The fields of image analysis, signal processing, and pattern recognition are fundamental pillars of computer science and electrical engineering, driving innovation from theoretical research to a vast array of industrial applications. Within these disciplines, Evolutionary Computation (EC)—encompassing evolutionary algorithms, swarm intelligence, and related paradigms—has established itself as a powerful and increasingly relevant problem-solving toolkit. The recent and powerful convergence of EC with deep learning, known as Evolutionary Deep Learning, is further pushing the boundaries of what is possible, giving rise to recognized specialties like Evolutionary Image Analysis and Evolutionary Computer Vision. The growth of this domain has been significantly accelerated by modern parallel computing hardware, such as GPUs, whose architecture is exceptionally well-suited to EC algorithms, mitigating their computational demands and even enabling real-time applications.

This tutorial offers a comprehensive framework for understanding and applying Evolutionary Computation (EC) and Evolutionary Deep Learning (EDL) to these key areas. We will begin by presenting a clear taxonomy of the field before demonstrating its practical efficacy through concrete examples in tasks such as edge detection, image segmentation, image classification, and object tracking. A dedicated focus will be placed on Evolutionary Deep Learning, illustrating how EC automates the design of deep learning systems by optimizing neural network architectures (Neural Architecture Search, NAS), tuning learning parameters, and refining transfer functions for Convolutional Neural Networks (CNN). Additionally, we will explore how non-neural-network-based approaches, such as constructing deep models via Genetic Programming (GP), can be effectively applied to signal and image data, offering powerful and versatile solutions to complex challenges.  Finally, the tutorial addresses current challenges and outlines promising directions for future research.