• Korean Journal of Remote Sensing
• /
• v.35 no.5_1
• /
• pp.681-692
• /
• 2019

In this paper, a hybrid deep learning model, called 2D convolution with bidirectional long short-term memory (2DCBLSTM), is presented that can effectively combine both spatial and temporal features for crop classification. In the proposed model, 2D convolution operators are first applied to extract spatial features of crops and the extracted spatial features are then used as inputs for a bidirectional LSTM model that can effectively process temporal features. To evaluate the classification performance of the proposed model, a case study of crop classification was carried out using multi-temporal unmanned aerial vehicle images acquired in Anbandegi, Korea. For comparison purposes, we applied conventional deep learning models including two-dimensional convolutional neural network (CNN) using spatial features, LSTM using temporal features, and three-dimensional CNN using spatio-temporal features. Through the impact analysis of hyper-parameters on the classification performance, the use of both spatial and temporal features greatly reduced misclassification patterns of crops and the proposed hybrid model showed the best classification accuracy, compared to the conventional deep learning models that considered either spatial features or temporal features. Therefore, it is expected that the proposed model can be effectively applied to crop classification owing to its ability to consider spatio-temporal features of crops.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.29 no.4
• /
• pp.632-640
• /
• 2002

The objective of the study was to apply the vibration technique to reduce the viscosity of bonding adhesives and thereby compare the bond strength and resin penetration into dentinal tubules achieved with those gained using the conventional technique. Eighty-eight noncarious extracted human permanent molar teeth were sectioned to remove the coronal enamel and were embedded in 1-inch PVC pipe with acrylic resin. The occlusal surfaces were placed so that the tooth and the embedding medium were at the same level to form one flat surface, and the samples were subsequently polished with silicon carbide abrasive papers. The samples were randomly assigned to 4 groups(n=22). On Group 1 and 2, Single Bond(3M-ESPE, St. Paul, USA) was used, and on Group 3 and 4, One-Step(Bisco Inc., Schaumburg, USA) was used, and each was applied according to its manufacturer's instructions. For Group 2 and Group 4, vibration was applied with ultrasonic scaler for 10 seconds, and the adhesive was light-cured for 10 seconds. Resin composite was condensed on to the prepared surface in two increments using a mold kit(Ultradent Products Inc., USA) and each was light-cured for 40 seconds. After 24 hours in tap water at room temperature the specimens were thermocycled, and shear bond strengths were measured with a universal testing machine(Instron 4465, Canton, USA). To investigate infiltration patterns of the adhesive materials, the surface of specimen was examined with scanning electron microscope. The results were as follows. 1. The shear bond strengths of vibration groups(Group 2, Group 4) were significantly greater than those of the non-vibration groups(Group 1, Group 3)(p<0.05). 2. The shear bond strengths of Single Bond and One-Step were not significantly different (p>0.05). 3. The vibration groups showed greater number of resin tags in tubules and lateral branches under SEM.

• The Journal of Korean Academy of Prosthodontics
• /
• v.58 no.3
• /
• pp.177-184
• /
• 2020

Purpose: The purpose of this study was to compare and evaluate the tensile bond strength of chairside reline resin to denture base resin fabricated by different methods (subtractive manufacturing, additive manufacturing, and conventional heat-curing). Materials and methods: Denture base specimens were fabricated as cuboid specimens with a width of 25 mm × length 25 mm × height 3 mm by subtractive manufacturing (VITA VIONIC BASE), additive manufacturing (NextDent Base) and conventional heat-curing (Lucitone 199). After storing the specimens in distilled water at 37℃ for 30 days and drying them, they were relined with polyethyl methacrylate (PEMA) chairside reline resin (REBASE II Normal). The subtractive and additive manufacturing groups were set as the experimental group, and the heat-curing group was set as the control group. Ten specimens were prepared for each group. After storing all bound specimens in distilled water at 37℃ for 24 hours, the tensile bond strength between denture bases and chairside reline resin was measured by a universal testing machine at a crosshead speed of 10 mm/min. The fracture pattern of each specimen was analyzed and classified into adhesive failure, cohesive failure, and mixed failure. Tensile bond strength, according to the fabrication method, was analyzed by 1-way ANOVA and Bonferroni's method (α=.05). Results: Mean tensile bond strength of the heat-curing group (2.45 ± 0.39 MPa) and subtractive manufacturing group (2.33 ± 0.39 MPa) had no significant difference (P>.999). The additive manufacturing group showed significantly lower tensile bond strength (1.23 ± 0.36 MPa) compared to the other groups (P<.001). Most specimens of heat-curing and subtractive manufacturing groups had mixed failure, but mixed failure and adhesive failure showed the same frequency in additive manufacturing group. Conclusion: The mean tensile bond strength of the subtractive manufacturing group was not significantly different from the heat-curing group. The additive manufacturing group showed significantly lower mean tensile bond strength than the other two groups.

• Journal of the Korean Society of Propulsion Engineers
• /
• v.20 no.4
• /
• pp.40-49
• /
• 2016

This study aims to confirm the characteristics of low swirl combustion at our low swirl model combustor. To do it, it is experimentally conducted by evaluating the flame shape, stability region and emissions according to the swirl angle. The most significant feature of low swirl combustion is a occurrence of lifted flame. Such lifted flames happen to combine exquisitely propagating feature of premixed flame with diverging flow. This feature of lifted flame was confirmed through a velocity flow field and visualized the flame in this model combustor. The visualized flame was classified according to the thermal power and equivalence ratio. The variation study in swirl angles showed that the lean flammable limit could be extended only by swirl angles. Also, as the swirl angle increased, it was confirmed that the NOx and CO emissions were decreased due to the mixing enhancement and shorter resident time.

• Journal of the Korean Academy of Child and Adolescent Psychiatry
• /
• v.7 no.2
• /
• pp.167-178
• /
• 1996

Objects : This study was conducted to investigate the relationships between parenting style and depressive trait in adolescents. Methods : 287 Middle school students were asked to complete self-report questionnaires including the Parental Bonding Instrument(PBI) and the Children's Depression Inventory (CDI). Results : Relations between two sets of variables were examined by means of the Pearson's correlation coefficient. It was suggested that the relationship with their mothers had more influence on the depressive trait of adolescents. The mean scores of children's depression inventory were significantly higher in group whose parents had been overprotective and less careful, than whose parents had been careful and not overprotective. Conclusion : The results support that the adolescents' depression is significantly related to the parenting style. Among parental bonding types, the high care-low overprotection type was most favorable and the low care-high overprotection type was worst.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.41 no.3
• /
• pp.167-167
• /
• 2004

The system combined layered space-time processing and space-time trellis codes (STTC) provide high transmission rate as well as diversity and coding gain without bandwidth expansion. In this paper, two layered receiver structures are proposed. One is the LSTT-MMSE in which received bit streams are decoupled by interference nulling and then decoded by separate STTC decoders. The decoded outputs are cancelled from the received signal before advancing to the next layer detection. The other is LSTT-Whitening employing whitening rather than nulling. The receiver employing whitening process shows several advantages on diversity gain and the required number of receive antennas compare to the convolutional coded space-time processing. The proposed receivers use different decoding order scheme according to each interference suwression. The (4, 3) LSTT-Whitening receiver still achieves 1㏈ gain over the (4, 4) LSTT-MMSE and the (4, 4) coded layered space-time processing.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.41 no.3
• /
• pp.9-14
• /
• 2004

The system combined layered space-time processing and space-time trellis codes (STTC) provide high transmission rate as well as diversity and coding gain without bandwidth expansion. In this paper, two layered receiver structures are proposed. One is the LSTT-MMSE in which received bit streams are decoupled by interference nulling and then decoded by separate STTC decoders. The decoded outputs are cancelled from the received signal before advancing to the next layer detection. The other is LSTT-Whitening employing whitening rather than nulling. The receiver employing whitening process shows several advantages on diversity gain and the required number of receive antennas compare to the convolutional coded space-time processing. The proposed receivers use different decoding order scheme according to each interference suwression. The (4, 3) LSTT-Whitening receiver still achieves 1㏈ gain over the (4, 4) LSTT-MMSE and the (4, 4) coded layered space-time processing.

• The korean journal of orthodontics
• /
• v.40 no.5
• /
• pp.342-348
• /
• 2010

Objective: The aim of this study was to compare the shear bond strengths (SBS) of orthodontic brackets bonded to enamel with a self-etching primer after bleaching, desensitizer application and combined treatment. Methods: Forty-eight premolars were randomly divided into four groups, each with n = 12 premolar samples. The four groups were; Group1: 15% hydrogen-peroxide office bleaching agent (Illumin$\acute{e}$ Office-IO), Group 2: IO + BisBlock Oxalate Dentin-Desensitizer, Group 3: Bis Block Oxalate Dentin-Desensitizer, Group 4: No treatment (control). Twenty-four hours after bonding, the specimens were tested in SBS at a crosshead speed of 5 mm/min until the brackets debonded. The failure mode of the brackets was determined by a modified adhesive remnant index. Results: Bleaching, bleaching and desensitizer treatment, and desensitizer treatment alone all significantly reduced SBS of the orthodontic brackets ($p$ = 0.001). No statistically significant difference was found between Group 1, Group 2 and Group 3 (Group 1-Group 2, $p$ = 0.564; Group 1-Group 3, $p$ = 0.371; Group 2-Group 3, $p$ = 0.133). The predominant mode of failure for the treatment groups (Group1, Group 2 and Group 3) was at the enamel-adhesive interface leaving 100% of the adhesive on the bracket base. Conclusions: Bleaching and desensitizer treatment should be delayed until the completion of orthodontic treatment.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.25 no.4
• /
• pp.377-385
• /
• 2015

This paper proposes a novel forward vehicle tracking algorithm based on the WMIL(Weighted Multiple Instance Learning) equipped with a particle filter. In the proposed algorithm Haar-like features are used to train a vehicle object detector to be tracked and the location of the object are obtained from the recognition result. In order to combine both the WMIL to construct the vehicle detector and the particle filter, the proposed algorithm updates the object location by executing the propagation, observation, estimation, and selection processes involved in particle filter instead of finding the credence map in the search area for every frame. The proposed algorithm inevitably increases the computation time because of the particle filter, but the tracking accuracy was highly improved compared to Ababoost, MIL(Multiple Instance Learning) and MIL-based ones so that the position error was 4.5 pixels in average for the videos of national high-way, express high-way, tunnel and urban paved road scene.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.10 no.1
• /
• pp.38-45
• /
• 2017

Attacking of computer and network is increasing as information processing technology heavily depends on computer and network. To prevent the attack of system and network, host and network based intrusion detection system has developed. But previous rule based system has a lot of difficulties. For this reason demand for developing a intrusion detection system which detects and cope with the attack of system and network resource in real time. In this paper we develop a real time intrusion detection system which is combination of APEX and LS-SVM classifier. Proposed system is for nonlinear data and guarantees convergence. While real time processing system has its advantages, such as memory efficiency and allowing a new training data, it also has its disadvantages of inaccuracy compared to batch way. Therefore proposed real time intrusion detection system shows similar performance in accuracy compared to batch way intrusion detection system, it can be deployed on a commercial scale.