2002 Fiscal Year Final Research Report Summary
Object Recognition Using a Multichannel Vision System
| Project/Area Number |
13680468
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Intelligent informatics
|
|---|
| Research Institution | Osaka Electro-Communication University |
|---|
Principal Investigator |
TOMINAGA Shoji Osaka Electro-Communication Univ., Dept. of Engineering Informatics, Professor, 総合情報学部, 教授 (10103342)
|
|---|
| Project Period (FY) |
2001 – 2002
|
| Keywords | vision system / multichannel / liquid-crystal tunable filter / spectrum / reflection model / color / object recognition |
|---|
| Research Abstract |
1. Analysis of surface reflection models
First, two types of dichromatic reflection models are proposed for representing the surface-spectral reflectances of a variety of hard copy objects printed by four printers of thermal dye-sublimation, ink jet, toner, and silver halide. Next, we present an experimental apparatus and set of calculations for determining various parameters of the surface reflection model. The apparatus consists of a lighting system to emit parallel beams, a goniometric rotating table, and a vision system with six spectral channels. The camera data are used to estimate various parameters such as spectral reflectance, illuminant spectrum, surface roughness, ratio of body to interface intensity, and index of refraction.
2. Object recognition using a multichannel vision system
First, we propose a multichannel imaging system and algorithms for identifying objects in a natural scene based on surface-spectral reflectances. The imaging system is composed of a liquid-crystal tu
… More
nable filter, a monochrome CCD camera, and a personal computer. Algorithms are presented for estimating both spectral functions of the illuminant spectral-power distribution and surface-spectral reflectance from the spectral image data. Next, we propose a method for identifying object materials on a raw circuit board. We have developed a multichannel camera system for observing tiny objects. We can distinguish metals and dielectrics based on the difference of reflectance in changing illumination geometries.
3. Recording and rendering of art paintings by using a multichannel vision system
We propose a method for recording and rendering of art paintings using only spectral reflectance data of the object surfaces. A multichannel camera system with six spectral channels of fixed wavelength bands is used for spectral imaging. First, a method for estimating spectral reflectance for body reflection component of a rough surface is described. Next a method is proposed for practical image rendering. As a result, the image of an art painting object illuminated from any direction is rendered using the reflectance data obtained for three illumination directions. Less
|
