With the rapid development of network and information technology, people are paying more attention to the security of information, particularly digital image protection, and numerous image encryption algorithms have been proposed. Because of the non-periodicity and sensitivity to the initial value, chaotic map seems to be a tool that can be used for image encryption. Recent work has shown that the quantum chaotic map is very sensitive to a slight change in the initial conditions. The sensitivity of the quantum chaotic map can extremely increase the complexity of the encryption algorithms. At present, a wide variety of image encryption algorithms based on DNA encoding have been proposed, where the image pixel values are encoded by four base pairs of DNA to achieve image pixel diffusion, but most of the methods in selecting the DNA encoding rules are fixed. In this paper, we present a new image encryption algorithm with the quantum chaotic map, the Lorenz chaotic map and DNA coding, which uses four base pairs of DNA dynamically select eight types of DNA encoding rules and eight types of DNA addition and XOR rules. This strategy has led to a significant enhancement in reliability and security. Through simulation experiments, the results of the histograms, correlations and the number of pixels change rate (NPCR) analyses indicate that the proposed algorithm possesses a high level of security and can successfully resist different attacks such as brute-force attacks and statistical attacks.