In image analysis and pattern recognition, the quality of the input data determines the quality of the output (e.g. accuracy), which is known as the GIGO (garbage in, garbage out) principle. For a given problem, the input data to any machine learning or data mining algorithm is almost always expressed by a number of features (attributes or variables) showing different properties of the problem. Therefore, the quality of the feature space is a key for successfully solving any image analysis and pattern recognition problem.
Computational intelligence techniques, mainly evolutionary computation, neural networks, and fuzzy systems, have been shown to be effective tools in image analysis and pattern recognition. However, their performance is still limited or influenced when the feature space is of poor quality, which may be that the dimensionality is too high (i.e. the number of features is too big) leading to the “curse of dimensionality”, features are not equally important, some features are irrelevant, redundant or even noisy, the original features are not informative enough, the features are not linearly separable, and so on. All these factors may lead to various performance limitations. For example in image classification problems, these will lead to low classification accuracy, a long training time, a complex classifier, etc.
The IEEE Symposium on Computational Intelligence in Feature Analysis, Selection, and Learning in Image and Pattern Recognition (FASLIP) aims to offer world-wide researchers in those fields as well as people from industry an opportunity to present their latest research and to discuss current developments and applications, besides fostering closer future interaction between members of the two communities. FASLIP welcomes contributions that investigate the new theories, methods or applications of different computational intelligence paradigms to feature analysis, selection, and learning in solving various image and pattern recognition tasks.
Topics of interest include but are not limited to:
Feature ranking/weighting
Feature selection
Feature construction/extraction
Multi-objective feature selection, construction or extraction
Feature analysis on high-dimensional and large-scale data
Analysis on computational intelligence for ffeature selection and construction/extraction algorithms
Evolutionary ccomputation for feature analysis
Neural networks for feature analysis
Fuzzy systems for feature analysis
Hybridisation of evolutionary computation and neural networks, and fuzzy systems for feature selection and construction
Hybridisation of evolutionary computation and machine learning, information theory, statistics, mathematical modelling, etc., for feature analysis
Feature analysis in classification, clustering, regression, image analysis, signal processing, and other tasks
12月06日
2016
12月09日
2016
初稿截稿日期
终稿截稿日期
注册截止日期
留言