The Re-Identification(Re-ID) is one of the most popular researches in the field of computer vision due to a variety of applications. To achieve a high-level re-identification performance, recently other methods have developed the deep learning based networks that are specialized for only person or vehicle. However, most of the current methods are difficult to be used in real-world applications that require re-identification of both person and vehicle at the same time. To overcome this limitation, this paper proposes a deep neural network learning method that combines triplet and softmax loss to improve performance and re-identify people and vehicles simultaneously. It's possible to learn the detailed difference between the identities(IDs) by combining the softmax loss with the triplet loss. In addition, weights are devised to avoid bias in one-side loss when combining. We used Market-1501 and DukeMTMC-reID datasets, which are frequently used to evaluate person re-identification experiments. Moreover, the vehicle re-identification experiment was evaluated by using VeRi-776 and VehicleID datasets. Since the proposed method does not designed for a neural network specialized for a specific object, it can re-identify simultaneously both person and vehicle. To demonstrate this, an experiment was performed by using a person and vehicle re-identification dataset together.

In this paper, we present a novel algorithm to generate a 3D model of patient-specific orbital implant, which is finally produced by the 3D printer. Given CT (computed tomography) scan data of the defective orbital wall or floor, we compose the depth image of the defect site by using the depth buffering, which is a computer graphics technology. From the depth image, we compute the 3D surface which fills the broken part by interpolating the points around the broken part. By thickening the 3D surface, we get the 3D volume mesh of the orbital implant. Our algorithm generates the patient-specific orbital implant whose shape is accurately coincident to the broken part of the orbit. It provides the significant time efficiency for manufacturing the implant with supporting high user convenience.

Outlier detection methods help one to identify unusual instances in data that may correspond to erroneous, exceptional, or surprising events or behaviors. This work studies conditional outlier detection, a special instance of the outlier detection problem, in the context of incorrect data label identification. Unlike conventional (unconditional) outlier detection methods that seek abnormalities across all data attributes, conditional outlier detection assumes data are given in pairs of input (condition) and output (response or label). Accordingly, the goal of conditional outlier detection is to identify incorrect or unusual output assignments considering their input as condition. As a solution to conditional outlier detection, this paper proposes the ratio-based outlier scoring (ROS) approach and its variant. The propose solutions work by adopting conventional outlier scores and are able to apply them to identify conditional outliers in data. Experiments on synthetic and real-world image datasets are conducted to demonstrate the benefits and advantages of the proposed approaches.

Gait recognition can identify people's identity from a long distance, which is very important for improving the intelligence of the monitoring system. Among many human features, gait features have the advantages of being remotely available, robust, and secure. Traditional gait feature extraction, affected by the development of behavior recognition, can only rely on manual feature extraction, which cannot meet the needs of fine gait recognition. The emergence of deep convolutional neural networks has made researchers get rid of complex feature design engineering, and can automatically learn available features through data, which has been widely used. In this paper,conduct feature metric learning in the three-dimensional space by combining the three-dimensional convolution features of the gait sequence and the Siamese structure. This method can capture the information of spatial dimension and time dimension from the continuous periodic gait sequence, and further improve the accuracy and practicability of gait recognition.

The Alzheimer's disease (AD) is a neurodegenerative disease commonly found in the elderly individuals. It is one of the most common forms of dementia; patients with AD suffer from a degradation of cognitive abilities over time. To correctly diagnose AD, compuated-aided system equipped with automatic classification algorithm is of great importance. In this paper, we propose a novel deep learning based classification algorithm that takes advantage of MRI biomarker images including brain areas of hippocampus and cerebrospinal fluid for the purpose of improving the AD classification performance. In particular, we develop a new approach that effectively applies MRI biomarker patch images as input to 3D Deep Convolution Neural Network. To integrate multiple classification results from multiple biomarker patch images, we proposed the effective confidence score fusion that combine classification scores generated from soft-max layer. Experimental results show that AD classification performance can be considerably enhanced by using our proposed approach. Compared to the conventional AD classification approach relying on entire MRI input, our proposed method can improve AD classification performance of up to 10.57% thanks to using biomarker patch images. Moreover, the proposed method can attain better or comparable AD classification performances, compared to state-of-the-art methods.

Ever since media contents took over the world, the importance of typography has increased, and the influence of fonts has be n recognized. Nevertheless, the current Hangul font system is very poor and is provided passively, so it is practically impossible to understand and utilize all the shape characteristics of more than six thousand Hangul fonts. In this paper, the characteristics of Hangul font shapes were selected based on the Hangul structure of similar fonts. The stroke element detection training was performed by fine tuning Faster R-CNN Inception v2, one of the deep learning object detection models. We also propose a system that automatically extracts the stroke element characteristics from characters by introducing an automatic extraction algorithm. In comparison to the previous research which showed poor accuracy while using SVM(Support Vector Machine) and Sliding Window Algorithm, the proposed system in this paper has shown the result of 10 % accuracy to properly detect and extract stroke elements from various fonts. In conclusion, if the stroke element characteristics based on the Hangul structural information extracted through the system are used for similar classification, problems such as copyright will be solved in an era when typography's competitiveness becomes stronger, and an automated process will be provided to users for more convenience.

We present in this work a segmentation method of E. coli bacterial images generated via phase contrast microscopy using a deep learning based hybrid feature generation. Unlike conventional machine learning methods that use the hand-crafted features, we adopt the denoising autoencoder in order to generate a precise and accurate representation of the pixels. We first construct a hybrid vector that combines original image, difference of Gaussians and image gradients. The created hybrid features are then given to a deep autoencoder that learns the pixels' internal dependencies and the cells' shape and boundary information. The latent representations learned by the autoencoder are used as the inputs of a softmax classification layer and the direct outputs from the classifier represent the coarse segmentation mask. Finally, the classifier's outputs are used as prior information for a graph partitioning based fine segmentation. We demonstrate that the proposed hybrid vector representation manages to preserve the global shape and boundary information of the cells, allowing to retrieve the majority of the cellular patterns without the need of any post-processing.

Aiming at the problem of complex feature extraction and low accuracy in human action recognition, this paper proposed a network structure combining batch normalization algorithm with GoogLeNet network model. Applying Batch Normalization idea in the field of image classification to action recognition field, it improved the algorithm by normalizing the network input training sample by mini-batch. For convolutional network, RGB image was the spatial input, and stacked optical flows was the temporal input. Then, it fused the spatio-temporal networks to get the final action recognition result. It trained and evaluated the architecture on the standard video actions benchmarks of UCF101 and HMDB51, which achieved the accuracy of 93.42% and 67.82%. The results show that the improved convolutional neural network has a significant improvement in improving the recognition rate and has obvious advantages in action recognition.

In this paper, we propose AnoVid, an automated video annotation tool based on deep neural networks, that automatically generates various meta data for each scene or shot in a long drama video containing rich elements. To this end, a novel meta data schema for drama video is designed. Based on this schema, the AnoVid video annotation tool has a total of six deep neural network models for object detection, place recognition, time zone recognition, person recognition, activity detection, and description generation. Using these models, the AnoVid can generate rich video annotation data. In addition, AnoVid provides not only the ability to automatically generate a JSON-type video annotation data file, but also provides various visualization facilities to check the video content analysis results. Through experiments using a real drama video, "Misaeing", we show the practical effectiveness and performance of the proposed video annotation tool, AnoVid.

Dataset is getting more import to develop a learning based algorithm. Quality of the algorithm definitely depends on dataset. So we introduce new dataset over 200 thousands images which are fully labeled multi modal sensor data. Proposed dataset was designed and constructed for researchers who want to develop detection, tracking, and action classification in outdoor environment for surveillance scenarios. The dataset includes various images and multi modal sensor data under different weather and lighting condition. Therefor, we hope it will be very helpful to develop more robust algorithm for systems equipped with difference kinds of sensors in outdoor application. Case studies with the proposed dataset are also discussed in this paper.

In this paper, we propose a novel knowledge distillation algorithm to create an compressed deep ensemble network coupled with the combined use of local and global features of face images. In order to transfer the capability of high-level recognition performances of the ensemble deep networks to a single deep network, the probability for class prediction, which is the softmax output of the ensemble network, is used as soft target for training a single deep network. By applying the knowledge distillation algorithm, the local feature informations obtained by training the deep ensemble network using facial subregions of the face image as input are transmitted to a single deep network to create a so-called compressed ensemble DCNN. The experimental results demonstrate that our proposed compressed ensemble deep network can maintain the recognition performance of the complex ensemble deep networks and is superior to the recognition performance of a single deep network. In addition, our proposed method can significantly reduce the storage(memory) space and execution time, compared to the conventional ensemble deep networks developed for face recognition.

The biggest feature of the online environment is anonymity. So it is difficult to identify the sender accurately in Email. So we use PGP to enhance the security of email. PGP is an e-mail encryption program that has become the standard for email security worldwide. But PGP has a weak point in security. PGP can identify users, but it is difficult to verify reliability. In PGP, the more reliable other users guarantee that user are, the more reliable they are. In the previous study, the PGP structure was implemented by utilizing the block chain to lay the foundation, and in this paper, the structure of the previous article is applied to the e-mail environment to measure up the sender's reliability, and a system is proposed that allows users to distinguish the reliability of the message.

In this study, in order to improve the implantable bone conduction transducer of the prototype proposed by Shin et al., the effect of the element parameters of the transducer on the frequency characteristics was analyzed using electromagnetic and mechanical vibration analysis. Electromagnetic analysis was performed on the size of the permanent magnet and the distance between the metal plate and the coil to derive an optimal structure that generates the maximum Lorentz force. In addition, mechanical vibration analysis was performed on the cantilever structure of the vibrational membrane in order to minimize the distortion of the transducer and to have a frequency characteristic suitable for conductive hearing loss compensation. The frequency characteristics of the transducer of the optimal structure derived through finite element method were compared with the simulation results of the previous transducer. As a result, the output magnitude (displacement) of the transducer designed with the optimal structure generated an average 8.8 times higher than the previous transducer, and the resonance frequency was generated at 0.9 kHz.

'Instagrammable' is a new term which means a photo or a series of pictures are worth posting on Instagram. Since Instagram is an image-oriented social media platform, it is important to give users proper awareness through images in order to be an instagrammable post. In this study, we explored the proper delivery method of messages within instagrammable posts through the use of hashtags(#). Specifically, we paid attention to the use of 'Card News', which involves a series of images that form a short narrative. Hashtags play an important role that they often describe sharing intention of the post, and we found analyzing the use of hashtags in Card News posts is a good indicator of users' Instagram activities. Currently, there are more than 580k posts are found with the search keyword Card News, and the number is increasing. In this study, we collected and analyzed more than 50k hashtags on Instagram to explore how news stories are posted from both the general users and news media accounts. Furthermore, we conducted interviews with journalists to analyze how news media are making use of Instagram as a legitimate place to share news stories with impact.

In hospitals, nurses are subjectively determining the urine status to check the kidneys and circulatory system of patients whose statuses are related to patients with kidney disease, critically ill patients, and nursing homes before and after surgery. To improve this problem, this paper proposes a urine spectrum analysis system which clusters urine test results based on a hybrid machine learning model consists of unsupervised learning and supervised learning. The proposed system clusters the spectral data using unsupervised learning in the first part, and classifies them using supervised learning in the second part. The results of the proposed urine spectrum analysis system using a mixed model are evaluated with the results of pure supervised learning. This paper is expected to provide better services than existing medical services to patients by solving the shortage of nurses, shortening of examination time, and subjective evaluation in hospitals.

High Level Architecture/Run-Time Infrastructure (HLA/RTI) is used to connect individual simulators on networks in order to interoperate heterogeneous simulators. In defense domain, Ship Air Defense Model (SADM) and Extended Air Defense Simulation (EADSIM) are two of most advanced simulation tools. To interoperate these SADM and EADSIM, this paper attempts to use HLA/RTI that helps to support a Composite Combat Mission Planning Simulation Environment (CCMPSE). The CCMPSE allows us to analyze a group of simulations for comprehensive and accurate experiments. For the first time, this paper analyzes guided missile simulations in EADSIM and SADM by comparing related simulation models in their parameters and considerations. It presents characteristics of these models in view of guided missile simulation perspectives. For the contributions of this paper, it provides insights to select guided missiles between SADM and EADSIM on the CCMPSE according to specific simulation purposes.

Aviation ICT technology is a convergence technology between aviation and electronics, and has a wide variety of applications, including navigation and education. Among them, in the field of aerial pilot training, there are many problems such as the possibility of accidents during training and the lack of coping skills for various situations. This raises the need for a simulated pilot training system similar to actual training. In this paper, pilot training data were collected in pilot training system using VR/AR to increase immersion in flight training, and Customized Pilot Training Platform with Collaborative Deep Learning in VR/AR Environment that can recommend effective training courses to pilots is proposed. To verify the accuracy of the recommendation, the performance of the proposed collaborative deep learning algorithm with the existing recommendation algorithm was evaluated, and the flight test score was measured based on the pilot's training data base, and the deviations of each result were compared. The proposed service platform can expect more reliable recommendation results than previous studies, and the user survey for verification showed high satisfaction.

This paper proposes an obstacle avoidance method using a depth polar grid. Depth information is a crucial factor for determining the safe path for collision-free navigation of unmanned aerial vehicles (UAVs) as it can perceive the distance to the obstacles effectively. However, the existing depth-camera-based approaches for obstacle avoidance require computational y expensive path planning algorithms. We propose a simple navigation method using the polar-grid of the depth information obtained from the camera with narrow field-of-view(FOV). The effectiveness of the approach was validated by a series of experiments using software-in-the-loop simulation in a realistic outdoor environment. The experimental results show that the proposed approach successfully avoids obstacles using a single depth camera with limited FOV.

The need for customized services providing real-time subway information through mobile devices for the blind using subway as their primary transportation method is becoming more essential. Therefore, this study is aimed at developing a mobile service to provide real-time subway information by displaying with braille and utilizing Dot Watch. The process of real-time subway information guidance mobile service is as follows: the route confirmation stage, the ride waiting & judgment stage, the on-the-go stage, and lastly, the off-road phase & wrong path state. The service provides real-time appropriate information to the visually impaired people for each situation. It receives subway information through API provided by Seoul Metropolitan Government and ODSay API. The service is developed and coded with in Java using Android Studio, and the communicating method with Dot Watch was done using Bluetooth. The usability evaluation was conducted in terms of efficiency, effectiveness, and satisfaction. The evaluation methods mentioned above were carried out by testees, and feedback was given. They assessed that the information provided was useful but still unstable to use the system. It is expected that actual usability will increase if our study is supplemented using voice output function.

Based on Maslow's hierarchy of needs theory, this research only employs whether to meet consumer needs as a measure, and focuses on the Collaboration Series between Uniqlo and KAWS to discuss the influence of the collaboration between fashion brands and artists on consumer needs. What's more, this paper also proposes whether the collaboration with artists as brands can meet the higher needs of consumers and influence new consumer groups. Finally, it is concluded that the collaboration between fashion brands and artists plays an active role in meeting the higher needs of consumers and can expand new consumer groups, which will provide reference and theoretical basis for enterprises to conduct collaboration with art, incorporate the needs of consumers into the scope of planning, and consider the collaboration with artists more comprehensively.

Train to Busan(2016) and Seoul Station(2016) contain trans-media properties in terms of the expansion of 'possible worlds' from the original story, and show the cultural and economic potential of the domestic trans-media contents with industrial systems added to the artist's competence of Dir. Yeon Shang-ho, an animator and film director. But they also have limits of the lack of continuity in time and characters between each media, various derived contents and suitable marketing strategy. For the development of Korean trans-media contents, it is necessary to extend the range of contents, find changes in production time and style, and improve the distribution environment to encourage a user to integrate contents from each platforms.