This paper proposes a Gaussian mixture based image enhancement method which uses particle swarm optimization
(PSO) to have an edge over other contemporary methods. The proposed method uses the guassian mixture model to
model the lightness histogram of the input image in CIEL*a*b* space. The intersection points of the guassian
components in the model are used to partition the lightness histogram. . The enhanced lightness image is generated by
transforming the lightness value in each interval to appropriate output interval according to the transformation function
that depends on PSO optimized parameters, weight and standard deviation of Gaussian component and cumulative
distribution of the input histogram interval. In addition, chroma compensation is applied to the resulting image to reduce
washout appearance. Experimental results show that the proposed method produces a better enhanced image compared to
the traditional methods. Moreover, the enhanced image is free from several side effects such as washout appearance,
information loss and gradation artifacts.
Changing illumination cause the measurements of object colors to be biased toward chromaticity of illuminants. Various color constancy algorithms are already exist to remove the chromaticity of illuminants in an image for improving image quality. Recently, NMFsc(nonnegative matrix factorization with sparseness constraint) was introduced to extract the illuminant and reflectance component in an image. NMFsc extract illuminant component and reflectance component by using nonnegative matrix decomposition and sparseness constraints. However, if an image has a chromaticity distribution dominated by a particular chromaticity, the sparse constraint values include that dominant chromaticity, thereby inducing color distortion. Therefore, the proposed method modified the matrix decomposition in NMFsc by using standard deviation and K-means algorithm in chromaticity space. Next, non-negative matrix decomposition and sparseness constraints are performed on an image. Subsequently, illumination is estimated by combining the low sparse constraint values that excludes the dominant chromaticity. The performance of the proposed method is evaluated by using angular error for Ciurea 11,346 image data set. Experimental results illustrate that the proposed method reduces the angular error over previous methods.
To accurately represent the colors in a real scene, a multi-channel camera system is necessary. One of the applications of
the data acquired with the multi-channel camera system is the spectral reflectance estimation. One of the most widely
used methods to estimate the spectral reflectance is the Wiener estimation. While simple and accurate in controlled
conditions, the Wiener estimation does not perform as well with real scene data. Therefore, the adaptive Wiener
estimation has been proposed to improve the performance of the Wiener estimation. The adaptive Wiener estimation
uses a similar training set that was adaptively constructed from the standard training set according to the camera
responses. In this paper, a new way of constructing such similar training set using the correlation between each spectral
reflectance in the standard training set and the first approximation of the spectral reflectance that was obtained by the
Wiener estimation is proposed. The experimental results showed that the proposed method is more accurate than the
conventional Wiener estimations.