Artificial neural networks (ANNs) are a set of mathematical nonlinear tools that allow to model complex systems. Although ANNs in their beginning were developed to mimic the functioning of human brain, it is from inclusion of nonlinear functions that are successfully applied to solve real-problems in different areas such as food science including quality control as well as analytical issues. The development of high-throughput and noninvasive analytical techniques in food science has the challenges of the variable processing related to the huge amount, noise, correlation, and mainly to the fact that functional relation to model is heavily nonlinear. Hyperspectral imaging technique (HIT) has been regarded as a promising analytical tool for food quality control through exploratory, classification, and regression analysis with advantages on traditional methods, such as nondestructive nature, and fast time and intensive data acquisition. HIT is an instrumental tool that integrates spectroscopic and imaging techniques to enable direct identification of different components and their spatial distribution in the tested sample. Spectral spatial data generated by HIT require the application of chemometric techniques for its analysis in order to carry out food quality control. However, where traditional chemometric techniques are unsatisfactory, ANNs are versatile and adaptive mathematical techniques in nonlinear problems providing a better model. In this chapter, the application of ANNs in food quality using HIT data is discussed by presenting a general, simple, and schematic description of the technique including the principles, neuron model, and architecture, its main application in food quality control, and an example of the full procedure of applying these techniques in the hardness modeling of Swiss-type cheese during ripening process, including an open access software developed in Octave.