scholarly journals HUMAN FACE DETECTION METHODS USING COMBINED CASCADE OF CLASSIFIERS

2014 ◽  
pp. 114-125
Author(s):  
Ihor Paliy
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Face Detection ◽  
Detection Methods ◽  
Classification Error ◽  
Asymmetric Structure ◽  
Human Face ◽  
High Detection Rate ◽  
Structure Generation ◽  
Human Face Detection

The paper presents the improved human face detection method using the combined cascade of classifiers with the improved face candidates’ verification approach, as well as methods and algorithms for the verification level (convolutional neural network) structure generation and training. The combined cascade shows a high detection rate with a very small number of false positives and the proposed candidates’ verification approach is in almost 3 times faster in comparison with the classic verification scheme. The network’s structure generation method allows creating the sparse asymmetric structure of the convolutional neural network automatically. The improved training method uses the adaptive training examples ratio to obtain a trained network with a very low classification error for the positive examples.

AN EFFICIENT APPROACH FOR FACE DETECTION AND FACIAL FEATURE LOCATION USING PRUNE-AND-SEARCH TECHNIQUE

2004 ◽  
Vol 18 (06) ◽  
pp. 987-1005 ◽  
Author(s):  
CHIN-CHEN CHANG ◽  
YUAN-HUI YU
Keyword(s):  
Face Detection ◽  
Facial Feature ◽  
Facial Features ◽  
Base Line ◽  
Human Face ◽  
Feature Location ◽  
Search Technique ◽  
Efficient Approach ◽  
Human Face Detection ◽  
Geometric Knowledge

This paper proposes an efficient approach for human face detection and exact facial features location in a head-and-shoulder image. This method searches for the eye pair candidate as a base line by using the characteristic of the high intensity contrast between the iris and the sclera. To discover other facial features, the algorithm uses geometric knowledge of the human face based on the obtained eye pair candidate. The human face is finally verified with these unclosed facial features. Due to the merits of applying the Prune-and-Search and simple filtering techniques, we have shown that the proposed method indeed achieves very promising performance of face detection and facial feature location.

scholarly journals Normal and Abnormal Human Face Detection Based on DCT and FFT Techniques - A Proposed Method

2021 ◽  
Author(s):  
Samir Bandyopadhyay ◽  
Shawni Dutta ◽  
Vishal Goyal ◽  
Payal Bose
Keyword(s):  
Face Detection ◽  
Fourier Transformation ◽  
Input Image ◽  
The Other ◽  
Fast Fourier Transformation ◽  
Length Variation ◽  
Human Face ◽  
Discrete Cosine Transformation ◽  
Cosine Transformation ◽  
Human Face Detection

In today’s world face detection is the most important task. Due to the chromosomes disorder sometimes a human face suffers from different abnormalities. For example, one eye is bigger than the other, cliff face, different chin-length, variation of nose length, length or width of lips are different, etc. For computer vision currently this is a challenging task to detect normal and abnormal face and facial parts from an input image. In this research paper a method is proposed that can detect normal or abnormal faces from a frontal input image. This method used Fast Fourier Transformation (FFT) and Discrete Cosine Transformation of frequency domain and spatial domain analysis to detect those faces.

scholarly journals Human Face Detection in Color Images

2021 ◽  
Author(s):  
Jun Gao
Keyword(s):  
Face Detection ◽  
Skin Color ◽  
Video Retrieval ◽  
Color Images ◽  
Human Face ◽  
Detection Scheme ◽  
Retrieval Systems ◽  
Human Face Detection ◽  
The Face ◽  
Video Retrieval Systems

Detection of human face has many realistic and important applications such as human and computer interface, face recognition, face image database management, security access control systems and content-based indexing video retrieval systems. In this report a face detection scheme will be presented. The scheme is designed to operate on color images. In the first stage of algorithm, the skin color regions are detected based on the chrominance information. A color segmentation stage is then employed to make skin color regions to be divided into smaller regions which have homogenous color. Then, we use the iterative luminance segmentation to further separate the detected skin region from other skin-colored objects such as hair, clothes, and wood, based on the high variance of the luminance component in the neighborhood of edges of objects. Post-processing is applied to determine whether skin color regions fit the face constrains on density of skin, size, shape and symmetry and contain the facial features such as eyes and mouths. Experimental results show that the algorithm is robust and is capable of detecting multiple faces in the presence of a complex background which contains the color similar to the skin tone.

Sign in / Sign up

Export Citation Format

Share Document