Conference Paper

Hussain, Z.M.: Contourlet structural similarity for facial expression recognition. ICASSP

Sch. of Electr. & Comput. Eng., RMIT Univ., Melbourne, VIC, Australia
DOI: 10.1109/ICASSP.2010.5495357 Conference: Acoustics Speech and Signal Processing (ICASSP), 2010 IEEE International Conference on
Source: IEEE Xplore


This paper presents a novel classification method based on perceptual image quality metrics for facial expression recognition. The features are extracted based on Contourlet sub-bands. Then, the optimum features are selected using minimum redundancy and maximum relevance algorithm (MRMR). The selected features are classified by structural similarity metric in contourlet domain. The proposed method has been extensively assessed using two different databases: the Cohn-Kanade database and the JAFFE database. A series of experiments have been carried out and a comparative study suggests the efficiency of the proposed method in enhancing the classification rates of a number of known algorithms.

Full-text preview

Available from:
  • Source
    • "For example, face detection has been used in many consumer still and video cameras. Face area data can be used for tuning exposure, focus or color [6] [7] [8] [9]. Face area data can also be utilized for image enhancement processing in a camera or post-processing in a computer [10] [11]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Face detection techniques are used for many different applications. For example, face detection is a basic component in many consumer still and video cameras. In this study, we compare the performance of face area data and freely selected local area data for predicting the sharpness of photographs. The local values were collected systematically from images, and for the analyses we selected only the values with the highest performance. The objective sharpness metric was based on the statistics of the wavelet coefficients for the selected areas. We used three image contents whose subjective sharpness values had been measured. The image contents were captured by 13 cameras, and the images were evaluated by 25 subjects. The quality of the cameras ranged from low-end mobile phone cameras to low-end compact cameras. The image contents simulated typical photos that consumers take with their mobile phones. The face area sizes on the images were approximately 0.4, 1.0 or 4.0 %. Based on the results, the face area data proved to be valuable for measuring the sharpness of the photographs if the face size was large enough. When the face area size was 1.0 or 4.0 %, the performance of the measured sharpness values was equal to or better than the sharpness values measured from the best local areas. When the face area was too small (0.4 %), the performance was low compared with the best local areas.
    Full-text · Article · Jan 2011 · Proceedings of SPIE - The International Society for Optical Engineering
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, we have investigated the performance of different multi-resolution transforms in the application of emotion recognition from facial images. Multi-resolution analysis of image provides frequency information along with time information in different scale, orientation and locations. The emotion information from facial images was being captured by different multiresolution algorithm such as Wavelet Transform, Curvelet Transform and Contourlet Transform. Wavelet transform mainly approximate frequency information along with time whereas curvelet transform is best to capture edges information with very few coefficients. Various statistical features obtained from different algorithms have been used to build reference model. The classification part was done using support vector machine (SVM) and K-Nearest Neighbor (KNN) classifier with JAFFE, a Japanese facial emotion and our own In-House facial emotion database. The individual as well as comparative study of different algorithms was done successfully.
    Full-text · Conference Paper · Jan 2011
  • [Show abstract] [Hide abstract]
    ABSTRACT: In this paper a low computation feature space has been proposed to recognize expressions of face images. The image is divided into number of blocks and binary pattern corresponding to each block is generated by modifying the Local Binary Pattern (LBP). The proposed method generates compressed binary pattern of images and therefore, reduced in size. Features are extracted from transformed image using block wise histograms with variable number of bins. For classification we use two techniques, template matching and Support Vector Machine (SVM). Experiments on face images with different resolutions show that the proposed approach performs well for low resolution images. Considering Cohn-Kanade database, the proposed method is compared with LBP feature based methods demonstrating better performance.
    No preview · Chapter · Jan 2015