Due to pixels or superpixels belonging to different objects in different images are not balanced, and it is difficult to use a limited number of features to describe classifying different objects, these algorithms can not achieve high-performance classification in these tasks.
This paper describes how we applied systems engineering principles to design a high performance embedded classification system in a systematic and well structured way.
Therefore, we can achieve higher performances using more resources (CPU cores, computer, etc).. Furthermore, with our sampling strategy we significantly speed-up the training process of the classifiers while yielding a high performance in classification accuracy.
From this point of view, feature selection can be considered as an essential component required for developing high performance pattern classification systems that use high dimensional data [1 3, 17].
In the field of pattern recognition, the combination of an ensemble of neural networks has been proposed as an approach to the development of high performance image classification systems.
We presented a novel approach that achieves high performances for classification tasks of fingerprint images.
Particularly a six-dimensional subset of the features showed reproducibly high performances in classification experiments.
Recently, we have noticed an emergence of work based on deep learning, which presents high performances on classification and detection tasks.
