Share Email Print
cover

Proceedings Paper

Implementation of cost-effective diffuse light source mechanism to reduce specular reflection and halo effects for resistor-image processing
Author(s): Yung-Sheng Chen; Jeng-Yau Wang
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Light source plays a significant role to acquire a qualified image from objects for facilitating the image processing and pattern recognition. For objects possessing specular surface, the phenomena of reflection and halo appearing in the acquired image will increase the difficulty of information processing. Such a situation may be improved by the assistance of valuable diffuse light source. Consider reading resistor via computer vision, due to the resistor’s specular reflective surface it will face with a severe non-uniform luminous intensity on image yielding a higher error rate in recognition without a well-controlled light source. A measurement system including mainly a digital microscope embedded in a replaceable diffuse cover, a ring-type LED embedded onto a small pad carrying a resistor for evaluation, and Arduino microcontrollers connected with PC, is presented in this paper. Several replaceable cost-effective diffuse covers made by paper bowl, cup and box inside pasted with white paper are presented for reducing specular reflection and halo effects and compared with a commercial diffuse some. The ring-type LED can be flexibly configured to be a full or partial lighting based on the application. For each self-made diffuse cover, a set of resistors with 4 or 5 color bands are captured via digital microscope for experiments. The signal-to-noise ratio from the segmented resistor-image is used for performance evaluation. The detected principal axis of resistor body is used for the partial LED configuration to further improve the lighting condition. Experimental results confirm that the proposed mechanism can not only evaluate the cost-effective diffuse light source but also be extended as an automatic recognition system for resistor reading.

Paper Details

Date Published: 22 September 2015
PDF: 14 pages
Proc. SPIE 9599, Applications of Digital Image Processing XXXVIII, 959928 (22 September 2015); doi: 10.1117/12.2187895
Show Author Affiliations
Yung-Sheng Chen, Yuan Ze Univ. (Taiwan)
Jeng-Yau Wang, Yuan Ze Univ. (Taiwan)
National Chung Shan Institute of Science and Technology (Taiwan)


Published in SPIE Proceedings Vol. 9599:
Applications of Digital Image Processing XXXVIII
Andrew G. Tescher, Editor(s)

© SPIE. Terms of Use
Back to Top