• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.38 no.1
• /
• pp.19-25
• /
• 2001

A novel approach for the feature extraction of handwritten characters is proposed by using spatial filtering neural networks with 4 layers. The proposed system first removes rough pixels which are easy to occur in handwritten characters. The system then extracts and removes the boundary information which have no influence on characters recognition. Finally, The system extracts feature information and removes the noises from feature information. The spatial filters adapted in the system correspond to the receptive fields of ganglion cells in retina and simple cells in visual cortex. With PE2 Hangul database, we perform experiments extracting features of handwritten characters recognition. It will be shown that the network can extract feature informations from handwritten characters successfully.

• The KIPS Transactions:PartB
• /
• v.10B no.5
• /
• pp.521-526
• /
• 2003

In this paper, we propose a geometric active contour model based on intensity information and curve evolution for detecting region boundaries. We put boundary extraction problem as the minimization of the difference between the average intensity of the region and the intensity of the expanding closed curves. We used level set theory to implement the curve evolution for optimal solution. It offered much more freedom in the initial curve position than a general active contour model. Our methods could detect regions whose boundaries are not necessarily defiened by gradient compared to general edge based methods and detect multiple boundaries at the same time. We could improve the result by using anisotropic diffusion filter in image preprocessing. The performance of our model was demonstrated on several data sets like CT and MRI medical images.

• Proceedings of the KOSOMBE Conference
• /
• v.1996 no.05
• /
• pp.175-178
• /
• 1996

The purpose of this thesis is the propose of various signal processing algorithm for the ECG(electrocardiogram) and the design of realtime automated ECG analyzer feasible with these algorithms. The algorithms are composed of (1)filtering procedure fer the estimation and removal of baseline drift, 60Hz power line interference, and muscle artifacts (2)detection procedure of QRS complex and P wave (3)typification procedure for the pattern classification according to the morphologies (4) selection of representative beat, significant point and wave boundary decision procedure and (5) parameter extraction and diagnosis procedure. All verifications are carried out between the algorithms proposed in this paper and other algorithms already proposed by many researchers, for the objective comparison in each procedure. The efficiency of proposed algorithms are demonstrated with the aid of internationally validated CSE database and the performances of filtering procedure are compared on artificial noise signal as well as actual ECG signals with appropriate noise components. for the comparison on the performance of designed automated ECG analyzer, the diagnosis results were compared with ECG analyzer manufactered by Fukuda denshi in Japan.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.3
• /
• pp.1418-1433
• /
• 2019

In computer graphics, 3D mesh segmentation is a challenging research field. This paper presents a 3D mesh model segmentation algorithm that focuses on removing exterior salient parts from the original 3D mesh model based on prominent feature points and marching plane. To begin with, the proposed approach uses multi-dimensional scaling to extract prominent feature points that reside on the tips of each exterior salient part of a given mesh. Subsequently, a set of planes intersect the 3D mesh; one is the marching plane, which start marching from prominent feature points. Through the marching process, local cross sections between marching plane and 3D mesh are extracted, subsequently, its corresponding area are calculated to represent local volumes of the 3D mesh model. As the boundary region of an exterior salient part generally lies on the location at which the local volume suddenly changes greatly, we can simply cut this location with the marching plane to separate this part from the mesh. We evaluated our algorithm on the Princeton Segmentation Benchmark, and the evaluation results show that our algorithm works well for some categories.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.5
• /
• pp.45-52
• /
• 2004

According tothe development of digital media technologies various algorithms for video sequence matching have been proposed to match the video sequences efficiently. A large number of video sequence matching methods have focused on frame-wise query, whereas a relatively few algorithms have been presented for video sequence matching or video shot matching. In this paper, we propose an efficientalgorithm to index the video sequences and to retrieve the sequences for video sequence query. To improve the accuracy and performance of video sequence matching, we employ the Cauchy function as a similarity measure between histograms of consecutive frames, which yields a high performance compared with conventional measures. The key frames extracted from segmented video shots can be used not only for video shot clustering but also for video sequence matching or browsing, where the key frame is defined by the frame that is significantly different from the previous fames. Several key frame extraction algorithms have been proposed, in which similar methods used for shot boundary detection were employed with proper similarity measures. In this paper, we propose the efficient algorithm to extract key frames using the cumulative Cauchy function measure and. compare its performance with that of conventional algorithms. Video sequence matching can be performed by evaluating the similarity between data sets of key frames. To improve the matching efficiency with the set of extracted key frames we employ the Cauchy function and the modified Hausdorff distance. Experimental results with several color video sequences show that the proposed method yields the high matching performance and accuracy with a low computational load compared with conventional algorithms.

• Journal of Korean Society of Forest Science
• /
• v.101 no.3
• /
• pp.365-373
• /
• 2012

This study purposed to extract and assess the accuracy of assessment for deforestation area in Wonju city using medium resolution satellite image. The total size of deforestation area during the last nine years (2000-2008) was about 467 ha, and it was occurred annually about 52 ha. The most frequent form of deforestation was settlements (72%). Ninety percent of the size of deforestation was less than 2 ha in size. In addition, 79 percent of deforestation area was found within 500 m from the road network and within 100 m of the Forest/Non-forest boundary. This study compared the deforestation based on the administrative information (GIS deforestationI) with the deforestation (RS deforestation) extracted from the satellite imagery by vegetation indices (NDVI, NBR, NDWI). Extraction accuracy, mean-standard deviation${\times}1.5$ applied 3 by 3 filtering, showed reliable accuracy 35.47% k-value 0.20. However, error could be occurred because of the difference of land-use change and land-cover change. The actual rate of land-cover change deforestation area was 32% on administrative information. The 7.52% of forest management activities area was misjudged as deforestation by RS deforestation. Finally, the comparison of land-cover change deforestation (GIS deforestationII) with the RS deforestation accuracy, as a result NDVI mean-standard deviation${\times}2$ applied 3 by 3 filtering, showed improved accuracy 61.23%, k-value 0.23.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.43 no.2 s.308
• /
• pp.21-32
• /
• 2006

The license plate recognition (LPR) unit consists of the following core components: plate region segmentation, individual character extraction, and character recognition. Out of the above three components, accuracy in the performance of plate region segmentation determines the overall recognition rate of the LPR unit. This paper proposes an algorithm for segmenting the license plate region on the front or rear of a vehicle in a fast and accurate manner. In the case of the proposed algorithm images are captured on the spot where unmanned monitoring of illegal parking and stowage is performed with a variety of roadway environments taken into account. As a means of enhancing the segmentation performance of the on-the-spot-captured images of license plate regions, the proposed algorithm uses a mathematical model for license plate colors to convert color images into digital data. In addition, this algorithm uses Gaussian smoothing and double threshold to eliminate image noises, one-pass boundary tracing to do region labeling, and MBR to determine license plate region candidates and extract individual characters from the determined license plate region candidates, thereby segmenting the license plate region on the front or rear of a vehicle through a verification process. This study contributed to addressing the inability of conventional techniques to segment the license plate region on the front or rear of a vehicle where the frame of the license plate is damaged, through processing images in a real-time manner, thereby allowing for the practical application of the proposed algorithm.

• Journal of the Korea Convergence Society
• /
• v.9 no.11
• /
• pp.61-68
• /
• 2018

In this paper, we propose a method of character detection by analyzing image enhancement and character type to detect characters in natural images that can be acquired in everyday life. The proposed method emphasizes the boundaries of the object part using the unsharp mask in order to improve the detection rate of the area to be recognized as a character in a natural image. By using the boundary of the enhanced object, the character candidate region of the image is detected using Maximal Stable Extermal Regions (MSER). In order to detect the region to be judged as a real character in the detected character candidate region, the shape of each region is analyzed and the non-character region other than the region having the character characteristic is removed to increase the detection rate of the actual character region. In order to compare the objective test of this paper, we compare the detection rate and the accuracy of the character region with the existing methods. Experimental results show that the proposed method improves the detection rate and accuracy of the character region over the existing character detection method.

• Membrane and Water Treatment
• /
• v.10 no.3
• /
• pp.207-212
• /
• 2019

Non-aqueous solvents (NASs) are generally known to be barely miscible, and reactive with polar compounds, such as water. However, water can interact with some NASs, which can be used as a new means for water recovery from saline water. This study explored the fate of water and salt in NAS, when saline water is mixed with NAS. Three amine solvents were selected as NAS. They had the same molecular formula, but were differentiated by their molecular structures, as follows: 1) NAS 'A' having the hydrophilic group ($NH_2$) at the end of the straight carbon chain, 2) NAS 'B' with symmetrical structure and having the hydrophilic group (NH) at the middle of the straight carbon chain, 3) NAS 'C' having the hydrophilic group ($NH_2$) at the end of the straight carbon chain but possessing a hydrophobic ethyl branch in the middle of the structure. In batch experiments, 0.5 M NaCl water was blended with NASs, and then water and salt content in the NAS were individually measured. Water absorption efficiencies by NAS 'B' and 'C' were 3.8 and 10.7%, respectively. However, salt rejection efficiency was 98.9% and 58.2%, respectively. NAS 'A' exhibited a higher water absorption efficiency of 35.6%, despite a worse salt rejection efficiency of 24.7%. Molecular dynamic (MD) simulation showed the different interactions of water and salts with each NAS. NAS 'A' formed lattice structured clusters, with the hydrophilic group located outside, and captured a large numbers of water molecules, together with salt ions, inside the cluster pockets. NAS 'B' formed a planar-shaped cluster, where only some water molecules, but no salt ions, migrated to the NAS cluster. NAS 'C', with an ethyl group branch, formed a cluster shaped similarly to that of 'B'; however, the boundary surface of the cluster looked higher than that of 'C', due to the branch structure in solvent. The MD simulation was helpful for understanding the experimental results for water absorption and salt rejection, by demonstrating the various interactions between water molecules and the salts, with the different NAS types.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.32 no.1D
• /
• pp.81-88
• /
• 2012

This present research was carried out by dividing two parts; field surveying and data processing, in order to analyze changed patterns of a shoreline. Firstly, the shoreline information measured by the precise GPS positioning during long duration was collected. Secondly, the algorithm for detecting an auto boundary with regards to the changed shoreline with multi-image data was developed. Then, a comparative research was conducted. Haeundae beach which is one of the most famous ones in Korea was selected as a test site. RTK-GPS surveying had been performed overall eight times from September 2005 to September 2009. The filed test by aerial Lidar was conducted twice on December 2006 and March 2009 respectively. As a result estimated from both sensors, there is a slight difference. The average length of shoreline analyzed by RTK-GPS is approximately 1,364.6 m, while one from aerial Lidar is about 1,402.5 m. In this investigation, the specific algorithm for detecting the shoreline detection was developed by Visual C++ MFC (Microsoft Foundation Class). The analysis result estimated by aerial photo and satellite image was 1,391.0 m. The level of reliability was 98.1% for auto boundary detection when it compared with real surveying data.