• Journal of Convergence for Information Technology
• /
• v.9 no.12
• /
• pp.17-23
• /
• 2019

In this paper, I propose a system development methodology that accepts images taken by camera attached to drone in real time while controlling mini drone and recognize and confirm the face of certain person. For the development of this system, OpenCV, Python related libraries and the drone SDK are used. To increase face recognition ratio of certain person from real-time drone images, it uses Deep Learning-based facial recognition algorithm and uses the principle of Triples in particular. To check the performance of the system, the results of 30 experiments for face recognition based on the author's face showed a recognition rate of about 95% or higher. It is believed that research results of this paper can be used to quickly find specific person through drone at tourist sites and festival venues.

• International Journal of Computer Science & Network Security
• /
• v.21 no.8
• /
• pp.288-296
• /
• 2021

Automated face recognition in a runtime environment is gaining more and more important in the fields of surveillance and urban security. This is a difficult task keeping in mind the constantly volatile image landscape with varying features and attributes. For a system to be beneficial in industrial settings, it is pertinent that its efficiency isn't compromised when running on roads, intersections, and busy streets. However, recognition in such uncontrolled circumstances is a major problem in real-life applications. In this paper, the main problem of face recognition in which full face is not visible (Occlusion). This is a common occurrence as any person can change his features by wearing a scarf, sunglass or by merely growing a mustache or beard. Such types of discrepancies in facial appearance are frequently stumbled upon in an uncontrolled circumstance and possibly will be a reason to the security systems which are based upon face recognition. These types of variations are very common in a real-life environment. It has been analyzed that it has been studied less in literature but now researchers have a major focus on this type of variation. Existing state-of-the-art techniques suffer from several limitations. Most significant amongst them are low level of usability and poor response time in case of any calamity. In this paper, an improved face recognition system is developed to solve the problem of occlusion known as FRS-OCC. To build the FRS-OCC system, the color and texture features are used and then an incremental learning algorithm (Learn++) to select more informative features. Afterward, the trained stack-based autoencoder (SAE) deep learning algorithm is used to recognize a human face. Overall, the FRS-OCC system is used to introduce such algorithms which enhance the response time to guarantee a benchmark quality of service in any situation. To test and evaluate the performance of the proposed FRS-OCC system, the AR face dataset is utilized. On average, the FRS-OCC system is outperformed and achieved SE of 98.82%, SP of 98.49%, AC of 98.76% and AUC of 0.9995 compared to other state-of-the-art methods. The obtained results indicate that the FRS-OCC system can be used in any surveillance application.

• Journal of Dental Anesthesia and Pain Medicine
• /
• v.19 no.1
• /
• pp.29-36
• /
• 2019

Background: An inability to cope with threatening dental stimuli, i.e., sight, sound, and sensation of airotor, manifests as anxiety and behavioral management problems. Behavior modification techniques involving pre-exposure to dental equipment will give children a first-hand experience of their use, sounds, and clinical effects. The aim of this study was to compare the techniques of Tell-Show-Play-doh, a smartphone dentist game, and a conventional Tell-Show-Do method in the behavior modification of anxious children in the dental operatory. Methods: Sixty children in the age group of 4-8 years, with Frankl's behavior rating score of 2 or 3, requiring Class I and II cavity restorations were divided into three groups. The groups were Group 1: Tell-Show-Play-doh; Group 2: smartphone dentist game; and Group 3: Tell-Show-Do technique and each group comprised of 20 children. Pulse rate, Facial Image Scale (FIS), Frankl's behavior rating scale, and FLACC (Face, Leg, Activity, Cry, Consolability) behavior scales were used to quantify anxious behavior. Operator compliance was recorded through a validated questionnaire. Results: The results showed lower mean pulse rates, lower FIS and FLACC scores, higher percentage of children with Frankl's behavior rating score of 4, and better operator compliance in both the Tell-Show-Play-doh and smartphone dentist game groups than in the conventional Tell-Show-Do group. Conclusion: The Tell-Show-Play-doh and smartphone dentist game techniques are effective tools to reduce dental anxiety in pediatric patients.

• Journal of Dental Anesthesia and Pain Medicine
• /
• v.19 no.4
• /
• pp.227-233
• /
• 2019

Background: Dental anxiety in children is a major barrier in patient management. If dental anxiety in pediatric patients is assessed during the first visit, it will not only aid in management but also help to identify patients who are in need of special care to deal with their fear. Nowadays, children and adults are highly interested in multimedia and are closely associated with them. Children usually prefer motion pictures on electronic devices than still cartoons on paper. Therefore, this study was conducted to evaluate a newly designed scale, the animated emoji scale (AES), which uses motion emoticons/animojis to assess dental anxiety in children during their first dental visit, and compare it with the Venham picture test (VPT) and facial image scale (FIS). Methods: The study included 102 healthy children aged 4-14 years, whose dental anxiety was measured using AES, VPT, and FIS during their first dental visit, and their scale preference was recorded. Results: The mean anxiety scores measured using AES, FIS, and VPT, represented as $mean{\pm}SD$, were $1.78{\pm}1.19$, $1.93{\pm}1.23$, and $1.51{\pm}1.84$, respectively. There was significant difference in the mean anxiety scores between the three scales (Friedman test, P < 0.001). The Pearson's correlation test showed a very strong correlation (0.73) between AES and VPT, and a strong correlation between AES and FIS (0.88), and FIS and VPT (0.69), indicating good validity of AES. Maximum number of children (74.5%) preferred AES. Conclusion: The findings of this study suggest that the AES is a novel and child-friendly tool for assessing dental anxiety in children.

• The Korean Journal of Applied Statistics
• /
• v.32 no.1
• /
• pp.41-68
• /
• 2019

Multi-view learning considers data from various viewpoints as well as attempts to integrate various information from data. Multi-view learning has been studied recently and has showed superior performance to a model learned from only a single view. With the introduction of deep learning techniques to a multi-view learning approach, it has showed good results in various fields such as image, text, voice, and video. In this study, we introduce how multi-view learning methods solve various problems faced in human behavior recognition, medical areas, information retrieval and facial expression recognition. In addition, we review data integration principles of multi-view learning methods by classifying traditional multi-view learning methods into data integration, classifiers integration, and representation integration. Finally, we examine how CNN, RNN, RBM, Autoencoder, and GAN, which are commonly used among various deep learning methods, are applied to multi-view learning algorithms. We categorize CNN and RNN-based learning methods as supervised learning, and RBM, Autoencoder, and GAN-based learning methods as unsupervised learning.

• Archives of Plastic Surgery
• /
• v.46 no.3
• /
• pp.248-254
• /
• 2019

Background No head to head comparison is available between surgical lip lifting and upper lip filler injections to decide which technique yields the best results in patients. Despite the growing popularity of upper lip augmentation, its effect on societal perceptions of attractiveness, successfulness and overall health in woman is unknown. Methods Blinded casual observers viewed three versions of independent images of 15 unique patient lower faces for a total of 45 images. Observers rated the attractiveness, perceived success, and perceived overall health for each patient image. Facial perception questions were answered on a visual analog scale from 0 to 100, where higher scores corresponded to more positive responses. Results Two hundred and seventeen random observers with an average age of 47 years (standard deviation, 15.9) rated the images. The majority of observers were females (n=183, 84%) of white race (n=174, 80%) and had at least some college education (n=202, 93%). The marginal mean score for perceived attractiveness from the natural condition was 1.5 points (95% confidence interval [CI], 0.9-2.18) higher than perceived attractiveness from the simulated upper lip filler injection condition, and 2.6 points higher (95% CI, 1.95-3.24) than the simulated upper lip lift condition. There was a moderate to strong correlation between the scores of the same observer. Conclusions Simulated upper lip augmentation is amenable to social perception analysis. Scores of the same observer for attractiveness, successfulness, and overall health are strongly correlated. Overall, the natural condition had the highest scores in all categories, followed by simulated upper lip filler, and lastly simulated upper lip lift.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.4
• /
• pp.381-387
• /
• 2019

The limitations of applying a variety of artificial intelligence to the medical community are, first, subjective views, extensive interpreters and physical fatigue in interpreting the image of an interpreter's illness. And there are questions about how long it takes to collect annotated data sets for each illness and whether to get sufficient training data without compromising the performance of the developed deep learning algorithm. In this paper, when collecting basic images based on acne data sets, the selection criteria and collection procedures are described, and a model is proposed to classify data into small loss rates (5.46%) and high accuracy (96.26%) in the sequential structure. The performance of the proposed model is compared and verified through a comparative experiment with the model provided by Keras. Similar phenomena are expected to be applied to the field of medical and skin care by applying them to the acne classification model proposed in this paper in the future.

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.35 no.6
• /
• pp.267-273
• /
• 2021

Mibyeong (sub-health) is a concept that represents the sub-health in traditional East Asian medicine. Assuming that the nose sizes and shapes are related to respiratory function, in this study, we hypothesized that the nose size and shape features are related to the self-rated health (SRH) level and self-rated Mibyeong severity, and aimed to assess this relationship using a fully automated image analysis system. The nose size features were evaluated from the frontal and profile face images of 810 participants. The nose size features consisted of five length features, one area feature, and one volume feature. The level of SRH and the Mibyeong severity were determined using a questionnaire. The normalized nasal height was negatively associated with the self-rated health score (SRHS) (partial ρ = -0.125, p = 3.53E-04) and the Mibyeong score (MBS) (partial ρ = -.172, p = 9.38E-07), even after adjustment for sex, age, and body mass index. The normalized nasal volume (ρ = -.105, p = 0.003), the normalized nasal tip protrusion length (ρ = -.087, p = 0.014), and the normalized nares width (ρ = -.086, p = .015) showed significant correlation with the SRHS. The normalized nasal area (ρ = -.118, p = 0.001), the normalized nasal volume (ρ = -.107, p = .002) showed significant correlation with the MBS. The wider, longer, and larger the nose, the lower the SRHS and MBS, indicating that health status can be estimated based on the size and shape features of the nose.

• Imaging Science in Dentistry
• /
• v.52 no.1
• /
• pp.43-51
• /
• 2022

Purpose: The aim of this study was to evaluate the reliability of the Vectra M3 (3D Imaging System; Canfield Scientific, Parsippany, NJ, USA) in detecting chin asymmetry, and to assess whether the automatic markerless tracking function is reliable compared to manually plotting landmarks. Materials and Methods: Twenty subjects (18 females and 2 males) with a mean age of 42.5±10.5 years were included. Three-dimensional image acquisition was carried out on all subjects with simulated chin deviation in 4 stages (1-4 mm). The images were analyzed by 2 independent observers through manually plotting landmarks and by Vectra software auto-tracking mode. Repeated-measures analysis of variance and the Tukey post-hoc test were performed to evaluate the differences in mean measurements between the 2 operators and the software for measuring chin deviation in 4 stages. The intraclass correlation coefficient (ICC) was calculated to estimate the intra- and inter-examiner reliability. Results: No significant difference was found between the accuracy of manually plotting landmarks between observers 1 and 2 and the auto-tracking mode (P=0.783 and P=0.999, respectively). The mean difference in detecting the degree of deviation according to the stage was <0.5 mm for all landmarks. Conclusion: The auto-tracking mode could be considered as reliable as manually plotted landmarks in detecting small chin deviations with the Vectra^® M3. The effect on the soft tissue when constructing a known dental movement yielded a small overestimation of the soft tissue movement compared to the dental movement (mean value<0.5 mm), which can be considered clinically non-significant.

• Journal of Korea Multimedia Society
• /
• v.25 no.10
• /
• pp.1375-1385
• /
• 2022

In this paper, we propose a semi-supervised domain adaptation solution to deal with practical face recognition (FR) scenarios where a single face image for each target identity (to be recognized) is only available in the training phase. Main goal of the proposed method is to reduce the discrepancy between the target and the source domain face images, which ultimately improves FR performances. The proposed method is based on the Domain Adatation network (DAN) using an MMD loss function to reduce the discrepancy between domains. In order to train more effectively, we develop a novel loss function learning strategy in which MMD loss and cross-entropy loss functions are adopted by using different weights according to the progress of each epoch during the learning. The proposed weight adoptation focuses on the training of the source domain in the initial learning phase to learn facial feature information such as eyes, nose, and mouth. After the initial learning is completed, the resulting feature information is used to training a deep network using the target domain images. To evaluate the effectiveness of the proposed method, FR performances were evaluated with pretrained model trained only with CASIA-webface (source images) and fine-tuned model trained only with FERET's gallery (target images) under the same FR scenarios. The experimental results showed that the proposed semi-supervised domain adaptation can be improved by 24.78% compared to the pre-trained model and 28.42% compared to the fine-tuned model. In addition, the proposed method outperformed other state-of-the-arts domain adaptation approaches by 9.41%.