数字图像处理方向学术论文
Abstract
Abstract
As one of the key technologies in image processing, image segmentation has received more and more attention since the 1970s .Image segmentation is the first step in image analysis, the foundation of computer vision and the important part in image understanding. This paper implements an image segmentation algorithm based on biological vision system and uses two evolutionary algorithms to optimize the parameters of the model.
The cognition and recognition of many landscapes and images are extremely intuitionistic and simply for biology, but it is a difficult task for the computer. In this paper, a novel image segmentation algorithm based on the biological vision system is implemented. In the pulse coupled neural network, each pixel in the image is excited not only by the brightness information of itself but the output of the pixels around it. Because of the variety of the image content, the parameters play a vital role in the segmentation process. We propose an objective function based on the histogram shape, clustering and entropy information fusion, uses two evolutionary strategies to optimize the parameters of the model, and chooses a set of most suitable parameters corresponding to each individual image in order to achieve the superior segmented result.
The experimental result indicates that, the image segmentation method based on the biological vision is more effective and efficient, and it also provides a new means for computer image understanding and cognition.
Keywords：Image segmentation Pulse coupled neural network Genetic algorithm
Particle swarm optimization algorithm