Open AccessHistograms of Oriented Gradients for cats-dogs detection
Author(s) -
Jiaxing Wu,
Zixuan Yang,
Ting Wang
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1314/1/012176
Subject(s) - pattern recognition (psychology) , histogram , kernel (algebra) , quadratic equation , support vector machine , principal component analysis , artificial intelligence , computer science , quadratic function , mathematics , image (mathematics) , combinatorics , geometry
We study the question based on grids of Histograms of Oriented Gradient (HOG) and support vector machine(SVM),adopting different kernel functions to distinguish cat from dog with robust visual object recognition [1]. We show experimentally that the quadratic descriptors significantly outperform others kernel functions and with the increasing number of the samples, the performance of the quadratic is getting better and better. Beside, in order to reduce the complexity, we also apply principal Component Analysis, which is known as pca, and we also study the influence of the number of features on performance, concluding that the line is quadratic.