Object detection using ensemble of linear classifiers with fuzzy adaptive boosting

Abstract

The Adaboost (Freund and Schapire, Eur. Conf. Comput. Learn. Theory 23-37, 1995) chooses a good set of weak classifiers in rounds. On each round, it chooses the optimal classifier (optimal feature and its threshold value) by minimizing the weighted error of classification. It also reweights training data so that the next round would focus on data that are difficult to classify. When determining the optimal feature and its threshold value, a process of classification is employed. The involved process of classification usually performs a hard decision (Viola and Jones, Rapid object detection using a boosted cascade of simple features, 2001; Joo et al., Sci. World J 2014: 1-17, 2014 Friedman et al., Ann. Stat 28:337-407, 2000). In this paper, we extend the process of classification to a soft fuzzy decision. We believe this extension could allow some flexibility to the Adaboost algorithm as well as a good performance especially when the size of a training data set is not large enough. The Adaboost algorithm, in general, assigns a same weight to each training datum on the first round of a boosting process (Freund and Schapire, Eur. Conf. Comput. Learn. Theory 23-37, 1995). We propose to assign different initial weights based on some statistical properties of involved features. In experimental results, we show that the proposed method yields higher performances compared to other ones.

Probability-driven scoring functions in combining linear classifiers

“…Using multiple classifiers instead of one has many advantages. As it was mentioned above, combining multiple classifiers allows building a more complex decision boundary [8]. What is more, it also improves the result stability (reduces variance) and robustness to outliers [14].…”

“…An example of such a technique is to build a multilayer neural network [11], a deep neural network in particular [41]. Another way is to use the multi-classifier approach and build an ensemble of linear classifiers [8].…”

“…For decades many algorithms for building linear classifiers have been developed [2,3,4,5,6,7]. Today they are still used by the machine learning society [8]. This is due to the relatively low computational complexity of training and predicting phases [9].…”

“…Additionally, for some linear classifiers, it is possible to obtain nonlinear decision boundaries using the kernel trick [7]. Nonlinear decision boundaries may also be obtained using ensembles of classifiers [8,11].…”

Real-time human detection in thermal infrared imaging at night using enhanced Tiny-yolov3 network

“…Object detection is an essential task in the computer vision field [1]. This task is to determine the different classes of objects, including a person.…”