（毕业论文）基于改进PCNN的图像分割和图像去噪.docVIP

摘 要 图像分割和图像去噪为图像识别和分析奠定了基础，也是计算机科学领域的关键步骤。作为图像预处理的基本技术，图像分割和图像去噪，一直是备受学者们关注的热点议题。PCNN模型是新兴的人工智能神经网络之一，广泛应用于图像处理和分析以及计算机视觉中。 本文以脉冲耦合神经网络（Pulse coupled neural network ，简称PCNN）为研究框架，结合螺旋优化算法对PCNN模型进行改进，并结合双边滤波模型在图像去噪方面的特点，提出基于改进PCNN的图像分割技术和基于改进PCNN和改进双边滤波模型的图像去噪技术。本文研究内容主要有以下几个方面： (1)简化PCNN模型。研究并分析基本PCNN模型的参数问题，通过对其分析，并结合图像自身所包含的信息，来对PCNN模型进行简化。 (2)基于改进PCNN的图像分割。深入研究PCNN模型，针对其参数取值单一，且要反复进行大量实验的问题，对某些关键参数，结合螺旋优化算法对其进行改进，并将其用于图像分割算法中。 (3)结合双边滤波模型和PCNN模型，给出新的图像去噪算法。探讨PCNN模型在图像去噪中的应用。通过分析PCNN模型针对脉冲噪声去除效果的优势，以及在去除其它噪声上不足的问题，引入双边滤波模型，并对该模型存在的问题进行改进，以弥补PCNN模型在去噪时的不足。通过与其它方法对比，证明了该算法的有效性。 关键词：图像去噪；图像分割；脉冲耦合神经网络；螺旋优化算法；双边滤波模型 Image Segmentation and Image De-noising Based on Improved PCNN Man Li (Control Science and Engineering) Directed by Professor Li Shurong Abstract Image de-noising and image segmentation have laid the foundation for the image recognition and analysis, but also are key steps in computer science. As a basic technology of image preprocessing, image de-noising and image segmentation have always been hot issues concerned by scholars. Pulse coupled neural network (PCNN) model is one of the emerging artificial intelligence neural networks, which is widely used in image processing, analysis, and computer vision. In this paper, we take the PCNN model as the research framework, and propose image segmentation technology based on improved PCNN model and image de-noising technology based on improved bilateral filtering model and PCNN model. The main contents of this paper are the following aspects. (1) A simplified PCNN model is proposed. Parameters of the basic PCNN model are researched and analyzed. Through the analysis on it, combining with the information contained in the image itself, the simplified PCNN model is given. (2) The improved PCNN model is put forward, which combines with spiral optimization algorithm. Parameters are manually selected by doing a lot of experiments repeatedly on some key parameters in image segmentation. (3) A new image de