• The Journal of the Convergence on Culture Technology
• /
• v.7 no.2
• /
• pp.411-417
• /
• 2021

The field that studies human-computer interaction is called HCI (Human-computer interaction). This field is an academic field that studies how humans and computers communicate with each other and recognize information. This study is a study on hand gesture recognition for human interaction. This study examines the problems of existing recognition methods and proposes an algorithm to improve the recognition rate. The hand region is extracted based on skin color information for the image containing the shape of the human hand, and the center of gravity profile is calculated using principal component analysis. I proposed a method to increase the recognition rate of hand gestures by comparing the obtained information with predefined shapes. We proposed a method to increase the recognition rate of hand gestures by comparing the obtained information with predefined shapes. The existing center of gravity profile has shown the result of incorrect hand gesture recognition for the deformation of the hand due to rotation, but in this study, the center of gravity profile is used and the point where the distance between the points of all contours and the center of gravity is the longest is the starting point. Thus, a robust algorithm was proposed by re-improving the center of gravity profile. No gloves or special markers attached to the sensor are used for hand gesture recognition, and a separate blue screen is not installed. For this result, find the feature vector at the nearest distance to solve the misrecognition, and obtain an appropriate threshold to distinguish between success and failure.

• Conservation Science in Museum
• /
• v.24
• /
• pp.55-74
• /
• 2020

The Sacred Bell of Great King Seongdeok is required digital precision recording of conservation conditions because of corrosion and partial abrasion of its patterns and inscriptions. Therefore, this study performed digital documentation of the bell using four types of scanning and unmanned aerial vehicle (UAV) photogrammetry technologies, and performed the various shape analyses through image processing. The modeling results of terrestrial laser scanning and UAV photogrammetry were merged and utilized as basic material for monitoring earthquake-induced structural deformation because these techniques can construct mutual spatial relationships between the bell and its tower. Additionally, precision scanning at a resolution four to nine times higher than that of the previous study provided highly valuable information, making it possible to visualize the patterns and inscriptions of the bell. Moreover, they are well-suited as basic data for identifying surface conservation conditions. To actively apply three-dimensional scanning results to the conservation of the original bell, the time and position of any changes in shape need to be established by further scans in the short-term. If no change in shape is detected by short-term monitoring, the monitoring should continue in medium- and long-term intervals.

• Journal of the Korea Computer Graphics Society
• /
• v.28 no.2
• /
• pp.1-9
• /
• 2022

In this paper, we propose a vector-based augmentation technique that can generate data required for crack detection and a ConvNet(Convolutional Neural Network) technique that can learn it. Detecting cracks quickly and accurately is an important technology to prevent building collapse and fall accidents in advance. In order to solve this problem with artificial intelligence, it is essential to obtain a large amount of data, but it is difficult to obtain a large amount of crack data because the situation for obtaining an actual crack image is mostly dangerous. This problem of database construction can be alleviated with elastic distortion, which increases the amount of data by applying deformation to a specific artificial part. In this paper, the improved crack pattern results are modeled using ConvNet. Rather than elastic distortion, our method can obtain results similar to the actual crack pattern. By designing the crack data augmentation based on a vector, rather than the pixel unit used in general data augmentation, excellent results can be obtained in terms of the amount of crack change. As a result, in this paper, even though a small number of crack data were used as input, a crack database can be efficiently constructed by generating various crack directions and patterns.

• The Journal of Korean Society for Radiation Therapy
• /
• v.28 no.2
• /
• pp.161-169
• /
• 2016

Purpose : To quantify the inter-fractional variation in prostate displacement and their dosimetric effects for prostate cancer treatment. Materials and Methods : A total of 176 daily cone-beam CT (CBCT) sets acquired for 6 prostate cancer patients treated with volumetric-modulated arc therapy (VMAT) were retrospectively reviewed. For each patient, the planning CT (pCT) was registered to each daily CBCT by aligning the bony anatomy. The prostate, rectum, and bladder were delineated on daily CBCT, and the contours of these organs in the pCT were copied to the daily CBCT. The concordance of prostate displacement, deformation, and size variation between pCT and daily CBCT was evaluated using the Dice similarity coefficient (DSC). Results : The mean volume of prostate was 37.2 cm3 in the initial pCT, and the variation was around ${\pm}5%$ during the entire course of treatment for all patients. The mean DSC was 89.9%, ranging from 70% to 100% for prostate displacement. Although the volume change of bladder and rectum per treatment fraction did not show any correlation with the value of DSC (r=-0.084, p=0.268 and r=-0.162, p=0.032, respectively), a decrease in the DSC value was observed with increasing volume change of the bladder and rectum (r=-0.230,p=0.049 and r=-0.240,p=0.020, respectively). Conclusion : Consistency of the volume of the bladder and rectum cannot guarantee the accuracy of the treatment. Our results suggest that patient setup with the registration between the pCT and daily CBCT should be considered aligning soft tissue.

• Radiation Oncology Journal
• /
• v.26 no.4
• /
• pp.280-288
• /
• 2008

Purpose: To identify the inter-fractional shift pattern and to assess an adequate treatment margin in the radiotherapy of a liver tumor using mega-voltage computed tomography (MVCT) of a tomotherapy unit. Materials and Methods: Twenty-six patients were treated for liver tumors by tomotherapy from April 2006 to August 2007. The MVCT images of each patient were analyzed from the $1^{st}$ to the $10^{th}$ fraction for the assessment of the daily liver shift by four groups based on Couinard's proposal. Daily setup errors were corrected by bony landmarks as a prerequisite. Subsequently, the anterior-, posterior-, right-, and left shifts of the liver edges were measured by maximum linear discrepancies between the kilo-voltage computed tomography (KVCT) image and MVCT image. All data were set in the 2-dimensional right angle coordinate system of the transverse section of each patient's body. Results: The liver boundary shift had different patterns for each group. In group II (segment 2, 3, and 4), the anterior mean shift was $2.80{\pm}1.73\;mm$ outwards, while the left mean shift was $2.23{\pm}1.37\;mm$ inwards. In group IV (segment 7 and 8), the anterior-, posterior-, right-, and left mean shifts were $0.15{\pm}3.93\;mm$ inwards, $3.15{\pm}6.58\;mm$ inwards, $0.60{\pm}3.58\;mm$ inwards, and $4.50{\pm}5.35\;mm$ inwards, respectively. The reduced volume in group II after MVCT reassessment might be a consequence of stomach toxicity. Conclusion: Inter-fractional liver shifts of each group based on Couinard's proposal were somewhat systematic despite certain variations observed in each patient. The geometrical deformation of the liver by respiratory movement can cause shrinkage in the left margins of liver. We recommend a more sophisticated approach in free-breathing mode when irradiating the left lobe of liver in order to avoid stomach toxicity.

• Journal of the Korean Wood Science and Technology
• /
• v.39 no.4
• /
• pp.318-325
• /
• 2011

Pitch pine (Pinus rigida Miller) retaining walls using Steel bar, of which the constructability and strength performance are good at the construction site, were manufactured and their strength properties were evaluated. The wooden retaining wall using Steel bar was piled into four stories stretcher and three stories header, which is 770 mm high, 2,890 mm length and 782 mm width. Retaining wall was made by inserting stretchers into Steel bar after making 18 mm diameter of holes at top and bottom stretcher, and then stacking other stretchers and headers which have a slit of 66 mm depth and 18 mm width. The strength properties of retaining walls were investigated by horizontal loading test, and the deformation of structure by image processing (AlCON 3D OPA-PRO system). Joint (Type-A) made with a single long stretcher and two headers, and joint (Type-B) made with two short stretchers connected with half lap joint and two headers were in the retaining wall using Steel bar. The compressive shear strength of joint was tested. Three replicates were used in each test. In horizontal loading test the strength was 1.6 times stronger in wooden retaining wall using Steel bar than in wooden retaining wall using square timber. The timber and joints were not fractured in the test. When testing compressive shear strength, the maximum load of type-A and Type-B was 130.13 kN and 130.6 kN, respectively. Constructability and strength were better in the wooden retaining wall using Steel bar than in wooden retaining wall using square timber.

• Korean Journal of Cognitive Science
• /
• v.3 no.1
• /
• pp.61-78
• /
• 1991

This paper descibes the study of application of a modified Neocognitron model with backward path for the recognition of Hangul(Korean) syllabic characters. In this original report, Fukushima demonstrated that Neocognitron can recognize hand written numerical characters of $19{\times}19$ size. This version accepts $61{\times}61$ images of handwritten Hangul syllabic characters or a part thereof with a mouse or with a scanner. It consists of an input layer and 3 pairs of Uc layers. The last Uc layer of this version, recognition layer, consists of 24 planes of $5{\times}5$ cells which tell us the identity of a grapheme receiving attention at one time and its relative position in the input layer respectively. It has been trained 10 simple vowel graphemes and 14 simple consonant graphemes and their spatial features. Some patterns which are not easily trained have been trained more extrensively. The trained nerwork which can classify indivisual graphemes with possible deformation, noise, size variance, transformation or retation wre then used to recongnize Korean syllabic characters using its selective attention mechanism for image segmentation task within a syllabic characters. On initial sample tests on input characters our model could recognize correctly up to 79%of the various test patterns of handwritten Korean syllabic charactes. The results of this study indeed show Neocognitron as a powerful model to reconginze deformed handwritten charavters with big size characters set via segmenting its input images as recognizable parts. The same approach may be applied to the recogition of chinese characters, which are much complex both in its structures and its graphemes. But processing time appears to be the bottleneck before it can be implemented. Special hardware such as neural chip appear to be an essestial prerquisite for the practical use of the model. Further work is required before enabling the model to recognize Korean syllabic characters consisting of complex vowels and complex consonants. Correct recognition of the neighboring area between two simple graphemes would become more critical for this task.

• Radiation Oncology Journal
• /
• v.28 no.3
• /
• pp.166-176
• /
• 2010

Purpose: To determine the appropriate prostate planning target volume (PTV) margins for 3-dimensitional (3D) conformal radiotherapy (CRT) and intensity-modulated radiation therapy (IMRT) patients treated with an endorectal balloon (ERB) under our institutional treatment condition. Materials and Methods: Patients were treated in the supine position. An ERB was inserted into the rectum with 70 cc air prior to planning a CT scan and then each treatment fraction. Electronic portal images (EPIs) and digital reconstructed radiographs (DRR) of planning CT images were used to evaluate inter-fractional patient's setup and ERB errors. To register both image sets, we developed an in-house program written in visual $C^{++}$. A new method to determine prostate PTV margins with an ERB was developed by using the common method. Results: The mean value of patient setup errors was within 1 mm in all directions. The ERB inter-fractional errors in the superior-inferior (SI) and anterior-posterior (AP) directions were larger than in the left-right (LR) direction. The calculated 1D symmetric PTV margins were 3.0 mm, 8.2 mm, and 8.5 mm for 3D CRT and 4.1 mm, 7.9 mm, and 10.3 mm for IMRT in LR, SI, and AP, respectively according to the new method including ERB random errors. Conclusion: The ERB random error contributes to the deformation of the prostate, which affects the original treatment planning. Thus, a new PTV margin method includes dose blurring effects of ERB. The correction of ERB systematic error is a prerequisite since the new method only accounts for ERB random error.

• Restorative Dentistry and Endodontics
• /
• v.35 no.2
• /
• pp.88-95
• /
• 2010

The purpose of this study was to compare the root canal shaping ability of 4 rotary NiTi instruments in simulated root canals. For the preparation of thirty two curved root canals, Mtwo instruments using "single length"technique, and Profile, ProTaper Universal, and K3 using crown-down technique (N = 8) were used. All canal samples were prepared by reaching an apical canal size of #30. Pre- and post-instrumentation digital images were recorded and an assessment of canal shape was determined using a computer image analysis program SigmaScan Pro (Systat Software Inc., San Jose, CA, USA). The changes of the dimension of inner walls of canals, (2) the changes of the dimension of outer walls of canals, and (3) the centering ratio were measured at 7 measuring points, and then data were statistically analyzed using one-way ANOVA and Duncan's test. The results were as below; 1. The root canal shaping ability of Profile was significantly faster than that of other rotary NiTi instruments (p < 0.05). 2. The deformation and fracture of all instruments used for this study were not experienced. 3. In the degree of changes of the dimension of inner walls of canals, Profile demonstrated the lowest changes of the dimension of inner walls of canals except at the measuring points of the 1 and 2 mm (p < 0.05). However, the ProTaper Universal showed the highest changes of the dimension of inner walls of canals at all measuring points (p < 0.05). 4. In the degree of changes of the dimension of outer walls of canals, Mtwo demonstrated the lowest changse of the dimension of outer walls of canals except at the measuring point of the 1 mm (p < 0.05). However, Profile exhibited the highest changes of the dimension of outer walls of canals at the measuring points of 3 and 4 mm and ProTaper Universal and K3 showed the largest changes of the dimension of outer walls of canals at the measuring points of 1, 2, 6, and 7 mm (p < 0.05). 5. In degree of centering ratio, Profile demonstrated the least centering ratio comparing with the centering ratio shown by other NiTi instruments at the measuring points of 1, 4, 5, and 6 mm. Results suggest that in the coronal part of canal preparation, active cutting files such as ProTaper Universal may efficiently flare the canal orifice and form a better taper, and in the apical part of the canal, files which have a better centering ability such as Profile may maintain the original canal curvature and reduce the shaping time.

• Restorative Dentistry and Endodontics
• /
• v.31 no.3
• /
• pp.179-185
• /
• 2006

This study was conducted to evaluate canal configuration after shaping by $ProTaper^{TM}$ with various rotational speed in J-shaped simulated resin canals. Forty simulated root canals were divided into 4 groups, and instrumented using by $ProTaper^{TM}$ at the rotational speed of 250, 300, 350 and 400 rpm. Pre-instrumented and post-instrumented images were taken by a scanner and those were superimposed. Outer canal width, inner canal width, total canal width, and amount of transportation from original axis were measured at 1, 2, 3, 4, 5, 6, 7 and 8 mm from apex. Instrumentation time, instrument deformation and fracture were recorded. Data were analyzed by means of one-way ANOVA followed by Scheffe's test. The results were as follows 1. Regardless of rotational speed, at the $1{\sim}2mm$ from the apex, axis of canal was transported to outer side of a curvature, and at 3~6 mm from the apex, to inner side of a curvature. Amounts of transportation from original axis were not sienifcantly different among experimental groups except at 5 and 6 mm from the apex. 2. Instrumentation time of 350 and 400 rpm was significantly less than that of 250 and 300 rpm (p<0.01). In conclusion the rotational speed of $ProTaper^{TM}$ files in the range of $250{\sim}400rpm$ does not affect the change of canal configuration, and high rotational speed reduces the instrumentation time. However appearance of separation and distortion of Ni-Ti rotary files can occur in high rotational speed.