Animal Recognition System Based on Convolutional Neural Network
In this paper, the performances of well-known image recognition methods such as Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), Local Binary Patterns Histograms (LBPH) and Support Vector Machine (SVM) are tested and compared with proposed convolutional neural network (CNN) fo...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
VSB-Technical University of Ostrava
2017-01-01
|
| Series: | Advances in Electrical and Electronic Engineering |
| Subjects: | |
| Online Access: | http://advances.utc.sk/index.php/AEEE/article/view/2202 |
| _version_ | 1797826968522063872 |
|---|---|
| author | Tibor Trnovszky Patrik Kamencay Richard Orjesek Miroslav Benco Peter Sykora |
| author_facet | Tibor Trnovszky Patrik Kamencay Richard Orjesek Miroslav Benco Peter Sykora |
| author_sort | Tibor Trnovszky |
| collection | DOAJ |
| description | In this paper, the performances of well-known image recognition methods such as Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), Local Binary Patterns Histograms (LBPH) and Support Vector Machine (SVM) are tested and compared with proposed convolutional neural network (CNN) for the recognition rate of the input animal images. In our experiments, the overall recognition accuracy of PCA, LDA, LBPH and SVM is demonstrated. Next, the time execution for animal recognition process is evaluated. The all experimental results on created animal database were conducted. This created animal database consist of 500 different subjects (5 classes/ 100 images for each class). The experimental result shows that the PCA features provide better results as LDA and LBPH for large training set. On the other hand, LBPH is better than PCA and LDA for small training data set. For proposed CNN we have obtained a recognition accuracy of 98%. The proposed method based on CNN outperforms the state of the art methods. |
| first_indexed | 2024-04-09T12:40:44Z |
| format | Article |
| id | doaj.art-5fc7870fc597419792c27ea5791b8698 |
| institution | Directory Open Access Journal |
| issn | 1336-1376 1804-3119 |
| language | English |
| last_indexed | 2024-04-09T12:40:44Z |
| publishDate | 2017-01-01 |
| publisher | VSB-Technical University of Ostrava |
| record_format | Article |
| series | Advances in Electrical and Electronic Engineering |
| spelling | doaj.art-5fc7870fc597419792c27ea5791b86982023-05-14T20:50:11ZengVSB-Technical University of OstravaAdvances in Electrical and Electronic Engineering1336-13761804-31192017-01-0115351752510.15598/aeee.v15i3.2202923Animal Recognition System Based on Convolutional Neural NetworkTibor Trnovszky0Patrik Kamencay1Richard Orjesek2Miroslav Benco3Peter Sykora4Department of multimedia and information-communication technologies, Faculty of Electrical Engineering, University of Zilina, Univerzitna 8215/1, 010 26 Zilina, SlovakiaUniversity of ZilinaDepartment of multimedia and information-communication technologies, Faculty of Electrical Engineering, University of Zilina, Univerzitna 8215/1, 010 26 Zilina, SlovakiaDepartment of multimedia and information-communication technologies, Faculty of Electrical Engineering, University of Zilina, Univerzitna 8215/1, 010 26 Zilina, SlovakiaDepartment of multimedia and information-communication technologies, Faculty of Electrical Engineering, University of Zilina, Univerzitna 8215/1, 010 26 Zilina, SlovakiaIn this paper, the performances of well-known image recognition methods such as Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), Local Binary Patterns Histograms (LBPH) and Support Vector Machine (SVM) are tested and compared with proposed convolutional neural network (CNN) for the recognition rate of the input animal images. In our experiments, the overall recognition accuracy of PCA, LDA, LBPH and SVM is demonstrated. Next, the time execution for animal recognition process is evaluated. The all experimental results on created animal database were conducted. This created animal database consist of 500 different subjects (5 classes/ 100 images for each class). The experimental result shows that the PCA features provide better results as LDA and LBPH for large training set. On the other hand, LBPH is better than PCA and LDA for small training data set. For proposed CNN we have obtained a recognition accuracy of 98%. The proposed method based on CNN outperforms the state of the art methods.http://advances.utc.sk/index.php/AEEE/article/view/2202animal recognition systemlbphneural networkspcasvm. |
| spellingShingle | Tibor Trnovszky Patrik Kamencay Richard Orjesek Miroslav Benco Peter Sykora Animal Recognition System Based on Convolutional Neural Network Advances in Electrical and Electronic Engineering animal recognition system lbph neural networks pca svm. |
| title | Animal Recognition System Based on Convolutional Neural Network |
| title_full | Animal Recognition System Based on Convolutional Neural Network |
| title_fullStr | Animal Recognition System Based on Convolutional Neural Network |
| title_full_unstemmed | Animal Recognition System Based on Convolutional Neural Network |
| title_short | Animal Recognition System Based on Convolutional Neural Network |
| title_sort | animal recognition system based on convolutional neural network |
| topic | animal recognition system lbph neural networks pca svm. |
| url | http://advances.utc.sk/index.php/AEEE/article/view/2202 |
| work_keys_str_mv | AT tibortrnovszky animalrecognitionsystembasedonconvolutionalneuralnetwork AT patrikkamencay animalrecognitionsystembasedonconvolutionalneuralnetwork AT richardorjesek animalrecognitionsystembasedonconvolutionalneuralnetwork AT miroslavbenco animalrecognitionsystembasedonconvolutionalneuralnetwork AT petersykora animalrecognitionsystembasedonconvolutionalneuralnetwork |