Shigeru Ando

Consistent gradient operators

We propose optimal gradient operators based on a newly derived consistency criterion. This criterion is based on an orthogonal decomposition of the difference between a continuous gradient and discrete gradients into the intrinsic smoothing effect and the self-inconsistency involved in the operator. We show that consistency assures the exactness of gradient direction of a locally 1D pattern in spite of its orientation, spectral composition, and sub-pixel translation. Stressing that inconsistency reduction is of primary importance, we derive an iterative algorithm which leads to accurate gradient operators of arbitrary size. We compute the optimum 3/spl times/3, 4/spl times/4, and 5/spl times/5 operators, compare them with conventional operators and examine the performance for one synthetic and several real images. The results indicate that the proposed operators are superior with respect to accuracy, bandwidth and isotropy.

Correlation image sensor: two-dimensional matched detection of amplitude-modulated light

In this paper, we describe a novel imaging device, the correlation image sensor, whose pixel circuit consists of a photodetector and source-coupled MOS transistors with individual capacitive loads and readout switches. It accepts differential or three-phase reference signals to produce one or two correlation integrals between the incident light intensity and the reference signals. The correlation results stored as charges in the capacitors are read out via MOS scanning circuits. We describe a functional model of the circuit and a quadrature detection method of amplitude-modulated light. We fabricated 64/spl times/64 and 100/spl times/100 pixel devices and obtained a correlation signal-to-noise ratio of 34 dB and a phase accuracy of 0.85/spl deg/ using only two-frame integration. We also describe application methods of this sensor to cross-correlation imagers, amplitude modulation-phase modulation demodulation imagers, and correlation matching imagers.

Image field categorization and edge/corner detection from gradient covariance

Edges, corners, and vertices in an image correspond to 1D (one-dimensional) and 2D discontinuities in the intensity surface of the underlying scene. Ridges and peaks correspond to 1D and 2D extrema in it. All of them can be characterized by the distribution of gradients, particularly by the dimensionality of it. The approach to image field categorization here is to construct a covariance matrix of the gradient vector in each small window and apply the canonical correlation analysis to it. Schwarzs inequality on the matrix determinant and the related differential equation is the key to this analysis. We obtain two operators P/sub EG/ and Q/sub EG/ to categorize the image field into a unidirectionally varying region (UNIVAR), an omidirectionally varying region (OMNIVAR), and a nonvarying region. We investigate the conditions under which their absolute maximum response, i.e. P/sub EG/=1 and Q/sub EG/=1, occurs in the small window and show that they are, respectively, the desired 1D and 2D discontinuities/extrema and OMNIVAR, is in many cases, a 1D pattern in polar coordinates. This leads to an algorithm to obtain further classification and accurate localization of them into edges, ridges, peaks, corners, and vertices through detailed analysis in the informative (varying) axis of them. We examined and compared the performance of the operators and the localization algorithm on various types of images and various noise levels. The results indicate that the proposed method is superior with respect to stability, localization, and resolution.

Automatic visual searching and reading of barcodes in 3-D scene

Automatic finding and reading barcodes in 3D scene has wide demands. The key problem is to search barcodes and supply them to a reading subsystem. In the previous papers (see IEEE Trans. PAMI, vol.22, no.2, p.179-90, 2000, and vol.22, no.3, p.252-65, 2000), we proposed a method of feature extraction after categorization and projection for edges, ridges, corners, and vertices. The categorization is based on uni- and omnidirectionality of significant local variation. In this paper, we describe an extension of this method that is more efficient to detailed images with closely spaced features like barcodes. Defining a barcode region with these features, we describe an application of it to extraction and reading of barcodes in 3-D scene.

A computer vision system for knowledge-based 3D scene analysis using radio-frequency tags

Visual tags, of which barcodes represent a successful effort for standardization, have often been used to deepen the level of automation in handling industrial objects. In the past few years, a new type of tags has been the object of intensive developments: radio-frequency (RF) Tags. We present the architecture of a machine vision system that uses RF tag identification (RFID). RFID enables detecting the presence of an object, which polyhedral model and properties are remotely retrieved from a knowledge network database. This model is used for 3D scene analysis by projective geometry. Our system not only facilitates the problem of object identification and registration to a model, but also allows new tag-based applications to be built around the concept of ubiquitous objects.

Tag-based vision: assisting 3D scene analysis with radio-frequency tags

Radio frequency (RF) tags are growing popular for many industrial applications such as product tracking and security. We investigate the benefit of RF tagging in an application of 3D object registration from single view. We present a machine vision system that uses a RF Tag device to detect objects prior to locating them visually. The tag system consists of a tag reader that can interrogate, and receive radio signals from, tags attached on objects and characterizing them. Laying the basis of an object model database shared on a network, we perform a knowledge-based recognition task where the information retrieved from the database query serves as the prior. The recognition algorithm used is a matching with projective invariants. Our architecture is a prototype that prefigures tag based applications where physical and logic representations of objects are combined.

Novel theoretical design and fabrication test of biomimicry directional microphone

Micromachined directional microphones (sound source localization sensor) mimicking an auditory organ of Ormia Ochracea have been intensively studied based on center-supported, marginally free 1-D or 2-D diaphragm structures. Based on an exact mathematical formulation, a Gaussian weight function on the diaphragm is shown to be inevitable in the transduction of 1st and 2nd moments. An analysis was performed to obtain the weight distribution assuming an arbitrary thickness distribution of a shell-structured diaphragm. The best result was a marginally-supported face-to-face cantilever structure. We fabricated this structure using SOI wafer and RIE. The experimental results show a strong dependence of the vibration pattern on both zenith and azimuth angles, and a greatly enhanced sensitivity to inclined sound pressure.

Optical vortex and correlation image sensor for networked deformation sensing of infrastructures

This paper outlines a new research project (CREST) for structure health monitoring. The purpose is to develop a networked sensor for static, full-axis deformation of large and long structures such as bridges, roads, and railways. We propose a novel technique based on the combined use of an optical vortex (OV) beam and correlation image sensors (CIS). Since the beam profile of OV involves definitive features known as zeros or phase singularities and they are invariant through propagation, we can encode and transmit information to determine geometrical relations between the sender and receiver nodes. By demodulating an amplitude and phase profile of OV using the CIS, all (three lateral and three rotational) displacement components can be determined. The basic optical setup is presented and the results of the fundamental experiments are shown.

Optical flow determination with complex-sinusoidally modulated imaging

In this paper, we study optical flow determination using a novel imaging device, correlation image sensor (CIS). The proposed method is based on the optical flow equation between intensity image and complex sinusoidally-modulated image captured simultaneously by CIS. Since it does not involve time derivative, only one frame is required for estimation. Moreover, the velocity limitation by the approximate frame differential is avoided

Real-time tracking experiment of higher-order Laguerre-Gaussian beam for remote six-axis deformation sensing

In INSS2008, we described a new principle for monitoring full-axis (three translational and three rotational) deformation of large and long structures such as bridges, roads, and railways. It is based on a combined use of an optical Laguerre-Gaussian (LG) beam and correlation image sensor (CIS). In this paper, we describe theoretically and examine experimentally several key issues of this technology. They are: 1) an optimum design of higher-order LG beam, 2) theoretical analysis on the relation between the beam profile and the deformation parameters, and 3) real-time tracking and parameter extraction algorithms of the LG beam and its parameters.