Hideaki Kawano

Sharpness preserving image enlargement by using self-decomposed codebook and Mahalanobis distance

We propose an image enlargement method preserving perceptual sharpness, which is achieved by augmenting a low resolution image with high-frequency components from a given image itself. The estimation of high-frequency image components is performed by a codebook built by a decomposition of the given image, i.e. a self-decomposed codebook. The rational that is exploited in this approach is the shape-invariant properties of edges across scales. As a distance measure for matching from the codebook, we employ the Mahalanobis distance which is a local distance measure incorporating pixel correlation. The effectiveness of the proposed method is verified by some image enlargement experiments. Consequently, the experimental results show that the performance of the proposed method is superior to other conventional image enlargement methods objectively and subjectively.

Analysis of Similarity and Differences in Brain Activities Between Perception and Production of Facial Expressions Using EEG Data and the NeuCube Spiking Neural Network Architecture

This paper is a feasibility study of using the NeuCube spiking neural network (SNN) architecture for modeling EEG brain data related to perceiving versus mimicking facial expressions. It is demonstrated that the proposed model can be used to study the similarity and differences between corresponding brain activities as complex spatio-temporal patterns. Two SNN models are created for each of the 7 basic emotions for a group of Japanese subjects, one when subjects are perceiving an emotional face and another, when the same subjects are mimicking this emotion. The evolved connectivity in the two models are then subtracted to study the differences. Analysis of the models trained on the collected EEG data shows greatest similarity in sadness, and least similarity in happiness and fear, where differences in the T6 EEG channel area were observed. The study, being based on the well-known mirror neuron concept in the brain, is the first to analyze and visualize similarity and differences as evolved spatio-temporal patterns in a brain-like SNN model.

Super-Resolution via Matching from Self-Decomposed Codebook with Local Distance Measure Incorporating Pixel Correlation

We propose a method for constructing an image of high resolution by augmenting a low resolution image with high frequency components from a given image itself. The estimation of high frequency image components is performed by a codebook built by a decomposition of the given image, i.e. self-decomposed codebook. The rational that is exploited in this approach is the shape-invariant properties of edges across scale. As a distance measure for matching from the codebook, we employ a local distance measure incorporating pixel correlation. The effectiveness of the proposed method is verified by some image super-resolution experiments. Consequently, the experimental results show that the performance of the proposed method is superior to other typical image super-resolution methods objectively and subjectively.

Sound target tracking in 3D using particle filter with 4 microphones

We propose a method to detect and to track a moving sound target in 3D using particle filter with four microphones, which are allocated not being on the same plane. We develop an elaborated state space model where state represents 3D location of the sound target, and observation is 6 sound directions obtained by 6 pairs of the 4 microphones. Each observed sound direction is calculated from TDOA measurements of corresponding microphone pair by taking maximum of cross correlation function of two sound signals. Through a simulation experiment of single sound target tracking in noisy environment, we show performance of proposed method.

Specification of signboard region and extraction of characters from scene picture

In this paper, a method which specifies the signboard region and extracts the characters inside the signboard is proposed. We usually take notes not to forget what we should leave to memory. But it is often that the work is too intrusive. Our aim is the development of a new input-interface so as to input texts from a picture. Most of signboards are composed of almost monochromatic region. On the basis of the observation, image segmentation using color information is applied, and then we get some binary images by applying threshold for each segmented region. Each binary image is enclosed by the smallest circumscribed square. The signboard region is specified according to distribution and area of the white pixels inside the square. As a result of experiment, we confirmed the effectiveness of the proposed method.

Pseudo-normal Image Synthesis from Chest Radiograph Database for Lung Nodule Detection

The purpose of this study is to develop a new computer aided diagnosis (CAD) system for a plain chest radiograph. It is difficult to distinguish lung nodules from a chest radiograph. Therefore, CAD systems enhancing the lung nodules have been actively studied. The most notable achievements are temporal subtraction (TS) based systems. The TS method can suppress false alarms comparatively because it uses the chest radiograph of the same person. However, the TS method cannot be applied to initial visitors because it requires the past chest radiograph of themselves. In this study, to overcome the absence of past image for a patient himself, a pseudo-normal image is synthesized from a database containing other patient’s chest radiographs that have already been diagnosed as normal by medical specialists. And then, the lung nodules are emphasized by subtracting the synthesized normal image from the target image.

Skeletonization of decorative characters by graph spectral decomposition

Decorative characters are widely used in various scenes. Practical optical character reader is required to deal with not only common fonts but also complex designed fonts. However, since the appearances of decorated characters are complicated, most general character recognition systems cannot give good performances on decorated characters. In this paper, an algorithm that can extract characters essential structure from a decorative character is proposed. This algorithm is applied in preprocessing of character recognition. Experimental results show character skeletons are clearly extracted from very complex decorative characters.

3D Rough Reconstruction of Buildings from Streetscape by Synergetic Stereo Matching

In this paper, a method to roughly reconstruct buildings in 3D space from a streetscape is proposed. The 3D reconstruction is based on a binocular stereo method called synergetic stereo matching. The proposed method is organized into three processes: extraction of buildings regions in each streetscape image, measurements of 3D distances for feature points in the buildings, and representation of a building by a matchbox model. The effectiveness of the proposed method was verified by experiments using actual streetscape images.

A Color Quantization Based on Vector Error Diffusion and Particle Swarm Optimization Considering Human Visibility

In this paper, we propose a new color quantization method for generation of the color-reduced images. The proposed method employs a vector error diffusion VED method and a particle swarm optimization PSO. VED method based on Floyd-Steinberg dithering is used for display of the color-reduced image. Furthermore, a color palette used in VED method is optimized by PSO. PSO generates the effective color palette with evaluating a human visibility of the color-reduced image on the display. The validity and the effectiveness of the proposed method are confirmed by some experiments.

Color conversion algorithm for color blindness using self-organizing map

The symptoms of “color blindness” are due to an innate lack or deficit of cone cells that recognize colors. People with color blindness have difficulty discriminating combinations of specific colors. In this paper, we propose a novel color conversion algorithm for color blindness. In the proposed method, the difficult colors for discrimination by color blindness are detected and converted for the legibility. This color conversion is consist of a color clustering using self-organizing map (SOM) and a analysis of the map structure. To validate the effectiveness of proposed method, it is applied to a image have various color combinations.