A Revisit Histogram of Oriented Descriptor for Facial Color Image Classification Based on Fusion of Color Information

Abstract

Histogram of Oriented Gradient (HOG) is a robust descriptor which is widely used in many real-life applications, including human detection, face recognition, object counting, and video surveillance. In order to extract HOG descriptor from color images whose information is three times more than the grayscale images, researchers currently apply the maximum magnitude selection method. This method makes the information of the resulted image is reduced by selecting the maximum magnitudes. However, after we extract HOG using the unselected magnitudes of the maximum magnitude selection method, we observe that the performance is better than using the maximum magnitudes in several cases. Therefore, in this paper, we propose a novel approach for extracting HOG from color images such as Color Component Selection and Color Component Fusion. We also propose the extended kernels in order to improve the performance of HOG. With our new approaches in the color component analysis, the experimental results of several facial benchmark datasets are enhanced with the increment from 3 to 10% of accuracy. Specifically, a 95.92% of precision is achieved on the Face AR database and 75% on the Georgia Face database. The results are better more than 10 times compared with the original HOG approach.

Machine Learning Approach for Facial Image Detection System

“…Good accuracy rates (90-93%) were obtained in Setiawan and Muttaqin [10], Silva et al [12], Moreano and Palomino [14], and Nguyen-Quoc and Hoang [15] using different feature extractions like PCA, LDA, and Gabor wavelets, with different classifiers like KNN, SVM, HOG, and GA. Using Euclidean distance and random forest, Barnouti et al [11], Szmuro and Osowski [17] obtained good accuracy of 87.5% and 83.3%, respectively.…”

“…Using Euclidean distance and random forest, Barnouti et al [11], Szmuro and Osowski [17] obtained good accuracy of 87.5% and 83.3%, respectively. With an accuracy of 77.78%, Zhou and Zhang [15] obtained a good result using novel collaborative representation at the NCRC. Rashed and Hamd [16] used the J48 classifier and reached 94.337%.…”

“…Color Component Selection and Color Component Fusion were proposed by Nguyen-Quoc and Hoang [15] as methods for extracting HOG from color images, as well as extended kernels to improve HOG performance. With our new approaches in color component analysis, the experimental results of several facial benchmark datasets are enhanced with high accuracy.…”

Efficient Iris Image Recognition System Based on Machine Learning Approach

“…F-measure: The harmonic medium value of recall and accuracy, where F1 is best at one and worst at zero. [32], [33]…”

Facial-Based Autism Classification Using Support Vector Machine Method

“…This process facilitates the extraction of distinctive features essential for robust object detection and recognition, making HOG a prominent choice in the field of computer vision. [21], [22], [23]. SURF (Speeded-Up Robust Features) represents an enhanced iteration of the Scale-Invariant Feature Transform (SIFT), renowned for its scale-invariant feature transformation capabilities.…”