In this course, the students will be introduced to advanced image analysis methods dedicated to enriching their geo-information problem-solving abilities. Image processing and analysis methods treated in previous courses, such as conventional hard pixel-based classification, do not take into account spatial correlations in images and, therefore, do not completely exploit the information contained in images. In this course, we aim to introduce more specialized image analysis methods. In particular, Support Vector Machine and Random Forest will be taught for multisource classification at the pixel level. Convolutional Neural Networks (CNNs) and Fully Convolutional Neural Networks (FCN) will be introduced for contextual classification. Advantages and challenges related to multi-temporal image analysis will also be discussed. The methods introduced in this course will be applied to real case studies.