On features ordering for rapid object detection

Shimon Cohen; Solomon Eyal Shimony; Ronen I. Brafman

doi:10.1109/EEEI.2008.4736580

On features ordering for rapid object detection

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Scopus citations

Abstract

Consider a real time classification problem which requires testing the value of a large number of features. Let T_j be the average computation time of the jth feature, and let R_j be its average rejection probability. In many real-world problems, such as defect detection on printed circuits or plasma screens, minimizing classifier execution time is of paramount importance because of its effect on the entire productions and QA process. We show that if features are independent, the best way to arrange them is according to the ratio R_j/T_j. This yields a minimal-time algorithm for features ordering in object detection framework when feature values are independent of each other, and motivates a simple algorithm for ordering features that are not independent. The algorithm obtains comparable classification results to [15], but is twice as fast.

Original language	English
Title of host publication	2008 IEEE 25th Convention of Electrical and Electronics Engineers in Israel, IEEEI 2008
Pages	509-513
Number of pages	5
DOIs	https://doi.org/10.1109/EEEI.2008.4736580
State	Published - 1 Dec 2008
Event	2008 IEEE 25th Convention of Electrical and Electronics Engineers in Israel, IEEEI 2008 - Eilat, Israel Duration: 3 Dec 2008 → 5 Dec 2008

Publication series

Name	IEEE Convention of Electrical and Electronics Engineers in Israel, Proceedings

Conference

Conference	2008 IEEE 25th Convention of Electrical and Electronics Engineers in Israel, IEEEI 2008
Country/Territory	Israel
City	Eilat
Period	3/12/08 → 5/12/08

Keywords

Feature ordering
Processing time
Rejection probability

ASJC Scopus subject areas

Electrical and Electronic Engineering

Access to Document

10.1109/EEEI.2008.4736580

Cite this

@inproceedings{3ea6e153664d4afbb6481fb241a5f8a0,

title = "On features ordering for rapid object detection",

abstract = "Consider a real time classification problem which requires testing the value of a large number of features. Let Tj be the average computation time of the jth feature, and let Rj be its average rejection probability. In many real-world problems, such as defect detection on printed circuits or plasma screens, minimizing classifier execution time is of paramount importance because of its effect on the entire productions and QA process. We show that if features are independent, the best way to arrange them is according to the ratio Rj/Tj. This yields a minimal-time algorithm for features ordering in object detection framework when feature values are independent of each other, and motivates a simple algorithm for ordering features that are not independent. The algorithm obtains comparable classification results to [15], but is twice as fast.",

keywords = "Feature ordering, Processing time, Rejection probability",

author = "Shimon Cohen and Shimony, \{Solomon Eyal\} and Brafman, \{Ronen I.\}",

year = "2008",

month = dec,

day = "1",

doi = "10.1109/EEEI.2008.4736580",

language = "English",

isbn = "9781424424825",

series = "IEEE Convention of Electrical and Electronics Engineers in Israel, Proceedings",

pages = "509--513",

booktitle = "2008 IEEE 25th Convention of Electrical and Electronics Engineers in Israel, IEEEI 2008",

note = "2008 IEEE 25th Convention of Electrical and Electronics Engineers in Israel, IEEEI 2008 ; Conference date: 03-12-2008 Through 05-12-2008",

}

Cohen, S, Shimony, SE & Brafman, RI 2008, On features ordering for rapid object detection. in 2008 IEEE 25th Convention of Electrical and Electronics Engineers in Israel, IEEEI 2008., 4736580, IEEE Convention of Electrical and Electronics Engineers in Israel, Proceedings, pp. 509-513, 2008 IEEE 25th Convention of Electrical and Electronics Engineers in Israel, IEEEI 2008, Eilat, Israel, 3/12/08. https://doi.org/10.1109/EEEI.2008.4736580

On features ordering for rapid object detection. / Cohen, Shimon; Shimony, Solomon Eyal ; Brafman, Ronen I.
2008 IEEE 25th Convention of Electrical and Electronics Engineers in Israel, IEEEI 2008. 2008. p. 509-513 4736580 (IEEE Convention of Electrical and Electronics Engineers in Israel, Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - On features ordering for rapid object detection

AU - Cohen, Shimon

AU - Shimony, Solomon Eyal

AU - Brafman, Ronen I.

PY - 2008/12/1

Y1 - 2008/12/1

N2 - Consider a real time classification problem which requires testing the value of a large number of features. Let Tj be the average computation time of the jth feature, and let Rj be its average rejection probability. In many real-world problems, such as defect detection on printed circuits or plasma screens, minimizing classifier execution time is of paramount importance because of its effect on the entire productions and QA process. We show that if features are independent, the best way to arrange them is according to the ratio Rj/Tj. This yields a minimal-time algorithm for features ordering in object detection framework when feature values are independent of each other, and motivates a simple algorithm for ordering features that are not independent. The algorithm obtains comparable classification results to [15], but is twice as fast.

AB - Consider a real time classification problem which requires testing the value of a large number of features. Let Tj be the average computation time of the jth feature, and let Rj be its average rejection probability. In many real-world problems, such as defect detection on printed circuits or plasma screens, minimizing classifier execution time is of paramount importance because of its effect on the entire productions and QA process. We show that if features are independent, the best way to arrange them is according to the ratio Rj/Tj. This yields a minimal-time algorithm for features ordering in object detection framework when feature values are independent of each other, and motivates a simple algorithm for ordering features that are not independent. The algorithm obtains comparable classification results to [15], but is twice as fast.

KW - Feature ordering

KW - Processing time

KW - Rejection probability

UR - https://www.scopus.com/pages/publications/62749144034

U2 - 10.1109/EEEI.2008.4736580

DO - 10.1109/EEEI.2008.4736580

M3 - Conference contribution

AN - SCOPUS:62749144034

SN - 9781424424825

T3 - IEEE Convention of Electrical and Electronics Engineers in Israel, Proceedings

SP - 509

EP - 513

BT - 2008 IEEE 25th Convention of Electrical and Electronics Engineers in Israel, IEEEI 2008

T2 - 2008 IEEE 25th Convention of Electrical and Electronics Engineers in Israel, IEEEI 2008

Y2 - 3 December 2008 through 5 December 2008

ER -

On features ordering for rapid object detection

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this