preprocess that the pixel with a mean value of skin-tone colors has highest level value (255) and the other pixels in the skin-tone region have values between 0 and 255 according to a normal distribution with a variance. This preprocess leads to an easy decision of the linking coefficient of Pulse-Coupled Neural Network.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.12 no.3
• /
• pp.171-175
• /
• 2008

Purpose: There have been something difficulties in locating focuses and quantitative analysis in case of pediatric patients because of the relatively small body compared to adults. This author of this study, therefore, evaluated the usefulness of DFOV (Display Field Of View) according to its changes in PET/CT image reconstruction by means of the phantom experiment and pediatric patients examination. Materials & Methods: 0.023 MBq/cc of $^{18}F$-FDG was put into the uniform NU2-94 phantom, and then emission scan was acquired for 10 minutes. For reconstruction, DFOV values were changed to 50, 45, 40, 35, 30, and 25 cm respectively. As for patient images, 20 patients who were diagnosed as the one or suspicion of the children tumor are targeted from Oct 2007 to Jan 2008. For image reconstruction, 50 cm was the basis of DFOV, and the value was adjusted to DFOV 45 cm to 25 cm respectively. In the phantom and the reconstruction image of pediatric patients, the changes in pixel size and $SUV_{max}$ according to DFOV changes were analyzed. Results: As DFOV decreased to 50, 45, 40, 35, 30, and 25 cm by means of the phantom, the pixel size was changed to 3.906, 3.515, 3.125, 2.734, 2.343, and 1.953 mm respectively. Besides, as a result of reconstruction DFOV in images of pediatric patients to 50, to 25 cm, the different values of $SUV_{max}$ are shown as 3.3, 7.3, 12, 14, 18% and 2.6, 4.3, 5.0, 7.0, 10.0% on respectively when 50 cm was the standard. Conclusion: In $SUV_{max}$ using the phantom, as DFOV decreased every 5 cm, the mean value gradually increased. With 50 cm as the standard, the increase rates were 3.7, 6.5, 11.2, 19.5, and 32.1% respectively. As for pediatric patients image too, as DFOV decreased, the rates increased as in the phantom experiment. In image reconstruction, since DFOV decrease regardless of matrix size change reduced the pixel size, the image quality can be improved. This would be more useful than reconstruction and enlarge images of pediatric patients in the same way of examining adults. However, when the value of 35 cm DFOV was applied, this may result in truncated artifact, and thus the application should be properly controlled. Change of DFOV may produce better image for pediatric patients, but changes of SUV values according to DFOV change should be considered in reading.

• Spatial Information Research
• /
• v.18 no.3
• /
• pp.73-83
• /
• 2010

This study presents a method to segment an image, employing the statistics observed at each pixel location across sequential frame images. In the acquisition and analysis of spatial information, utilization of digital image processing technique has very important implications. Various image segmentation techniques have been presented to distinguish the area of digital images. In this study, based on the analysis of the spectroscopic characteristics of sequential frame images that had been previously researched, an image segmentation method was proposed by using the randomness occurring among a sequence of frame images for a same scene. First of all, we computed the mean and standard deviation values at each pixel and found reliable pixels to determine seed points using their standard deviation value. For segmenting an image into individual regions, we conducted region growing based on a T-test between reference and candidate sample sets. A comparative analysis was conducted to assure the performance of the proposed method with reference to a previous method. From a set of experimental results, it is confirmed that the proposed method using a sequence of frame images segments a scene better than a method using a single frame image.

• The Journal of the Korea Contents Association
• /
• v.3 no.2
• /
• pp.11-18
• /
• 2003

In this paper, we propose a fast variable block matching algorithm for motion estimation based on bit-patterns. Motion estimation in the proposed algorithm is peformed after the representation of image sequence is transformed 8-bit pixel values into 1-bit ones by the mean pixel value of search block, which brings a short searching time by reducing the computational complexity. Moreover, adaptive searching methods according to the motion information of the block make the procedure of motion estimation efficient by eliminating unnecessary searching processes of low motion block and deepening a searching procedure in high motion block. Experimental results show that the proposed algorithm provides bettor performance - average 0.5dB PSNR improvement and about 99% savings in the number of operations - than full search Hock matching algorithm with a fixed block size.

• Journal of the Korean Society of Radiology
• /
• v.14 no.5
• /
• pp.603-612
• /
• 2020

This study has a purpose to look into the effect of the relationship between the Tube current (mA) and SOD(Source to Object Distance), which is a parameter of lower limb angiography examination, and the dilution rate of the contrast medium concentration (300, 320, 350) on the image. To that end, using 3 mm vessel model water phantom, a vessel model custom made in the size of peripheral vessel diameter, this study measured relationships between change of parameters, such as tube current (mA), SOD and varying concentrations (300, 320, 350) of contrast medium dilution into SNR and CNR values while analyzing the coefficients of variance(cv<10). The software used to measure SNR and CNR values was Image J 1.50i from NIH (National Institutes of Health, USA). MPV (mean pixel value) and SD (standard deviation) were used after verifying numerically the image signal for region of interest (ROI) and background on phantom from the DICOM (digital imaging and communications in medicine) 3.0 file transmitted to PACS. As to contrast medium dilution by the change of tube current, when 146 mA and 102 mA were compared, For both SNR and CNR, the coefficient of variation value was less than 10 until the section of CM: N/S dilution (100% ~ 30% : 70%) but CM: N/S dilution rate (20%: 80% ~ 10% : 90%) the coefficient of variation was 10 or more. As to contrast medium dilution by concentration for SOD change, when SOD's (32.5 cm and 22.5 cm) were compared,For both SNR and CNR, the coefficient of variation value was less than 10 until the section of CM: N/S dilution (100% ~ 30% : 70%) but CM: N/S dilution rate (20%: 80% ~ 10% : 90%) the coefficient of variation was 10 or more. As to contrast medium dilution by concentration for SOD change, when SOD's (32.5 cm and 12.5 cm) were compared,For both SNR and CNR, the coefficient of variation value was less than 10 until the section of CM: N/S dilution (100% ~ 30% : 70%) but CM: N/S dilution rate (20%: 80% ~ 10% : 90%) the coefficient of variation was 10 or more. As a result, set a low tube current value in other tests or procedures including peripheral angiography of the lower extremities in the intervention, and make the table as close as possible to the image receiver, and adjust the contrast agent concentration (300) to CM: N/S dilution (30%: 70%). ) Is suggested as the most efficient way to obtain images with an appropriate concentration while simultaneously reducing the burden on the kidney and the burden on exposure.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.2
• /
• pp.907-923
• /
• 2020

For the severe haze situation in the Beijing-Tianjin-Hebei region, conventional fine particulate matter (PM2.5) concentration prediction methods based on pollutant data face problems such as incomplete data, which may lead to poor prediction performance. Therefore, this paper proposes a method of predicting the PM2.5 concentration based on image analysis technology that combines image data, which can reflect the original weather conditions, with currently popular machine learning methods. First, based on local parameter estimation, autoregressive (AR) model analysis and local estimation of the increase in image blur, we extract features from the weather images using an approach inspired by free energy and a no-reference robust metric model. Next, we compare the coefficient energy and contrast difference of each pixel in the AR model and then use the percentages to calculate the image sharpness to derive the overall mass fraction. Furthermore, the results are compared. The relationship between residual value and PM2.5 concentration is fitted by generalized Gauss distribution (GGD) model. Finally, nonlinear mapping is performed via the wavelet neural network (WNN) method to obtain the PM2.5 concentration. Experimental results obtained on real data show that the proposed method offers an improved prediction accuracy and lower root mean square error (RMSE).

• Journal of the Institute of Electronics Engineers of Korea TE
• /
• v.37 no.3
• /
• pp.45-52
• /
• 2000

In this study, we propose the image segmentation algorithm for facial region segmentation. The proposed algorithm separates the mean image of low frequency band from the differential image of high frequency band in order to make a boundary using HWT, and then we reduce the isolation pixels, projection pixels, and overlapped boundary pixels from the low frequency band. Also the boundaries are detected and simplified by the proposed boundary detection algorithm, which are cleared on the thinning process of 1 pixel unit. After extracting facial image boundary by using the proposed algorithm, we make the mask and segment facial image through matching original image. In the result of facial region segmentation experiment by using the proposed algorithm, the successive facial segmentation have 95.88% segmentation value.