• The Korean Journal of Nuclear Medicine Technology
• /
• v.28 no.1
• /
• pp.57-69
• /
• 2024

Purpose: Quantitative analysis through count measurement in nuclear medicine planar images is limited by analysis techniques that are useful for obtaining various clinical information or by organ overlap or artifacts in actual clinical practice. On the other hand, the use of SPECT tomography images is quantitative analysis using volume rather than planar, which is not only free from problems such as projection overlap, but also has excellent quantitative accuracy. In the use of developing SPECT quantitative analysis technology, this study aims to compare the accuracy of quantitative analysis between ROI of the conventional planar images and VOI of the SPECT tomographic images in evaluating the count change happened by the volume change of the source. Materials and Methods: A ^99mTcO₄^- source(200.17 MBq) was filled with sterilized water in the syringe to create a phantom with an inner diameter volume of 60 cc, and a planar image and a SPECT image were obtained by reducing the volume by 15 cc (25%) respectively. ROI and VOI(threshold: 1~45%, 5% interval) were set for each image obtained to estimate true count and measure the total count, and compared with the preseted volumetric change rate(%). Results: When volume changes of 25%, 50%, and 75% occurred in the initial volume of 60 cc(100%) of the phantom, the average count changes of the measured planar image were 26.8%, 53.2%, 77.5%, and the average count changes of the SPECT image were 24.4%, 50.9%, and 76.8%. In this case, the VOI size(cm³) set showed an average change rate of 25.4%, 51.1%, and 76.6%. The highest threshold value for the accuracy of radioactive concentration by VOI size (average error -1.03%) was 35%, and the VOI size of the same threshold had an error of -17.1% on average compared to the actual volume. Conclusion: On average, the count-based volumetric change rate in nuclear medicine images was able to track changes more accurately using VOI than ROI, but there was no significant difference with relatively similar value. However, the accuracy of radioactive concentration according to individual VOI sizes did not match, but it is considered that a relatively accurate quantitative analysis can be expected when the size of VOI is set smaller than the actual volume.

• The KIPS Transactions:PartB
• /
• v.15B no.4
• /
• pp.259-268
• /
• 2008

In an effort to increase the data capacity of one-dimensional symbology, 2D barcodes have been proposed a decade ago. In this paper, we present an effective 2D barcode detection algorithm from gray-level images, especially for the handheld 2D barcode recognition system. To locate the symbol inside the image, a criteria based on the block contrast is adopted, and a gray-scale projection with sub-pixel operation is utilized to segment the symbol precisely from the region of interest(ROI). Finally, the segmented ROI is normalized using the inverse perspective transformation for the following decoding processes. We also introduce the post-processing steps for decoding the QR-code. The proposed method ensures high performances under various lighting/printing conditions and strong perspective deformations. Experiments shows that our method is very robust and efficient in detecting the code area for the various types of 2D barcodes in real time.

• Journal of Internet Computing and Services
• /
• v.19 no.1
• /
• pp.105-111
• /
• 2018

This paper presents the analysis of image quality with FHD(Full HD) resolution camera equipped DIC(Digital Image Correlation) platform for the measurement of the architectural structure's relative displacement. DIC platform was designed based on i.MX6 of Freescale. Displacement measurement based on DIC method, the error is affected by image quality factors as pixel number, brightness, contrast, and SNR[dB](Signal to Noise Ratio). The effect were analyzed. The displacement of ROI(Region Of Interest) area within the image was measured by sub-pixel units based on DIC method. The non-contact telemetry property of DIC method, it can be used to long distance non-contact measurement. The various displacement results was measured and analyzed with the image quality factor adjustment according to the distance(25m, 35m, 50m).

• International Journal of Control, Automation, and Systems
• /
• v.6 no.6
• /
• pp.800-808
• /
• 2008

A fingerprint verification system based on a set of invariant moment features and a nonlinear Back Propagation Neural Network(BPNN) verifier is proposed. An image-based method with invariant moment features for fingerprint verification is used to overcome the demerits of traditional minutiae-based methods and other image-based methods. The proposed system contains two stages: an off-line stage for template processing and an on-line stage for testing with input fingerprints. The system preprocesses fingerprints and reliably detects a unique reference point to determine a Region-of-Interest(ROI). A total of four sets of seven invariant moment features are extracted from four partitioned sub-images of an ROI. Matching between the feature vectors of a test fingerprint and those of a template fingerprint in the database is evaluated by a nonlinear BPNN and its performance is compared with other methods in terms of absolute distance as a similarity measure. The experimental results show that the proposed method with BPNN matching has a higher matching accuracy, while the method with absolute distance has a faster matching speed. Comparison results with other famous methods also show that the proposed method outperforms them in verification accuracy.

• Korean Journal of Remote Sensing
• /
• v.27 no.2
• /
• pp.151-161
• /
• 2011

PGA(Phase Gradient Autofocus) is a representative autofocus technique to improve the SAR(Synthetic Aperture Radar) image quality. PGA can estimate high order phase errors and have good robustness in noisy environments. However, PGA is not suitable to apply to the stripmap mode data directly because it is based on the spotlight mode operation. In this paper, the PGA implementation technique for stripmap mode data and the method of ROI(Region of Interest) selection that affects severely on PGA performance have been proposed. The proposed technique was verified by the point target simulation first, and was applied to the real SAR signal data acquired by the flight test. Finally, the significant improvements in focusing quality were shown in the processed SAR images using the proposed method.

• Korean Journal of Remote Sensing
• /
• v.40 no.2
• /
• pp.167-177
• /
• 2024

With advancements in satellite technology, interest in target detection and identification is increasing quantitatively and qualitatively. Synthetic Aperture Radar(SAR) images, which can be acquired regardless of weather conditions, have been applied to various areas combined with machine learning based detection algorithms. However, conventional studies primarily focused on the detection of stationary targets. In this study, we proposed a method to identify moving targets using an algorithm that integrates sub-aperture SAR images and cosine similarity calculations. Utilizing a transformer-based deep learning target detection model, we extracted the bounding box of each target, designated the area as a region of interest (ROI), estimated the similarity between sub-aperture SAR images, and determined movement based on a predefined similarity threshold. Through the proposed algorithm, the quantitative evaluation of target identification capability enhanced its accuracy compared to when training with the targets with two different classes. It signified the effectiveness of our approach in maintaining accuracy while reliably discerning whether a target is in motion.

• Progress in Medical Physics
• /
• v.30 no.4
• /
• pp.139-149
• /
• 2019

Purpose: With neurodegeneration, the signal intensity of the cerebrospinal fluid (CSF) in the brain increases. The objective of this study was to evaluate chemical exchange saturation transfer (CEST) signals with and without the contribution of CSF signals in elderly human brains using two different 3T magnetic resonance imaging (MRI) sequences Methods: Full CEST signals were acquired in ten subjects (Group I) with a three-dimensional (3D)-segmented gradient-echo echo-planar imaging (EPI) sequence and in ten other subjects (Group II) with a 3D gradient and spin-echo (GRASE) sequence using two different 3T MRI systems. The segmented tissue compartments of gray and white matter were used to mask the CSF signals in the full CEST images. Two sets of magnetization transfer ratio asymmetry (MTR_asym) maps were obtained for each offset frequency in each subject with and without masking the CSF signals (masked and unmasked conditions, respectively) and later compared using paired t-tests. Results: The region-of-interest (ROI)-based analyses showed that the MTR_asym values for both the 3D-segmented gradient-echo EPI and 3D GRASE sequences were altered under the masked condition compared with the unmasked condition at several ROIs and offset frequencies. Conclusions: Depending on the imaging sequence, the MTR_asym values can be overestimated for some areas of the elderly human brain when CSF signals are unmasked. Therefore, it is necessary to develop a method to minimize this overestimation in the case of elderly patients.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.8
• /
• pp.1885-1894
• /
• 2014

There are various high-performance algorithms in the area of the existing VDSs (vehicle detection systems). However, they requires a large amount of computational time-complexity and their systems generally are very expensive and consumes high-power. This paper proposes real-time traffic information detection algorithm that can be applied to low-cost, low-power, and open development platform such as Android. This algorithm uses a vehicle's sub-shadow to set ROI(region of interest) and to count vehicles using a location of the sub-shadow and the vehicle. The proposed algorithm is able to count the vehicles per each roads and each directions separately. The experiment result show that the detection rate for going-up vehicles is 94.1% and that for going-down vehicles is 97.1%. These results are close to or surpasses 95%, the detection rate of commercial loop detectors.

• Journal of radiological science and technology
• /
• v.46 no.5
• /
• pp.401-408
• /
• 2023

Thyroid uptake measurements can be subject to measurement errors due to the scoping and positioning of the thyroid gland. To compensate for these limitations, the clinical utility of the thyroid simultaneous counting method as an alternative to thyroid uptake measurement was analyzed and evaluated experimentally through quantitative analysis of images acquired after thyroid scanning. Experimental data were obtained using a Gamma camera (GE infinia), a thyroid uptake system (KOROID 1), and a thyroid neck phantom. Based on the thyroid uptake rate of 1-5% according to the protocol of thyroid scan test (^99mTcO₄ - , 370 MBq) in normal results, ^99mTcO₄ - was set in the range of 3.7-18.5 MBq (Matrix: 256×256, Scan time: 1 min, collimator: pin hole, phantom-collimator distances: 7 cm). The acquired images were corrected for the attenuation of isotopes due to the set-up time and half-life by applying the Auto Region of interest (ROI) drawing system, and the significance of the experimental results was evaluated by Multiple linear regression analysis (SPSS, ver. 22, IBM). The thyroid uptake rate showed a significant correlation between the dose and the measured counts when using the thyroid uptake system equipment. Meanwhile, the quantitative analysis counts of phantom images using Gamma camera also showed a significant correlation. Thus confirmed that the correlation between these two experiments was statistically significant (P<0.05). The simultaneous counting protocol, which indirectly measures thyroid uptake from thyroid scans, is likely to be clinically relevant if complemented by additional studies with different variables in patients with thyroid disease.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2010.07a
• /
• pp.260-263
• /
• 2010

본 논문에서는 주행 중인 차량에서 전방을 향해 장착된 카메라를 통해 입력된 영상에서 측면에 부분적으로 나타나는 차량을 검출하는 방법을 제안한다. 기존 연구에서는 모션 벡터를 이용하여 주변 배경과 관측되는 차량 사이의 모션 벡터 차이를 이용하여 측면 차량을 검출하고 있다. 그러나 모션 벡터를 이용할 경우 정지된 차량이나 전방에서 다가오는 차량의 경우 검출하기 어려운 점이 있다. 이러한 문제를 해결하기 위해 본 논문에서는 모션 벡터를 사용하지 않고 차량 측면 모습에서 특징 정보를 추출하여 SVM 분류기를 통해 측면 차량을 검출하는 방법을 제안한다. 차량 측면 모습의 특징을 뽑기 위해 영상의 밝기 변화에 강인한 국부 이진 패턴을 사용하였고 ROI영역 내에서 차량이 나타나는 위치에 상관없이 차량의 측면 모습을 찾아내기 위해 국부 이진 패턴의 히스토그램을 이용하였다. 실험결과에서는 제안하는 방법이 정지된 차량을 포함하여 88.5%의 정확도로 측면 차량을 검출하는 것을 확인하였다.