Proposal of Novel Histogram Features for Face Detection
نویسندگان
چکیده
This paper presents novel features for face detection in the paradigm of AdaBoost algorithm. Features are multi-dimensional histograms computed from a set of rectangles in the filtered images, and they represent marginal distributions of these rectangles. The filter banks consist of intensity, Laplacian of Gaussian (Difference of Gaussians), and Gabor filters, aiming at capturing spatial and frequency properties of human faces at different scales and different orientations. The best features selected by AdaBoost, pairs of filter and rectangle, can thus be interpreted as boosted marginal distributions of human faces. The result of preliminary experiments demonstrate that the selected features are much more powerful to describe the face pattern than the simple features of Viola and Jones and some variants which can only capture several moments of ONE dimensional histogram in intensity images.
منابع مشابه
Joint Integral Histogram based Adaboost for Face Detection System
Face detection is a crucial step in many vision applica-tions. Since the Viola and Jones face detector, many feature extraction approches based Adaboost are proposed.This paper presents a novel approach to extract effective features for face detection system. Both LBP and three Patch LBP (TPLBP) with joint integral histogram are used to extract features. The joint integral histogram was firstly...
متن کاملSpatial Histogram Features for Face Detection in Color Images
This paper presents a novel face detection approach in color images. We employ spatial histograms as robust features for face detection. The spatial histograms consist of marginal distribution of color image information. Facial texture and shape are preserved by the spatial histogram representation. A hierarchical classifier combining histogram matching and support vector machine is utilized to...
متن کاملA Novel Face Detection Method Based on Over-complete Incoherent Dictionary Learning
In this paper, face detection problem is considered using the concepts of compressive sensing technique. This technique includes dictionary learning procedure and sparse coding method to represent the structural content of input images. In the proposed method, dictionaries are learned in such a way that the trained models have the least degree of coherence to each other. The novelty of the prop...
متن کاملEffect of Histogram Equalization in Face Detection Using Spatial Histogram Features
Many information of an object can simultaneously be conserved by histogram based features. Because of the different intensities of pixels and different lighting conditions in an image, histograms which are employed for extracting features are better to be first equalized. In this paper, the effect of equalization over the histograms used for feature extraction is studied. Experimental results o...
متن کاملImproved Face Detection Using Spatial Histogram Features
In this paper, we improve an object detection approach using spatial histogram features, by applying classifier ensemble. The spatial histogram features can preserve texture and shape information of an object, simultaneously. We train a hierarchical classifier by combining cascade histogram matching and the combination of Multi Layer Perceptrons. The cascade histogram matching is trained via au...
متن کامل