2014
DOI: 10.1109/tnnls.2013.2293418
|View full text |Cite
|
Sign up to set email alerts
|

Feature Learning for Image Classification Via Multiobjective Genetic Programming

Abstract: Feature extraction is the first and most critical step in image classification. Most existing image classification methods use hand-crafted features, which are not adaptive for different image domains. In this paper, we develop an evolutionary learning methodology to automatically generate domain-adaptive global feature descriptors for image classification using multiobjective genetic programming (MOGP). In our architecture, a set of primitive 2-D operators are randomly combined to construct feature descriptor… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

1
226
0
2

Year Published

2018
2018
2021
2021

Publication Types

Select...
5
1

Relationship

0
6

Authors

Journals

citations
Cited by 288 publications
(229 citation statements)
references
References 41 publications
1
226
0
2
Order By: Relevance
“…It is significance noting that the depth of scene d is the most important information. Since the scattering coefficient β can be regarded as a constant in homogeneous atmosphere condition [23], the average transmission t can be approximated easily according to Equation (2) if the depth of the scene is known. Moreover, in the ideal case, the range of d(x) is[0, +∞) as the scenery objects that show in the image can be very far away from the spectator, and we have:…”
Section: Atmospheric Scattering Modelmentioning
confidence: 99%
See 4 more Smart Citations
“…It is significance noting that the depth of scene d is the most important information. Since the scattering coefficient β can be regarded as a constant in homogeneous atmosphere condition [23], the average transmission t can be approximated easily according to Equation (2) if the depth of the scene is known. Moreover, in the ideal case, the range of d(x) is[0, +∞) as the scenery objects that show in the image can be very far away from the spectator, and we have:…”
Section: Atmospheric Scattering Modelmentioning
confidence: 99%
“…Note that, if d(x) is large enough, t (x) tends to be very small according to Equation (2), and I(x) equals A approximately. Therefore, instead of calculating the atmospheric light A by Equation (3), we can approximation A by the following equation given a threshold d thresold :…”
Section: Atmospheric Scattering Modelmentioning
confidence: 99%
See 3 more Smart Citations