Thresholding is one of the methods of image segmentation, and metaheuristics have got much attention to find optimal thresholds. However, falling into local optima and lack of ability to thoroughly search in complicated search spaces like those in image thresholding problems, have always been a challenge for metaheuristics. In this paper, we applied a dynamic quantum-inspired genetic algorithm (DQGA) that utilizes the probabilistic representation of qubits in quantum computing, an adaptive look-up table for quantum gates, and introduced the lengthening chromosomes in order to promote the search power and local optima avoidance of the algorithm. Image segmentation using the proposed algorithm yielded promising results in comparison with the other competitive algorithms, which proves the merits of DQGA.