Takahito Kawanishi

Generation of high-resolution stereo panoramic images by omnidirectional imaging sensor using hexagonal pyramidal mirrors

We have developed a high-resolution omnidirectional stereo imaging sensor that can take images at video-rate. The sensor system takes an omnidirectional view by a component constructed of six cameras and a hexagonal pyramidal mirror and acquires stereo views by symmetrically connecting two sensor components. The paper describes a method of generating stereo panoramic images by using our sensor. First, the sensor system is calibrated; that is, twelve cameras are correctly aligned with pyramidal mirrors and the Tsais method restores the radial distortion of each camera image. Stereo panoramic images are then computed by registering the camera images captured at the same time.

A fast template matching algorithm with adaptive skipping using inner-subtemplates' distances

This paper proposes a new fast template matching algorithm that skips comparison between a template and search windows neighboring an already compared dissimilar sub-window. Comparison skipping is executed when a lower bound of distance between the template and a window exceeds a threshold. The lower bound of distance between the template and the window is determined by the triangular inequality in distances: the distance between a subtemplate and a subwindow and that between inter-subtemplates. Experimental results demonstrate that the proposed method is up to five times faster than the conventional fast exhaustive search method (sequential similarity detection algorithm), strictly guaranteeing the same accuracy.

Dynamic Active Search for quick object detection with pan-tilt-zoom camera

This paper proposes a search method for detecting known objects quickly in 3D environments with a pan-tilt-zoom camera. In our previous work, we proposed an algorithm named Active Search that greatly reduces the number of calculations required to obtain a match between a reference object and an input image using color histograms. We describe two improvements we have made to Active Search for such practical applications as robots and surveillance. First, we increased the robustness as regards the color histogram changes that result from different lighting conditions and camera angles by using multiple reference images and a pixel color vector quantization. Second, we reduced the number of camera operations (pan, tilt and zoom) by using a best-direction-first and upper bound pruning strategies. We call this camera control Dynamic Active Search. Experiments show an improvement in object detection accuracy and a 78% reduction in detection time.

Stereo one-shot six-band camera system for accurate color reproduction

Abstract. For accurate color reproduction of motion pictures or still pictures of moving objects, we developed a different one-shot six-band image-capturing and visualization system that combines multiband and stereo imaging techniques. The proposed system consists of two consumer-model digital cameras and an interference filter whose spectral transmittance is comb-shaped. Subpixel correspondence search between the stereo image pairs is conducted and image transformation whose parameters are estimated from the correspondence is used to correct the geometric relationship between the images. The Wiener estimation method is used for color reproduction. For experiments, we have constructed two six-band camera systems to evaluate the quality of resultant images. One is for capturing high-resolution images using digital single-lens reflex cameras. The other is for capturing motion pictures using digital video cameras, for which all image processing steps after image capture are implemented on graphics processing units and the frame rate of the system is 30 fps when image size is XGA. For both systems, the average color difference between the measurement data and estimation results for 24 color patches of the Macbeth ColorChecker™ is dEa*b*=1.21 at maximum.

Acceleration of similarity-based partial image retrieval using multistage vector quantization

We propose a new method for quick and accurate partial image retrieval from a huge number of images based on a predefined distance measure. The proposed method utilizes vector quantization (VQ) on multiple layers, namely color, block, and feature layers. This can greatly reduce the amount of calculation needed for partial image retrieval. Experiments indicate that the proposed method can detect partial images that are similar to queries through 1000 images within 4 seconds. This is approximately 30 times faster than the method to which multistage VQ is not applied.

Quick 3D object detection and localization by dynamic active search with multiple active cameras

This paper proposes a method for detecting known objects in 3D environments and estimating their positions with multiple pan-tilt-zoom cameras. Our search method, Dynamic Active Search, reduces the number of camera operations by predicting the existence of a target in wide angles, zooming-in a promising area, and confirming the target. Even when many reference images taken under various object views and various scales need to be searched for, Dynamic Active Search can detect the object efficiently. With multiple cameras, object detection and object localization become more efficient. Experiments show that Dynamic Active Search with four cameras is 2.5 times faster and 2 times more accurate than with a single camera.

Similarity-based partial image retrieval guaranteeing same accuracy as exhaustive matching

We propose a new framework for quick and accurate partial image retrieval from a huge number of images based on a predefined distance measure. Finding partial similarities generally requires a huge amount of storage space for indexes due to the large number of portions of images. The proposed method extracts portions from each database image at a constant spacing, while it extracts all possible portions from a query image. In this way, the proposed method can greatly reduce the size of indexes while theoretically guaranteeing the same accuracy as exhaustive matching.

A stereo one-shot multi-band camera system for accurate color reproduction

In the digital archiving for cultural heritage preservation, in the medical field, and in some industrial fields, high-fidelity reproduction of color, gloss, texture, and shape are very important. Multiband or full-spectrum imaging technology is a solution for accurate color reproduction. Although several types of multi band camera systems have been developed [Yamaguchi 2000, Tominaga 2000, Helling 2004, Hashimoto 2008], all of them are multi-shot systems and they cannot take images of moving objects. Ohsawa et al. [2004] have developed a six-band HDTV camera system. However, the system requires very expensive customized equipment. In order to make multiband technology pervasive, equipment costs must be reduced and the systems have to be able to take images of moving objects. To meet these requirements, we developed a novel multiband image capturing system that combines multiband and stereo imaging techniques. This system can acquire both spectral color information and depth information at the same time. In this paper, we focus on the generation of six-band images from a pair of stereo image.

Scene text recognition with high performance CNN classifier and efficient word inference

The recognition of text in natural scene images is a practical yet challenging task due to the large variations in backgrounds, textures, fonts, and illumination conditions. In this paper, we propose a highly accurate character recognition model by utilizing the representational power of a specially designed Convolutional Neural Network (CNN). Based on the recognition model, we also develop an efficient post processing approach for error correction and hypothesis re-verification. Character and word image recognition experiments on two public datasets, namely the ICDAR 2003 Robust Reading dataset and the Street View Text (SVT) dataset both show that the proposed approach provides superior or comparable results to the state-of-the-art techniques.

High-definition and multispectral capturing for digital archiving of large 3D woven cultural artifacts

This paper describes a 3D measurement system with wheel-rail, a capturing system with multi-band camera, and a 3D modeling of large woven cultural artifacts, and show a high-resolution 3D model with multi-band image.