Image processing, among its vast applications, has proven particular efficiency in quality control systems. Quality control systems such as the ones in the food industry, fruits and meat industries, pharmaceutic, and hardness testing are highly dependent on the accuracy of the algorithms used to extract image feature vectors and process them. Thus, the need to build better quality systems is tied to the progress in the field of image processing.  Color histograms have been widely and successfully used in many computer vision and image processing applications. However, they do not include any spatial information. We propose statistical models to integrate both color and spatial information. Our first model is based on finite mixture models which have been applied to different computer vision, image processing and pattern recognition tasks. The majority of the work done concerning finite mixture models has focused on mixtures for continuous data. However, many applications involve and generate discrete data for which discrete mixtures are better suited. In this thesis, we investigate the problem of discrete data modeling using finite mixture models. We propose a novel, well motivated mixture that we call a multinomial generalized Dirichlet mixture.  Our second model is based on finite multiple-Bernoulli mixtures. For the estimation of the model's parameters, we use a maximum a posteriori (MAP) approach through deterministic annealing expectation maximization (DAEM). Smoothing priors to the components parameters are introduced to stabilize the estimation. The selection of the number of clusters is based on stochastic complexity