Abstract : Metaheuristics are a class of algorithms that deploy strategies used to guide search processes to find optimal solutions to optimization problems for which it is difficult to find exact solutions for either due to their inherent complexity or due to limited available resources. Metaheuristics are mostly inspired from nature by the collective behavior of such things as ants, flocks of birds, chromosomes and others. Such metaheuristics try to simulate these effective cooperative behaviors to explore better solutions in the search space. In nature, humans are effective explorers who act collectively and cooperatively to explore the world around them. Still, to make themselves even better explorers, humans have devised many imaging technologies such as maps, telescopes, astrolabes, radars and satellites to speed up and increase the level of success of their search expeditions. In this research we propose, develop and use imaging techniques to support the search effort of metaheuristics in general during the phases of population initialization as well as in the main search phase of a given metaheuristic. The proposed approach includes a set of basic imaging techniques, four of which are to be used during the initialization phase of a metaheuristic and one of which to be used during the main search phase of a metaheuristic. To evaluate these techniques, they have been implemented to extend the particle swarm optimization algorithm used to tune the weights and biases of multiple neural networks trained to classify objects in seven different datasets. The performance of the proposed approaches has been evaluated against each other, and against the particle swarm optimization algorithm alone. The evaluation criteria are based on the performance classification accuracy and training runtime. The results show that the extended particle swarm optimization using the proposed approach generally outperforms particle swarm optimization algorithm alone.