论文部分内容阅读
为提高图像分割精度,并针对传统模糊均值聚类算法存在的聚类中心选取问题,提出一种改进果蝇算法优化模糊均值聚类算法的图像分割算法。首先,根据上一代最优味道浓度判断值和当前迭代次数来自适应调整果蝇算法进化步长,在初期的步长大可避免陷入局部最优,进化后期果蝇移动步长变小可获得更高的收敛精度,加快收敛速度;然后,采用改进果蝇算法选择模糊均值聚类算法的初始聚类中心,实现图像分割;最后,采用仿真实验测试算法的性能,实验结果表明,相对于对比算法,算法在分割正确率、分割速度及鲁棒性上均更优,具有更广的应用前景。
In order to improve the accuracy of image segmentation and to solve the problem of clustering center selection based on traditional fuzzy mean clustering algorithm, an image segmentation algorithm based on improved fuzzy-mean clustering algorithm is proposed. Firstly, the evolutionary step size of Drosophila algorithm is adaptively adjusted according to the optimal taste concentration of the previous generation and the current iteration times. The initial step size can avoid falling into the local optimum, and the fader move step size becomes smaller in the late stage of evolution to obtain more High convergence precision and fast convergence rate. Then, the improved Drosophila algorithm is used to select the initial clustering center of the fuzzy mean clustering algorithm to achieve image segmentation. Finally, the performance of the algorithm is tested by simulation experiments. The experimental results show that, compared with the contrast algorithm , The algorithm is better in segmentation accuracy, segmentation speed and robustness, and has a broader application prospect.