A Novel Feature Selection in Vehicle Detection Through the Selection of Dominant Patterns of Histograms of Oriented Gradients (DPHOG)

Abstract

This paper proposes a novel method that addresses the selection of the dominant patterns of the histograms of oriented gradients (DPHOGs) in vehicle detection. HOG features lead to an expensive classification with high misclassification rates since HOG generates a long vector containing both redundant and ambiguous features (similarities between the vehicle and non-vehicle images). Several modifications of HOG were proposed to resolve these issues such as the vertical histograms of oriented gradient and one that includes position and intensity with HOG; however, these methods still contain some ambiguous features. A feature selection method can exclude these ambiguous features, allowing for better classification rates and a reduction in classification times. The proposed method uses the ideal vectors of the vehicle and non-vehicles images for selecting features in dominant patterns. The segments indicating the differences between the vehicle and non-vehicle classes are the dominant patterns, in which the length of the feature vector is shortened. We performed DPHOG on three standard datasets, in which the kernel extreme learning machine, the support vector machine, K-nearest neighbor, random forest, and deep neural network were used as classifiers. We then compared the performance of the DPHOG with eight well-known feature selection methods and three existing feature extraction methods for vehicle detection. In evaluations with each comparative method concerning the accuracy, true positive, false positive, and F1-score, the DPHOG presented the highest performances with less running time in each dataset.

Improved meta‐heuristic algorithm for selecting optimal features: A big data classification model

“…The selected best feature has the least number of dimensions and hence contributes to better accuracy. 15,20,21 In the wrapper method of FS, it is necessary to have a predetermined algorithm for predicting the best feature subset. This method of filtering produces better accuracy results, but it is expensive for large dataset.…”

“…Apart from this, it is insufficient to handle numeric class problems and it works well only when the datasets are less in dimensions. 20,21 Meta-heuristic algorithms have received wide attention and have been employed to solve various optimization problems. 26 In addition, the fast correlation-based filter algorithm reduces the dimensionality effectively and it cannot handle the feature redundancy.…”

“…The FS is also referred as subset selection, which makes use of the machine learning techniques to select the ideal data applicable for the learning algorithm. The selected best feature has the least number of dimensions and hence contributes to better accuracy 15,20,21 …”

“…However, it limits the reintroduction of redundant features. Apart from this, it is insufficient to handle numeric class problems and it works well only when the datasets are less in dimensions 20,21 . Meta‐heuristic algorithms have received wide attention and have been employed to solve various optimization problems 26 .…”

Analysis and Comparison of FPGA-Based Histogram of Oriented Gradients Implementations

“…The main idea behind the HOG algorithm is to compute gradients as local descriptors and normalize them locally, and then obtain location invariant features which are robust to illumination changes in the image. HOG features have many applications such as face recognition [42], [43], texture classification [44], vehicle detection [45], and human activity The associate editor coordinating the review of this manuscript and approving it for publication was Liang-Bi Chen .…”