Digital image correlation is a well-known optical measurement method for full-field deformation and strain measurements. The quality of speckle images used in digital image correlation calculation can directly affect the measurement accuracy of digital image correlation. In most practical measurement circumstances, a uniform illumination environment is usually required to illuminate the detected object in order to capture speckle images upon different deformed states with uniform background intensity. However, the tested object becomes so large that the adopted light source cannot cover all the interested area with uniform illumination, and the speckle images acquired by CCD camera may have non-uniform background intensity distributions. In this paper, the influence of non-uniform illumination is first analyzed in detail by means of a comparison of experimental results of digital image correlation using speckle patterns with both uniform and non-uniform intensity distributions. Then, a new correctional method based on the combination of the basic retinex theory and the illumination formulae of a point light source is proposed. Finally, a real experiment with non-uniform illumination is implemented to verify the effectiveness of this method.
2
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
To deal with illumination variations in face recognition, a novel two-stage illumination normalization method is proposed in this paper. Firstly, a discrete cosine transform (DCT) is used on the original images in logarithm domain. DC coefficient is set based on the average pixel value of all the within-class training samples and some low frequency AC coefficients are set to zero to eliminate illumination variations in large areas. Secondly, local normalization method, which can minimize illumination variations in small areas, is used on the inverse DCT images. This makes the pixel values on the processed images be close to or equal to that of the normal illumination condition. Experimental results, both on Yale B database and Extended Yale B database, show that the proposed method can eliminate effect of illumination variations effectively and improve performance of face recognition methods significantly. The present method does not demand modeling step and can eliminate the effect of illumination variations before face recognition. In this way, it can be used as a preprocessing step for any existing face recognition method.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.