Image segmentation refers to extracting regions of interest from images. Due to the characteristics of human tissues, the boundaries of medical images are blurred and the contrast is low, so it is difficult to segment medical images [1]. Literature [2] proposes a multi-level threshold segmentation method for brain images based on binary crossover real-coded genetic algorithm. Literature [3] proposes a two-dimensional entropy multi-threshold image segmentation algorithm based on firefly algorithm, which can effectively improve the image segmentation speed, but the image segmentation accuracy is low due to the limitation of search space. Literature [4] uses particle swarm optimization to optimize the parameter q of two-dimensional Tsallis entropy, which can segment images well. Literature [5] Aiming at the problem of large computation of two-dimensional maximum entropy segmentation image, artificial bee colony algorithm is applied to two-dimensional maximum entropy optimization, and the results show that this method has strong anti-noise ability and fast convergence speed.

In order to improve the effect of medical image segmentation, aiming at the influence of the selection of parameter Q in two-dimensional Tsallis entropy threshold method on the image segmentation effect, a medical image segmentation algorithm based on firefly algorithm to optimize two-dimensional Tsallis entropy is proposed. Finally, the effectiveness of the algorithm is proved by simulation research.

Author information:

Xu Hao 1, Wang Shuang 2

(1. Eye Optometry Hospital affiliated to Wenzhou Medical University, Wenzhou, Zhejiang 325000; 2. Shaanxi Xi 'an University of Science and Technology 7 10054)