Novel algorithms for optimal compression using classification metrics

Bei Xie, Tamal Bose, Erzsébet Merényi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

In image processing, classification and compression are very common operations. Compression and classification algorithms are conventionally independent of each other and performed sequentially. However, some class distinctions may be lost after a minimum distortion compression. In this paper, two new schemes are developed that combine the compression and classification operations in order to optimize some classification metrics. In other words, the compression systems are improved under classification constraints. In the first scheme, compression is achieved by using Adaptive Differential Pulse Code Modulation (ADPCM). Optimization of filter coefficients is done by using a simple genetic algorithm (GA). In the second scheme, compression is achieved by image transform and quantization. The parameters in transform and quantization are adapted to improve the compression system and reduce the classification errors. Computer simulations are performed on hyperspectral images. The results are promising and illustrate the performance of the algorithms under various classification constraints and compression schemes.

Original languageEnglish (US)
Title of host publicationIEEE Aerospace Conference Proceedings
DOIs
StatePublished - 2008
Externally publishedYes
Event2008 IEEE Aerospace Conference, AC - Big Sky, MT, United States
Duration: Mar 1 2008Mar 8 2008

Other

Other2008 IEEE Aerospace Conference, AC
CountryUnited States
CityBig Sky, MT
Period3/1/083/8/08

Fingerprint

Differential pulse code modulation
Image processing
Genetic algorithms
Computer simulation

ASJC Scopus subject areas

  • Aerospace Engineering

Cite this

Xie, B., Bose, T., & Merényi, E. (2008). Novel algorithms for optimal compression using classification metrics. In IEEE Aerospace Conference Proceedings [4526393] https://doi.org/10.1109/AERO.2008.4526393

Novel algorithms for optimal compression using classification metrics. / Xie, Bei; Bose, Tamal; Merényi, Erzsébet.

IEEE Aerospace Conference Proceedings. 2008. 4526393.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Xie, B, Bose, T & Merényi, E 2008, Novel algorithms for optimal compression using classification metrics. in IEEE Aerospace Conference Proceedings., 4526393, 2008 IEEE Aerospace Conference, AC, Big Sky, MT, United States, 3/1/08. https://doi.org/10.1109/AERO.2008.4526393
Xie, Bei ; Bose, Tamal ; Merényi, Erzsébet. / Novel algorithms for optimal compression using classification metrics. IEEE Aerospace Conference Proceedings. 2008.
@inproceedings{4f1583f1b12742f3aad068256993189e,
title = "Novel algorithms for optimal compression using classification metrics",
abstract = "In image processing, classification and compression are very common operations. Compression and classification algorithms are conventionally independent of each other and performed sequentially. However, some class distinctions may be lost after a minimum distortion compression. In this paper, two new schemes are developed that combine the compression and classification operations in order to optimize some classification metrics. In other words, the compression systems are improved under classification constraints. In the first scheme, compression is achieved by using Adaptive Differential Pulse Code Modulation (ADPCM). Optimization of filter coefficients is done by using a simple genetic algorithm (GA). In the second scheme, compression is achieved by image transform and quantization. The parameters in transform and quantization are adapted to improve the compression system and reduce the classification errors. Computer simulations are performed on hyperspectral images. The results are promising and illustrate the performance of the algorithms under various classification constraints and compression schemes.",
author = "Bei Xie and Tamal Bose and Erzs{\'e}bet Mer{\'e}nyi",
year = "2008",
doi = "10.1109/AERO.2008.4526393",
language = "English (US)",
isbn = "1424414881",
booktitle = "IEEE Aerospace Conference Proceedings",

}

TY - GEN

T1 - Novel algorithms for optimal compression using classification metrics

AU - Xie, Bei

AU - Bose, Tamal

AU - Merényi, Erzsébet

PY - 2008

Y1 - 2008

N2 - In image processing, classification and compression are very common operations. Compression and classification algorithms are conventionally independent of each other and performed sequentially. However, some class distinctions may be lost after a minimum distortion compression. In this paper, two new schemes are developed that combine the compression and classification operations in order to optimize some classification metrics. In other words, the compression systems are improved under classification constraints. In the first scheme, compression is achieved by using Adaptive Differential Pulse Code Modulation (ADPCM). Optimization of filter coefficients is done by using a simple genetic algorithm (GA). In the second scheme, compression is achieved by image transform and quantization. The parameters in transform and quantization are adapted to improve the compression system and reduce the classification errors. Computer simulations are performed on hyperspectral images. The results are promising and illustrate the performance of the algorithms under various classification constraints and compression schemes.

AB - In image processing, classification and compression are very common operations. Compression and classification algorithms are conventionally independent of each other and performed sequentially. However, some class distinctions may be lost after a minimum distortion compression. In this paper, two new schemes are developed that combine the compression and classification operations in order to optimize some classification metrics. In other words, the compression systems are improved under classification constraints. In the first scheme, compression is achieved by using Adaptive Differential Pulse Code Modulation (ADPCM). Optimization of filter coefficients is done by using a simple genetic algorithm (GA). In the second scheme, compression is achieved by image transform and quantization. The parameters in transform and quantization are adapted to improve the compression system and reduce the classification errors. Computer simulations are performed on hyperspectral images. The results are promising and illustrate the performance of the algorithms under various classification constraints and compression schemes.

UR - http://www.scopus.com/inward/record.url?scp=49349111167&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=49349111167&partnerID=8YFLogxK

U2 - 10.1109/AERO.2008.4526393

DO - 10.1109/AERO.2008.4526393

M3 - Conference contribution

SN - 1424414881

SN - 9781424414888

BT - IEEE Aerospace Conference Proceedings

ER -