Object recognition by cost minimization

Liu Lu; Fang Luo; Nanno J. Mulder

doi:10.1117/12.182904

17 August 1994 Object recognition by cost minimization

Liu Lu, Fang Luo, Nanno J. Mulder

Proceedings Volume 2357, ISPRS Commission III Symposium: Spatial Information from Digital Photogrammetry and Computer Vision; (1994) https://doi.org/10.1117/12.182904
Event: Spatial Information from Digital Photogrammetry and Computer Vision: ISPRS Commission III Symposium, 1994, Munich, Germany

Abstract

Since the analogy between images and statistical mechanics systems was made, numerous research projects have been done to solve problems of image processing by the use of iterative methods. In this paper, we investigate cost functions in image processing and present a general expression and common requirements for cost functions. Meanwhile, we construct cost functions for object recognition. Although the cost functions are defined with geometric errors or radiometric errors, they measure errors in the two domains and their minimum corresponds to the same prediction without any errors in both domains. An optimization procedure is performed to reach the expected result with the minimum cost. The most noticeable characteristic of the method is that it is not sensitive to noise and works remarkably well in the presence of high noise level.

Citation Download Citation

Liu Lu, Fang Luo, and Nanno J. Mulder "Object recognition by cost minimization", Proc. SPIE 2357, ISPRS Commission III Symposium: Spatial Information from Digital Photogrammetry and Computer Vision, (17 August 1994); https://doi.org/10.1117/12.182904

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available