Training Artificial Neural Network Using Back-Propagation & Particle Swarm Optimization for Image Skin Diseases
This work is devoted to compression Image Skin Diseases by using Discrete Wavelet Transform (DWT) and training Feed-Forward Neural Networks (FFNN) by using Particle Swarm Optimization(PSO) and compares it with Back-Propagation (BP) neural networks in terms of convergence rate and accuracy of results...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Unviversity of Technology- Iraq
2011-09-01
|
| Series: | Engineering and Technology Journal |
| Subjects: | |
| Online Access: | https://etj.uotechnology.edu.iq/article_32977_cad064c50e4688f8f84122236d88d244.pdf |
| _version_ | 1797325457336565760 |
|---|---|
| author | Hanan A. R. Akkar Samem Abass Salman |
| author_facet | Hanan A. R. Akkar Samem Abass Salman |
| author_sort | Hanan A. R. Akkar |
| collection | DOAJ |
| description | This work is devoted to compression Image Skin Diseases by using Discrete Wavelet Transform (DWT) and training Feed-Forward Neural Networks (FFNN) by using Particle Swarm Optimization(PSO) and compares it with Back-Propagation (BP) neural networks in terms of convergence rate and accuracy of results .The comparison between the two techniques will be mentioned. A MATLAB 6.5 program is used in simulation. The structure Artificial Neural Network (ANN) of training image skin diseases is proposed as follows: 1- The proposed structure of NN that performs three compressions Images Skin training by BP algorithms with log sigmoid activation function, and three neurons in output layer. 2- The proposed structure of FFNN using PSO that performs three compressions Images Skin with hardlim activation function, and three neurons in output layer. The results obtained using PSO are compared to those obtained using BP. Learning iterations (602-4700 epoch), convergence time (1sec.- 100 sec.), number of initial weights (1set - 75set), number of derivatives (0 - 38 derivatives) and accuracy (60% - 100%) are used as performance measurements. The obtained Mean Square Error (MSE) is 7 10 - to check the performance of algorithms. The results of the proposed neural networks performed indicate that PSO can be a superior training algorithm for neural networks, which is consistent with other research in the area. |
| first_indexed | 2024-03-08T06:09:46Z |
| format | Article |
| id | doaj.art-b17c9c1b78ad4f4681434af8b5f607dc |
| institution | Directory Open Access Journal |
| issn | 1681-6900 2412-0758 |
| language | English |
| last_indexed | 2024-03-08T06:09:46Z |
| publishDate | 2011-09-01 |
| publisher | Unviversity of Technology- Iraq |
| record_format | Article |
| series | Engineering and Technology Journal |
| spelling | doaj.art-b17c9c1b78ad4f4681434af8b5f607dc2024-02-04T17:42:24ZengUnviversity of Technology- IraqEngineering and Technology Journal1681-69002412-07582011-09-0129132739275510.30684/etj.29.13.1232977Training Artificial Neural Network Using Back-Propagation & Particle Swarm Optimization for Image Skin DiseasesHanan A. R. AkkarSamem Abass SalmanThis work is devoted to compression Image Skin Diseases by using Discrete Wavelet Transform (DWT) and training Feed-Forward Neural Networks (FFNN) by using Particle Swarm Optimization(PSO) and compares it with Back-Propagation (BP) neural networks in terms of convergence rate and accuracy of results .The comparison between the two techniques will be mentioned. A MATLAB 6.5 program is used in simulation. The structure Artificial Neural Network (ANN) of training image skin diseases is proposed as follows: 1- The proposed structure of NN that performs three compressions Images Skin training by BP algorithms with log sigmoid activation function, and three neurons in output layer. 2- The proposed structure of FFNN using PSO that performs three compressions Images Skin with hardlim activation function, and three neurons in output layer. The results obtained using PSO are compared to those obtained using BP. Learning iterations (602-4700 epoch), convergence time (1sec.- 100 sec.), number of initial weights (1set - 75set), number of derivatives (0 - 38 derivatives) and accuracy (60% - 100%) are used as performance measurements. The obtained Mean Square Error (MSE) is 7 10 - to check the performance of algorithms. The results of the proposed neural networks performed indicate that PSO can be a superior training algorithm for neural networks, which is consistent with other research in the area.https://etj.uotechnology.edu.iq/article_32977_cad064c50e4688f8f84122236d88d244.pdfartificial neural networkparticle swarm optimizationbackpropagationwavelet transforms |
| spellingShingle | Hanan A. R. Akkar Samem Abass Salman Training Artificial Neural Network Using Back-Propagation & Particle Swarm Optimization for Image Skin Diseases Engineering and Technology Journal artificial neural network particle swarm optimization back propagation wavelet transforms |
| title | Training Artificial Neural Network Using Back-Propagation & Particle Swarm Optimization for Image Skin Diseases |
| title_full | Training Artificial Neural Network Using Back-Propagation & Particle Swarm Optimization for Image Skin Diseases |
| title_fullStr | Training Artificial Neural Network Using Back-Propagation & Particle Swarm Optimization for Image Skin Diseases |
| title_full_unstemmed | Training Artificial Neural Network Using Back-Propagation & Particle Swarm Optimization for Image Skin Diseases |
| title_short | Training Artificial Neural Network Using Back-Propagation & Particle Swarm Optimization for Image Skin Diseases |
| title_sort | training artificial neural network using back propagation particle swarm optimization for image skin diseases |
| topic | artificial neural network particle swarm optimization back propagation wavelet transforms |
| url | https://etj.uotechnology.edu.iq/article_32977_cad064c50e4688f8f84122236d88d244.pdf |
| work_keys_str_mv | AT hananarakkar trainingartificialneuralnetworkusingbackpropagationparticleswarmoptimizationforimageskindiseases AT samemabasssalman trainingartificialneuralnetworkusingbackpropagationparticleswarmoptimizationforimageskindiseases |