Using Non-Additive Entropy to Enhance Convolutional Neural Features for Texture Recognition
Here we present a study on the use of non-additive entropy to improve the performance of convolutional neural networks for texture description. More precisely, we introduce the use of a local transform that associates each pixel with a measure of local entropy and use such alternative representation...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-09-01
|
| Series: | Entropy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1099-4300/23/10/1259 |
| _version_ | 1797514683107770368 |
|---|---|
| author | Joao Florindo Konradin Metze |
| author_facet | Joao Florindo Konradin Metze |
| author_sort | Joao Florindo |
| collection | DOAJ |
| description | Here we present a study on the use of non-additive entropy to improve the performance of convolutional neural networks for texture description. More precisely, we introduce the use of a local transform that associates each pixel with a measure of local entropy and use such alternative representation as the input to a pretrained convolutional network that performs feature extraction. We compare the performance of our approach in texture recognition over well-established benchmark databases and on a practical task of identifying Brazilian plant species based on the scanned image of the leaf surface. In both cases, our method achieved interesting performance, outperforming several methods from the state-of-the-art in texture analysis. Among the interesting results we have an accuracy of 84.4% in the classification of KTH-TIPS-2b database and 77.7% in FMD. In the identification of plant species we also achieve a promising accuracy of 88.5%. Considering the challenges posed by these tasks and results of other approaches in the literature, our method managed to demonstrate the potential of computing deep learning features over an entropy representation. |
| first_indexed | 2024-03-10T06:35:04Z |
| format | Article |
| id | doaj.art-cb7a6b0f9c5747a2bdfb1aaf28dabb87 |
| institution | Directory Open Access Journal |
| issn | 1099-4300 |
| language | English |
| last_indexed | 2024-03-10T06:35:04Z |
| publishDate | 2021-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Entropy |
| spelling | doaj.art-cb7a6b0f9c5747a2bdfb1aaf28dabb872023-11-22T18:10:15ZengMDPI AGEntropy1099-43002021-09-012310125910.3390/e23101259Using Non-Additive Entropy to Enhance Convolutional Neural Features for Texture RecognitionJoao Florindo0Konradin Metze1Institute of Mathematics, Statistics and Scientific Computing, University of Campinas, Campinas 13083-859, BrazilFaculty of Medical Sciences, State University of Campinas (UNICAMP), Campinas 13083-894, BrazilHere we present a study on the use of non-additive entropy to improve the performance of convolutional neural networks for texture description. More precisely, we introduce the use of a local transform that associates each pixel with a measure of local entropy and use such alternative representation as the input to a pretrained convolutional network that performs feature extraction. We compare the performance of our approach in texture recognition over well-established benchmark databases and on a practical task of identifying Brazilian plant species based on the scanned image of the leaf surface. In both cases, our method achieved interesting performance, outperforming several methods from the state-of-the-art in texture analysis. Among the interesting results we have an accuracy of 84.4% in the classification of KTH-TIPS-2b database and 77.7% in FMD. In the identification of plant species we also achieve a promising accuracy of 88.5%. Considering the challenges posed by these tasks and results of other approaches in the literature, our method managed to demonstrate the potential of computing deep learning features over an entropy representation.https://www.mdpi.com/1099-4300/23/10/1259texture recognitionconvolutional neural networksnon-additive entropyimage descriptors |
| spellingShingle | Joao Florindo Konradin Metze Using Non-Additive Entropy to Enhance Convolutional Neural Features for Texture Recognition Entropy texture recognition convolutional neural networks non-additive entropy image descriptors |
| title | Using Non-Additive Entropy to Enhance Convolutional Neural Features for Texture Recognition |
| title_full | Using Non-Additive Entropy to Enhance Convolutional Neural Features for Texture Recognition |
| title_fullStr | Using Non-Additive Entropy to Enhance Convolutional Neural Features for Texture Recognition |
| title_full_unstemmed | Using Non-Additive Entropy to Enhance Convolutional Neural Features for Texture Recognition |
| title_short | Using Non-Additive Entropy to Enhance Convolutional Neural Features for Texture Recognition |
| title_sort | using non additive entropy to enhance convolutional neural features for texture recognition |
| topic | texture recognition convolutional neural networks non-additive entropy image descriptors |
| url | https://www.mdpi.com/1099-4300/23/10/1259 |
| work_keys_str_mv | AT joaoflorindo usingnonadditiveentropytoenhanceconvolutionalneuralfeaturesfortexturerecognition AT konradinmetze usingnonadditiveentropytoenhanceconvolutionalneuralfeaturesfortexturerecognition |