This paper proposes new frame rate up-conversion algorithms. Adaptive motion estimation based on block complexity information are used to obtain more accurate motion vectors. Because the information on block complexity is extracted from the motion estimation prediction size from the original frame, additional computational complexity is not imparted. In experimental results, the proposed algorithms provide robust frame interpolation performance for whole test sequences. Also, the computational complexity of the proposed algorithm is reduced to a benchmark algorithm.

Image Enhancement is used in various applications. Among them, unsharp masking methods can improve the contrast with a simple operation. However, it has problems of noise enhancement and halo effect caused by the use of a single filter. To solve this problems, adaptive processing using multi-scale and bilinear filters is being studied. These methods are effective for improving the halo effect, but it require a lot of calculation time. In this paper, we want to simplify adaptive filtering by generating a weight map based on local brightness. This weight map enables adaptive processing that eliminates the halo effect through a single multiplication operation. Through experiments, we confirmed the suppression of the halo effect through the result image of the proposed algorithm and existing algorithm.

This study attempts to detect a shot boundary in films(or dramas) based on the length of a sequence. As films or dramas use scene change effects a lot, the issues regarding the effects are more diverse than those used in surveillance cameras, sports videos, medical care and security. Visual techniques used in films are focused on the human sense of aesthetic therefore, it is difficult to solve the errors in shot boundary detection with the method employed in surveillance cameras. In order to define the errors arisen from the scene change effects between the images and resolve those issues, the mixed algorithm based upon color histogram, edge histogram, and optical flow was implemented. The shot boundary data from this study will be used when analysing the configuration of meaningful shots in sequences in the future.

Monitoring technology of machining has a long history since unmanned machining was introduced. Despite the long history, many researchers have presented new approaches continuously in this area. Sound based machine fault diagnosis is the process consisting of detecting automatically the damages that affect the machines by analyzing the sounds they produce during their operating time. The collected sound is corrupted by the surrounding work environment. Therefore, the most important part of the diagnosis is to find hidden elements inside the data that can represent the error pattern. This paper presents a feature extraction methodology that combines various digital signal processing and pattern recognition methods for the analysis of the sounds produced by tools. The magnitude spectrum of the sound is extracted using the Fourier analysis and the band-pass filter is applied to further characterize the data. Statistical functions are also used as input to the nonlinear classifier for the final response. The results prove that the proposed feature extraction method accurately captures the hidden patterns of the sound generated by the tool, unlike the conventional features. Therefore, it is shown that the proposed method can be applied to a sound based automatic diagnosis system.

In this paper, we propose an automatic composition method using time series embedding of RNN Auto-Encoder. RNN Auto-Encoder can learn existing songs and can compose new songs from the trained RNN decoder. If one song is fully trained in the RNN Auto-Encoder, the song is embedded into the vector values of RNN nodes in the Auto-Encoder. If we train a lot of songs and apply a specific vector to the decoder of Auto-Encoder, then we can obtain a new song that combines the features of trained multiple songs according to the given vector. From extensive experiments we could find that our method worked well and generated various songs by selecting of the composition vectors.

In this paper, we propose and implement a multi-level convolutional neural network (CNN) algorithm to identify the sexually explicit and lewdness of various image files, and verify its effectiveness by using unauthorized image files generated in the actual military. The proposed algorithm increases the accuracy by applying the convolutional artificial neural network step by step to minimize classification error between similar categories. Experimental data have categorized 20,005 images in the real field into 6 authorization categories and 11 non-authorization categories. Experimental results show that the overall detection rate is 99.51% for the image files. In particular, the excellence of the proposed algorithm is verified through reducing the identification error rate between similar categories by 64.87% compared with the general CNN algorithm.

In this paper, we propose a measurement method for height of white light scanning interferometer using deep learning. In order to measure the fine surface shape, a three-dimensional surface shape measurement technique is required. A typical example is a white light scanning interferometer. In order to calculate the surface shape from the measurement image of the white light scanning interferometer, the height of each pixel must be calculated. In this paper, we propose a neural network for height calculation and use virtual data generation method to train this neural network. The accuracy was measured by inputting 57 actual data to the neural network which had completed the learning. We propose two new functions for accuracy measurement. We have analyzed the cases where there are many errors among the accuracy calculation values, and it is confirmed that there are many errors when there is no interference fringe or outside the learned range. We confirmed that the proposed neural network works correctly in most cases. We expect better results if we improve the way we generate learning data.

In this paper, we propose an robust CAMShift method to track stage objects with a camera. In order to solve the problem of tracking object misdetection in existing CAMShift technique, MBR region is detected to separate the background and the subject, and the subject size of the region of interest is calculated to solve the problem of erroneously detecting a large region having a similar color distribution ratio. Also, by applying the color corelogram and MB-LBP to the part that can not be solved by the color ratio and the size limitation, accurate texture tracking is enabled by reflecting the texture characteristics. Experimental results show that the proposed method has good tracking performance for objects that do not deviate from the size of the subject set in the area of interest and accurately extracts the texture characteristics of different subjects with similar color distribution ratios.

In this paper, we report detection and recognition of vehicle logo from images captured from street CCTV. Image data includes both the front and rear view of the vehicles. The proposed method is a two-step process which combines image preprocessing and faster region-based convolutional neural network (R-CNN) for logo recognition. Without preprocessing, faster R-CNN accuracy is high only if the image quality is good. The proposed system is focusing on street CCTV camera where image quality is different from a front facing camera. Using perspective transformation the top view images are transformed into front view images. In this system, the detection and accuracy are much higher as compared to the existing algorithm. As a result of the experiment, on day data the detection and recognition rate is improved by 2% and night data, detection rate improved by 14%.

In this paper, we propose a security framework for a cluster drones network using the MAVLink (Micro Air Vehicle Link) application protocol based on FANET (Flying Ad-hoc Network), which is composed of ad-hoc networks with multiple drones for IoT services such as remote sensing or disaster monitoring. Here, the drones belonging to the cluster construct a FANET network acting as WTRP (Wireless Token Ring Protocol) MAC protocol. Under this network environment, we propose an efficient algorithm applying the Lightweight Encryption Algorithm (LEA) to the CTR (Counter) operation mode of WPA2 (WiFi Protected Access 2) to encrypt the transmitted data through the MAVLink application. And we study how to apply LEA based on CBC (Cipher Block Chaining) operation mode used in WPA2 for message security tag generation. In addition, a modified Diffie-Hellman key exchange method is approached to generate a new key used for encryption and security tag generation. The proposed method and similar methods are compared and analyzed in terms of efficiency.

Cloud storage services have many advantages. As a result, the amount of data stored in the storage of the cloud service provider is increasing rapidly. This increase in demand forces cloud storage providers to apply deduplication technology for efficient use of storages. However, deduplication technology has inherent security and privacy concerns. Several schemes have been proposed to solve these problems, but there are still some vulnerabilities to well-known attacks on deduplication techniques. In this paper, we examine some of the existing schemes and analyze their security weaknesses.

Recently, researches using LBP and SVM have been performed as one of the image - based methods for facial emotion recognition. LBP, introduced by Ojala et al., is widely used in the field of image recognition due to its high discrimination of objects, robustness to illumination change, and simple operation. In addition, CS(Center-Symmetric)-LBP was used as a modified form of LBP, which is widely used for face recognition. In this paper, we propose a method to detect four facial expressions such as expressionless, happiness, surprise, and anger using deep neural network. The validity of the proposed method is verified using accuracy. Based on the existing LBP feature parameters, it was confirmed that the method using the deep neural network is superior to the method using the Adaboost and SVM classifier.

In this paper, we propose a new smart ordering application system for granting the customer numbers. The proposed system assigns sequence number of customers who visit at offline store as using speaker and microphone of smart device. This system has more advantage than the existing ordering system. Because the proposed system does not need any customer information like as phone number or SNS ID, it can protect customer privacy information. In this system, we use high frequency which is inaudible to the human ear as communication signal between speaker and microphone. To confirm performance evaluation, we perform a test with ten smart devices like as iPhone 6, 7, 8, Galaxy s8 and the result shown a success rate of 98.7 percent. Therefore, the proposed system can be useful service technology at the various offline store which need to assign a sequence number for customers, because many customers visit at the store.

There are some limits on cost and efficiency for large amount of data in RDBMS, and NoSQL is starting to gain popularity. In medical institutions, data forms are different between organizations, and that makes difficulty for interoperability between organizations. In this paper we focused on performance issues between RDMBS and NoSQL in medical documents. We had built two different environment and had experiment comparative analysis of NoSQL with RDBMS based on medical data. We used medical HL7 FHIR as a medical data standard. Also YCSB benchmark tool was used for performance comparison. Experiments shows that NoSQL has better performance in large amounts of medical data processing systems that have over 10,000~100,000 records.

In these days, 360 degree videos, which is provided via VR, have high resolution and quality of 8K. These kinds of videos inevitably require streaming technology which guarantees QoS. Therefore, we suggest MPEG DASH HTTP protocol which stably provides streaming services of high quality videos in 8K, which are 360 degree tile-encoded, on low-spec devices such as OTT and IPTV Settop and Tile Segment Cache management structure for servers operating streaming services.

Clustering is one of the most fundamental algorithms in machine learning. The performance of clustering is affected by the distribution of data, and when there are more data or more dimensions, the performance is degraded. For this reason, we use a stacked auto encoder, one of the deep learning algorithms, to reduce the dimension of data which generate a feature vector that best represents the input data. We use k-means, which is a famous algorithm, as a clustering. Sine the feature vector which reduced dimensions are also multi dimensional, we use the Euclidean distance as well as the cosine similarity to increase the performance which calculating the similarity between the center of the cluster and the data as a vector. A deep clustering networks combining a stacked auto encoder and k-means re-trains the networks when the k-means result changes. When re-training the networks, the loss function of the stacked auto encoder and the loss function of the k-means are combined to improve the performance and the stability of the network. Experiments of benchmark image ad document dataset empirically validated the power of the proposed algorithm.

The mobile visual search service uses a query image to acquire linkage information through pre-constructed DB search. From the standpoint of this purpose, it would be more useful if you could perform a search on a web-based keyword search system instead of a pre-built DB search. In this paper, we propose a representative query extraction algorithm to be used as a keyword on a web-based search system. To do this, we use image classification labels generated by the CNN (Convolutional Neural Network) algorithm based on Deep Learning, which has a remarkable performance in image recognition. In the query extraction algorithm, dictionary meaningful words are extracted using Wikipedia, and hierarchical categories are constructed using WordNet. The performance of the proposed algorithm is evaluated by measuring the system response time.

Recent advances in machine learning approaches such as deep neural network and reinforcement learning offer significant performance improvements in generating detailed and varied motions in physically simulated virtual environments. The optimization methods are highly attractive because it allows for less understanding of underlying physics or mechanisms even for high-dimensional subtle control problems. In this paper, we propose an efficient learning method for stochastic policy represented as deep neural networks so that agent can generate various energetic motions adaptively to the changes of tasks and states without losing interactivity and robustness. This strategy could be realized by our novel trajectory search method motivated by the trust region policy optimization method. Our value-based trajectory smoothing technique finds stably learnable trajectories without consulting neural network responses directly. This policy is set as a trust region of the artificial neural network, so that it can learn the desired motion quickly.

In recent years, many wearable devices and mobile apps related to life care have been developed, and a service for measuring the movement during walking and showing the amount of exercise has been provided. However, they do not measure walking in detail, so there may be errors in the total calorie consumption. If the user's behavior is measured by a multi-axis sensor and learned by a machine learning algorithm to recognize the kind of behavior, the detailed operation of walking can be autonomously distinguished and the total calorie consumption can be calculated more than the conventional method. In order to verify this, we measured activities and created a model using a machine learning algorithm. As a result of the comparison experiment, it was confirmed that the average accuracy was 12.5% or more higher than that of the conventional method. Also, in the measurement of the momentum, the calorie consumption accuracy is more than 49.53% than that of the conventional method. If the activity recognition is performed using the wearable device and the machine learning algorithm, the accuracy can be improved and the energy consumption calculation accuracy can be improved.

In the context of the 4th Industrial Revolution, IoT, Big Data, and VR are rapidly emerging as core sectors of future industries. In particular, the VR has been under the limelight as a new media content appealing to new generation. And the VR user is not merely a 'spectator', but the 'actor'. In other words, the newness of VR is not in the 'more likely representation of the virtual reality', but in the 'making it act more virtual (more technically, 'interactive')' in the virtual world. In this paper, we examine the VR cinema in terms of film composition unit, montage, cinematic space, mise-en-$sc{\grave{e}}ne$ and perspective. The VR cinema, which is in the early stage of evolution, is basically based on $360^{\circ}$ image that strengthens the autonomy of the audience's point of view, but other factors like haptic or sonic immersion are becoming increasingly important. In addition, the VR cinema will be combined with AR, MR, SR, and Interactive technologies, and will expand its horizon as it is produced in various forms. Therefore, it is expected that more detailed viewpoint will be applied in the subsequent study on VR cinema.

Virtual reality (VR) technologies, which have been improved of late, offer experiences closer to reality than before. While VR technologies are expected to be an effective medium if used in learning content that cannot be offered in real life, distraction factors in VR technologies may prevent the users' presence and flow, making these technologies ineffective. Therefore, this study, through experiment and observation, analyzed the effect of distraction factors on the user's experience perception and learning effect when using HMD-based immersive VR. Experimental results revealed that the distraction factors were shown to have a negative, albeit not generally significant (except for tactile interactivity), effect on vividness, tactile interactivity, locomotive interactivity, presence, and flow. Ultimately, they were shown to have a negative effect on learning.

3D SCAN enables easy human body measurement via a digital method in the process of film costume production which used to be done manually. Software-oriented computer graphic, which integrates 3D SCAN data in the process of manual film costume production, can induce quick and diverse design outcomes. While, 3D PRINT, which integrates computer graphic data in the process of manual film costume production, can automate the process of special costume production using a digital method. Integration of 3D Scan + Computer Graphic + 3D Print using integrated platforms for tailored costume production as developed in this study allows significant reduction of costume production period and costs. It also allows efficient integration of costume production outcomes in various industries related with OSMU contents in particular. In other words, using it, we can create a new business market that integrates multiple areas of film content, drama content and game content.