Face Identification Based on Contrast Limited Adaptive Histogram

Equalization (CLAHE).

Gibran Benitez-Garcia, Jesus Olivares-Mercado, Gualberto Aguilar-Torres,

Gabriel Sanchez-Perez and Hector Perez-Meana

Mechanical and Electrical Engineering School of National Polytechnic

Institute of Mexico. Mexico, Mexico D.F.

Introduction

This paper proposes a face identification method based on Contrast Limited Adaptive Histogram Equalization (CLAHE) robust to facial expressions, occlusion (නිරෝධනය) and especially to illumination (ආලෝකය) changes.

Approach

This is based on Eigenphases algorithm for feature extraction, the Principal Components Analysis (PCA) and the Phase Spectrum was used as feature extraction stage, and Support Vector Machine (SVM) as a classifier.

Method

• Capture: - Capture the pattern (data acquisition of this method consists in taking a picture).
• Feature Extraction: - Extract the feature of captured pattern (Pattern is converted into vector feature).
• Classification: - Generate template based on vector feature (SVM).

Strengthes

• The results were obtained using a database that includes face images of 120 subjects (60 males and 60 females) with illumination changes, facial expressions and partial occlusion. The proposed method provides results with a correct recognition up to 97%.
•  Result obtained are greater than 90% and in the best cases obtained a 97.36% which is acceptable for a face recognition system.

Advantages

• The proposed system shown to be robust to changes in the database used, which are illumination changes and partial occlusion by using sunglasses.

Download Research Paper

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.217.8286&rep=rep1&type=pdf

Mood Detection: Implementing a facial expression recognition system

 Neeraj Agrawal, Rob Cosgriff and Ritvik Mudur

Introduction

The purpose of this research is to develop a mood detection system which can achieve high recognition rates of facial expression across multiple databases.

Approach

An approach using eigenface masks was developed and implemented to classify facial expressions across publicly available databases. These learned masks along with feature vectors of the image were used to train the SVM (Support Vector Machines).

Method

• Pre-processing image data by normalizing and applying a simple mask.
• Extracting certain (facial) features using PCA (Principal Component Analysis) and Gabor filters.
• Using SVMs for classification and recognition of expressions.

Feature extraction

a.     Normalized pixel intensities (Every image in training set normalized by subtracting the mean of all training set images. The masked region is then converted to a column vector which forms the feature vector).

b.     Gabor filter representations (Gabor filters are often used in image processing and are based on physiological studies of the human visual cortex. The use of Gabor filtered facial images has been shown to result in improved accuracy for facial expression recognition).

Eigenface masks (eigenfaces are used to generate a mask that eliminates pixels that vary little across training samples in different labels. In this system, they modify the approach to generate a separate mask for each expression class).

Strength

When apply Eigenface mask the results were similar for both the normalized pixel value and the gabor filtered feature vectors.

Advantages

• This approach works reasonably well in capturing expression-relevant facial information across all databases.

Conclusion

The results were similar for both the normalized pixel value and the gabor filtered feature vectors, with neither representation being clearly superior to the other. The performance across databases is indicative of the method’s robustness to variations in facial structure and skin tone when recognizing the expression.

However, both representations of the feature vectors end up being very high dimensional. This hurts the run time of the algorithm hampering the ability to use it in real time, as well as leading to possible over fitting of the data during training. Increasing the number of available training images would help to compensate for over fitting.

Download Research Paper

http://cs229.stanford.edu/proj2009/AgrawalCosgriffMudur.pdf

Facial Expression recognition using Fuzzy Inference System

Maedeh Rasoulzadeh

Rasoulzade.m@gmail.com

Introduction

This paper proposes a novel fuzzy method for facial expression recognition on still images of the face.

Approach

The new technique involves in extracting mathematical data from some special regions of the face and fed them to a fuzzy rule-based system. Fuzzification operation uses triangular membership functions for both input and output.

Method

1. Input image is preprocessed by wiener filter for smoothing and more distinction between face and background.
2. 5 basic regions were extracted on face area of the preprocessed image by defining 10 lines.
3. Feature extraction is performed (image energy, mean and variance, were calculated and considered as features that are fed to mamdani-type fuzzy system for expression recognition).

Strengths

·        Average recognition rate of expression is 92.3%.

·        Superiority of proposed system to existing ones.

 

Advantages

• This system is capable of recognizing 6 basic human facial expressions that are happiness, surprise, anger, fear, disgust and sadness.
• System is simple & highly accurate.
•  Experimental results on JAFFEE database indicate good performance of the developed technique.

Disadvantages

•  It needs static image as input for giving expression as the output.

Future work

Will include emotion recognition based on considering more facial regions, improving rules and combining other classifiers to the fuzzy system and for better performance.

Download Research Paper

http://www.ijeit.com/vol%201/Issue%204/IJEIT1412201204_50.pdf