Analysis of Multi-Threshold Image Segmentation Method Based on Improved Sparrow Search Algorithm

Image segmentation, as an important part of network image processing, is mainly used to obtain the target region of interest. Once proposed, it has been widely used in multi-field image processing work. The common methods include Kapur entropy, maximum inter-class variance method, etc. According to the practical research, the traditional multi-threshold image segmentation method does not meet the current technical innovation requirements, and the actual segmentation speed and accuracy are not high, so the researchers put forward a multi-threshold image segmentation method based on the improved sparrow search algorithm. On the one hand, the sparrow search algorithm is continuously optimized by using the flight idea in the bird swarm algorithm, on the other hand, the segmentation effect of Kapur entropy and maximum inter-class variance method is compared and analyzed. Standard deviation and objective function are selected for evaluation and analysis. The final results show that the algorithm in this paper has strong pioneering and searching power, and can effectively improve the speed and accuracy of image segmentation.