| Autor: |
P. Babu Aurtherson, M. P. Flower Queen, B A Athira Lekshmi, J. Arul Linsely |
| Rok vydání: |
2018 |
| Předmět: |
|
| Zdroj: |
2018 International Conference on Emerging Trends and Innovations In Engineering And Technological Research (ICETIETR). |
| DOI: |
10.1109/icetietr.2018.8529083 |
| Popis: |
In the system proposed here an algorithm for feature extraction and classification of image by using particle swarm optimization (PSO) is implemented. The effective performance in detection or sorting tasks by training of a desired model depends mainly on the features, which has been used in the training phase. The objective of feature design is to briefly define a certain features and key points which are needed to locate or classify required images. An area expert has been performing the feature detection or extraction since now, which in majority of the cases are costly and difficult to utilize and to discover. To automate these process image descriptors has been emerged by particle swarm optimization and to determine the measurement lengthwise for the feature vector continuously using only minimal instances from each class. A group of facial feature classification datasets is being used, also the performance of this method will be evaluated. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
