• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.19 no.5
• /
• pp.178-188
• /
• 2020

This paper proposes a lane recognition CNN network using split-attention network as a backbone to extract feature. Split-attention is a method of assigning weight to each channel of a feature map in the CNN feature extraction process; it can reliably extract the features of an image during the rapidly changing driving environment of a vehicle. The proposed deep neural networks in this paper were trained and evaluated using the Tusimple data set. The change in performance according to the number of layers of the backbone network was compared and analyzed. A result comparable to the latest research was obtained with an accuracy of up to 96.26, and FN showed the best result. Therefore, even in the driving environment of an actual vehicle, stable lane recognition is possible without misrecognition using the model proposed in this study.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.41 no.3
• /
• pp.115-124
• /
• 2004

The content-based retrieval using low-level features can hardly provide the retrieval result that corresponds with conceptual demand of user for intelligent retrieval. Video includes not only moving picture data, but also audio or closed-caption data. Knowledge-based video retrieval is able to provide the retrieval result that corresponds with conceptual demand of user because of performing automatic indexing with such a variety data. In this paper, we present the knowledge-based video retrieval system using Korean closed-caption. The closed-caption is indexed by Korean keyword extraction system including the morphological analysis process. As a result, we are able to retrieve the video by using keyword from the indexing database. In the experiment, we have applied the proposed method to news video with closed-caption generated by Korean stenographic system, and have empirically confirmed that the proposed method provides the retrieval result that corresponds with more meaningful conceptual demand of user.

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• v.41 no.1
• /
• pp.52-56
• /
• 2015

Clinical features of masticator-space abscess (MSA) are very similar to those of parotitis or temporomandibular disorder (TMD), making early differential diagnosis difficult. Local causes of MSA include nerve block anesthesia, infection after tooth extraction, and trauma to the temporomandibular joint (TMJ); the systemic cause is immunodeficiency. Odontogenic causes account for most etiologies, but there are also unusual causes of MSA. A 66-year-old male patient visited the emergency room (ER) presenting with left-side TMJ pain three days after receiving an acupressure massage. He was tentatively diagnosed with conventional post-trauma TMD and discharged with medication. However, the patient returned to the ER with increased pain. At this time, his TMD diagnosis was confirmed. He made a third visit to the ER during which facial computed tomographic (CT) images were taken. CT readings identified an abscess or hematoma in the left masticator space. After hospitalizing the patient, needle aspiration confirmed pus in the infratemporal and temporal fossa. Antibiotics were administered, and the abscess was drained through an incision made by the attending physician. The patient's symptoms decreased, and he was discharged.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.3
• /
• pp.328-335
• /
• 2007

Ontology is represented by the shared concepts and relations among those. Many studies have been actively working on sharing human's knowledge with that of systems by using it. For a typical example, there is the design and implementation of ontology system for image understanding. However conventional studies on ontology-based image understanding have proposed not concrete methods but conceptual idea. In this paper, we propose an ontology-based image understanding system with following four processes: i)knowledge representation of a specific domain by the ontology, ii)feature extraction of objects through image processing and image analysis, iii)image interpretation by object features, and iv)reduction of ambiguity existing in image interpretation by ontology reasoning. We implement an image understanding system based on the proposed processed, and show the effectiveness of the proposed system from experimental results in a specific domain.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.6
• /
• pp.852-859
• /
• 2009

Normally, to extract the defect in TFT-LCD inspection system, the image is obtained by using line scan camera or area scan camera which is achieved by CCD or CMOS sensor. Because of the limited dynamic range of CCD or CMOS sensor as well as the effect of the illumination, these images are frequently degraded and the important features are hard to decern by a human viewer. In order to overcome this problem, the feature vectors in the image are obtained by using the average intensity difference between defect and background based on the weber's law and the standard deviation of the background region. The defect detection method uses non-linear SVM (Supports Vector Machine) method using the extracted feature vectors. The experiment results show that the proposed method yields better performance of defect classification methods over conveniently method.

• Maxillofacial Plastic and Reconstructive Surgery
• /
• v.14 no.3
• /
• pp.185-193
• /
• 1992

The calcifying odontogenic cyst was identified as a pathological entity by Gorlin & his associates in 1962. This lesion is one of the rarest and most disputable cysts in the oral region. The calcifying odontogenic cyst has variable clinical and radiological features. We review the previous literatures and report 2 cases of calcifying odontogenic cyst at Department of Oral and Maxillofacial Surgery, Kyung-Hee University. The 1st case was as follows. The patient vas 22 year old female. The past dental history revealed extraction of prolonged retained #73 tooth about 15days ago. She complained a painful swelling on the lower anterior teeth area. There were chin and vestibular swelling on the lower anterior teeth area, tenderness and missing of #33 tooth. The radiograph revealed well-demarcated unilocular radiolucency containing radiopaque calcific flecks around impacted #33 tooth. The clinical diagnosis was COC, so surgical enucleation was done. There was no recurrence and COC was confirmed by pathologist. The second case was as follows. The patient was 72 year old male. The past history revealed inactive tuberculosis, bronchial asthma and denture construction. The chief complaint was rapidly growing mass on the lower left anterior edentulous area. The clinical findings were chin swelling protruding mass with surface ulceration, fluctuation and a few bloody fluid in aspiration. The radiograph revealed well-demarcated radiolucency mimiking the residual cyst. The biopsy result was COC. The surgical excision was done, but the lesion was recurred 10 months later. The treatment was surgical excision with aggressive peripheral bone grinding and FTSG form groin area. There was no problem during the postoperative period.

• Journal of IKEEE
• /
• v.24 no.1
• /
• pp.200-207
• /
• 2020

In this paper, an area-efficient feature extractor was proposed for security surveillance radar systems and FPGA-based implementation results were presented. In order to reduce the memory requirements, features extracted from Doppler profile for FFT window-size are used, while those extracted from total spectrogram for frame-size are excluded. The proposed feature extractor was design using Verilog-HDL and implemented with Xilinx Zynq-7000 FPGA device. Implementation results show that the proposed design can reduce the logic slice and memory requirements by 58.3% and 98.3%, respectively, compared with the existing research. In addition, security surveillance radar system with the proposed feature extractor was implemented and experiments to classify car, bicycle, human and kickboard were performed. It is confirmed from these experiments that the accuracy of classification is 93.4%.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39C no.9
• /
• pp.811-819
• /
• 2014

Activity recognition is a key component in identifying the context of a user for providing services based on the application such as medical, entertainment and tactical scenarios. Instead of applying numerous sensor devices, as observed in many previous investigations, we are proposing the use of smartphone with its built-in multimodal sensors as an unobtrusive sensor device for recognition of six physical daily activities. As an improvement to previous works, accelerometer, gyroscope and magnetometer data are fused to recognize activities more reliably. The evaluation indicates that the IBK classifier using window size of 2s with 50% overlapping yields the highest accuracy (i.e., up to 99.33%). To achieve this peak accuracy, simple time-domain and frequency-domain features were extracted from raw sensor data of the smartphone.

• Smart Structures and Systems
• /
• v.22 no.4
• /
• pp.433-440
• /
• 2018

Pavements porous concrete is a noble structure design in the urban management development generally enabling water to be permeated within its structure. It has also capable in the same time to cater dynamic loading. During the technology development, the quality and quantity of waste materials have led to a waste disposal crisis. Using recycled materials (secondary) instead of virgin ones (primary) have reduced landfill pressure and extraction demanding. This study has reviewed the waste materials (Recycled crushed glass (RCG), Steel slag, Steel fiber, Tires, Plastics, Recycled asphalt) used in the pavement porous concretes and report their respective mechanical, durability and permeability functions. Waste material usage in the partial cement replacement will cause the concrete production cost to be reduced; also, the concretes' mechanical features have slightly affected to eliminate the disposal waste materials defects and to use cement in Portland cement (PC) production. While the cement has been replaced by different industrial wastes, the compressive strength, flexural strength, split tensile strength and different PC permeability mixes have depended on the waste materials' type applied in PC production.

• Maxillofacial Plastic and Reconstructive Surgery
• /
• v.13 no.1
• /
• pp.69-76
• /
• 1991

A 9-year old boy suffering from cleidocranial dysplasia associated with impacted 4 supernumerary teeth and unerupted all permanent teeth is presented with his mother. The pedigree showed autosomal dominant pattern of inheritance, and the raiographic features of them were very similar in clavicle, skull, vertebrae, peivis and extremities. Although almost of the skeleton was involved with this syndrome, they did not recognize any other problem but except dental problem. In mother, who was wearing removable partial dentures leaving 24 impacted teeth in her jaws, the radiographic abnormalities like cystic lesion were not detected. And in the son, who showed impacted 4 supernumerary and all permanent teeth, we have attempted surgical extraction of the supernumerary teeth and periodic surgical opening of the alveolar bone covering the permanent dentition to induce the eruption of permanent teeth at the proper position, Orthodontic treatment has also been combined to correct class III malocclusion state.