Chapter 6 provides a comprehensive review of data/image processing methods, computer science, and digital techniques. Indeed, data processing is one of the most important parts in remote sensing. A number of available and efficient algorithms including statistical, spectral, wavelet, correlation techniques, fusion, multiresolution analysis, and machine vision are discussed and analyzed in detailed. Image enhancement, segmentation, classification, and feature extraction are critical issues for detection purposes. The term “features” means not only extracted image morphological attributes, but also products of the digital processing. Among them Fourier spectra, fractal irregularity, wavelet maps, fused patterns, synthesized textures, and others products are perceived as computer vision signatures related to certain geophysical phenomena. Synergy algorithm, involving both global and local image processing is the most promising computer technique for analysis of complex multispectral (hyperspectral) optical data. Ultimately, synergy techniques facilitate new prospective developments in automatic detection technology based on remote sensing observations. Methods and algorithms of data/image processing are constantly improved and become more sophisticated to achieve higher level of understanding, interpretation, and application of remotely sensed data. The choice of robust image processing is defined, first of all, by significant scientific experience, knowledge, and establishment of computer vision methods. In particular, nonacoustic detection technology requires the integration of digital processing and physics-based modeling of remotely sensed data (signals, images, signatures). The implementation is based on the state-of-the-art diversification methods and algorithms. Most of them are considered in this chapter.