We propose a rendering method that represents contour drawing techniques in sumi-e. Our method focuses on both “mokkotsu” and “horinuri” techniques in sumi-e. We developed a rendering system which can generate sumi-e-like images based on these techniques.

抄録全体を表示

This paper statistically models 3-D DFT coefficients of a moving image with our previously presented 2-component spherically-symmetric Gaussian mixture distribution, defines dissimilarity between moving images on the basis of geometric metrics in the model space, constructs hierarchical clustering with the dissimilarity, and applies the clustering to ITE standard moving images.

抄録全体を表示

Three-dimensional mean value separation type short time DFT is effective for movie restoration. We modeled the ST-DFT coefficient as two component spherically-symmetric Gaussian mixture distribution. We proposed the estimation approach of the model parameters using the moment method. If we know the amount of noise mixing in the movie, parameter estimation procedure can executed without iteration. In this purpose, we propose the estimation method of noise variance value in moving-image by using three dimension DFT coefficients.

抄録全体を表示

We present an edge-aware filter using multi-scale box filter (MSBF). In our method, multiple filtered results are adaptively weighted, which depend on image edge and gradient, and summed. The proposed filter achieves better result in 1)image enhancement and 2)image denoising application compared with guided filter, i.e. 1)avoiding luminance saturation, 2)preserving edges.

抄録全体を表示

Mountain tunnel construction is different greatly from other infrastructure construction, as the construction method of the tunnel is depend on the rock condition. However, the condition is evaluated by the subjective indexes. We propose a new objective evaluation method of rock condition displayed by the distribution of regional directional component for a tunnel face images. We confirmed the usefulness of the proposed system for some real tunnel face images.

抄録全体を表示

UHD-TV display is measured to faithful reproduction of color on UHD-TV display. When measuring, colorimetric measurements were made by displaying RGB monochrome and color mixture on the display. Mutual influence of RGB signals was investigated on the base of measurement results.

抄録全体を表示

ZnInGaP/ZnS quantum dots (QDs), which were low-toxic and had excellent PL characteristics, were refined and evaluated by the thermo-gravimetric analysis in order to apply them to the QD light emitting diodes (LEDs). The uniformity of films and the luminescence property were improved in the QD-LEDs by using the refined quantum dots.

抄録全体を表示

We are studying a W spin domain wall type optical modulation device in order to realize a wide viewing zone dynamic holographic display device. In order to simplify a manufacturing process, we proposed a technique which controls a coercivity. Moreover, we manufactured the device experimentally and demonstrated an optical modulation.

抄録全体を表示

We present the layout design of an optical phased array (OPA) using electro-optic (EO) polymers. The designed OPA has cascaded 1x2 multimode interference beam splitters, and was optimized in the propagation loss. We fabricated and evaluated the designed OPA, and then found a sharply defined optical beam and a 22.5 degrees deflection on far-field patterns.

抄録全体を表示

We report on development of prototype hybrid cast application using 4k live streaming and companion screen. A recoded 4k video of Koshien games in summer 2017 was encoded in real time at adaptive bit rate (three different rate), packaged in MPEG-DASH and uploaded to the cloud. The 4k video was delivered to a television receiver and played using the IPTV-F dashNX player. In addition, we also developed an application which controls playback of 4k live video on TV terminal from smartphone by using companion screen architecture of Hybridcast and confirmed that trick play control by it was operated stably.

抄録全体を表示

In the case of live distribution of 4K video related to terrestrial broadcasting using a hybrid casting system, HEVC system compressible encoder, file division machine (MPEG-DASH conversion),Distribution server is required separately. Therefore, preparation required expertise, time and effort. We cooperated with manufacturers to integrate these functions and developed small and lightweight products.

抄録全体を表示

We are developing “Hybridcast Connect Library”. In addition to the existing TV-centric control, the mobile-centric control is available to work with this library. New services can be expected by various applications and IoT devices built on this library.

抄録全体を表示

TBS Television has developed a Hybridcast application which leads consumers to a real-time TV viewing.

抄録全体を表示

In the projection mapping, it is necessary to deform the image according to the surface shape of the three-dimensional object which is a screen. In this research, we developed a method to detect corrected coordinates of all pixels on a simple system using Web camera and projector scanning line. In this paper, we describe this correction method and an example of real time image deformation realized by using the obtained correction coordinates and FPGA.

抄録全体を表示

We propose a system that can create and share virtual objects in real time by using 3D position of monocular camera. Using this system, we can easily embody our intention with a virtual object .

抄録全体を表示

We are studying about the hybrid sensor which detects a movement of a camera to composite CG to the real image shot by the camera. This sensor is constructed with MEMS, Laser Optical gyro and a wire mechanism. In the basic experiment, we recognized the advantage comparing traditional sensor.

抄録全体を表示

We are studying about the hybrid sensor which detects a movement of a camera to composite CG to the real image shot by the camera. This sensor is constructed with MEMS, Laser Optical gyro and a wire mechanism. In the basic experiment, we recognized the advantage comparing traditional sensor.

抄録全体を表示

A video search method when keywords are not clear is considered. A few cloud services that attach semantic metadata to video clips are evaluated. Experiment using such services suggests effectiveness of semantic metadata to identify video clips whether to want to watch or not at the time of relaxing.

抄録全体を表示

Conventional separable histogram equalization method often produces unnatural colors. In this paper, we propose a method for equalizing a common histogram for three channels: red, green and blue. Experimental results show that the proposed method produces more natural colors than the ones produced by the conventional separable method.

抄録全体を表示

We proposed a method of two-in-one image steganography using error diffusion for grayscale images at the 70th JCEEE in Kyusyu last year [1]．In this paper, we improve the method to make it also available for color images．

抄録全体を表示

This paper proposes a frequency-domain realization of the depth from defocus(DFD) technique based on cross-reblurring, which is based on the property that two blurred images coincide if the others point spread functions is applied each other. The frequency domain approach can reduce the computation cost, which are demonstrated by the experimental results given later in the paper.

抄録全体を表示

We examined how much the immersive feeling and sense of presence given by the user changes as the sound is zoomed in accordance with the zoom of the image in the VR system. For verification, we have developed VR system that can record omnidirectional images and sound. Our system cut out sound and images in the target direction in real time. Experimental results show that the user can obtain more immersive and sense of presence by zooming the sound according to the zoom of the images.

抄録全体を表示

We are conducting research on advanced terrestrial broadcasting system under the auspices of the Ministry of Internal Affairs and Communications. Because of applying LDPC code and high-order-modulation OFDM, spectral efficiency is improved. We have developed the prototype modulator and demodulator and evaluated its performance in AWGN environment by laboratory experiment.

抄録全体を表示

We developed prototype modulator and demodulator based on “Next Generation Terrestrial Broadcasting System” and evaluated the transmission characteristics in SFN environment by laboratory experiments. As the results of the experiments, the required C/N degradation of “Next Generation Terrestrial Broadcasting System” became smaller than that of ISDB-T.

抄録全体を表示

LDPC (Low Density Parity Check) codes are used for the next generation of the digital terrestrial broadcasting. In general, the BER (Bit Error Rate ) performance of LDPC codes have a waterfall region and an error floor It is difficult to evaluate the BER performance in the error floor region because of the huge amount of the computation. In order to evaluate it , we conducted the laboratory experiment using the prototype modulator and demodulator in which LDPC codes are implemented. This paper reports that the BER performance in the error floor region is under BER= 1.0×10^-7.

抄録全体を表示

Toward the advancement of broadcasting services t he authors are conducting research on the next generation terrestrial broadcasting. The next generation terrestrial broadcasting system defined by proposed specification is able to provide mobile reception service by partial reception such as one segment service of ISDB-T. Moreover in order to improve the frequency inter leaving effect the partial reception bandwidth is expanded approximately 3.5 times compared to ISDB-T. This paper describes the frequency interleave method suitable for the expanded partial reception and the evaluation results of the proposed method by laboratory experiments.

抄録全体を表示

NHK has been researching multiplexing scheme for IP packets in next-generation terrestrial broadcasting system. In the actual SFN environment, we conducted fundamental experiments for SFN using remux equipment and modulators that support the multiplexing scheme.

抄録全体を表示

Technical verification for automatically determining telop mistakes in TV programs.

抄録全体を表示

In this paper, we tackle to identify near-miss traffic incidents, which is close to traffic accidents, on vehicle camera video recorded by event data recorder. We study three types of identification method: typical video feature descriptor, feature extraction extraction based on DNN composed from CNN and RNN, and combination them. From the results of quantitative evaluation using actual event recorder data, we clarify the effectiveness with each video feature extraction technique.

抄録全体を表示

This paper proposes a search results re-ranking method for illegal video copy detection system. We estimate illegal probability scores using illegal video profile data and re-rank search results by estimated order. The experimental results indicate that the proposed method outperformed the order of search results.

抄録全体を表示

The purpose of this research is to predict water levels of rivers without building detailed geo logical models. Sparse prediction was applied to rainfall sensors, water level sensors and gridded rainfall data. Gridded rainfall brings better accuracy on water level prediction. Moreover, 2D Fused Lasso shows better results to normal Lasso.

抄録全体を表示

In the next-generation DTTB, a combination of polarized MIMO and ultra-multilevel OFDM is proposed to improve transmission capacity. High accuracy channel estimation is required for detection. In this paper, the reception characteristics using the proposed channel estimation scheme of amplitude, delay time and polarization angle are evaluated.

抄録全体を表示

In the next-generation DTTB, a combination of polarized MIMO and ultra-multilevel OFDM is proposed to improve transmission capacity. High accuracy channel estimation is required for detection. In this paper, the reception characteristics using the proposed channel estimation scheme of amplitude, delay time and polarization angle are evaluated.

抄録全体を表示

Dynamic broadcasting transmission band assignment scheme is proposed. In this scheme, quality of experiments (QoE) is calculated by real-time viewing rate and transmission bands are dynamically assigned based on the QoE. The simulation results show that the QoE of proposal scheme is constantly higher than that of static assignment scheme.

抄録全体を表示

Dynamic RF/IP switching system for TV program over distribution over Cable TV network is proposed. To evaluate the system, we developed emulator environments and the emulation results show that dynamic RF/IP switching contribute to improvements in quality of experience (QoE) compared to the results under no switching.

抄録全体を表示

Multicast join together scheme is proposed. In this scheme, all multi-cast streams are joined in one time to accelerate picture transition for IPTV channel zapping whereas each multicast separately is joined in general scheme. It is calculated that the latency of proposal scheme is shorter than that of general scheme.

抄録全体を表示