基于机器视觉的蘑菇单体检测定位算法及其边界描述

    Algorithm for locating individual mushroom and description of its contour using machine vision

    • 摘要: 介绍的算法是以单体蘑菇为研究对象,通过该算法的研究将为蘑菇采摘机器人图像处理方法的实现奠定理论基础。研究内容包括,蘑菇图像的数字特征;提取边界的算法;对蘑菇边界进行离散傅立叶变化。提出仅需利用蘑菇的边界信息求蘑菇形心坐标的新方法,而且傅立叶描述可以进行蘑菇边界的平移、旋转和缩放操作,具有很强的边界形状重建功能。最后通过对获取的蘑菇图像进行分析,表明该算法和边界描述是非常有效的。

       

      Abstract: An algorithm which was used to locate individual mushroom was put forward in this paper. By using the algorithm, it is helpful to complete image processing for mushroom picking robot, and it supplies the theoretic basis for image processing to locate mushrooms in the growing bed. The main contents include the followings: digital characteristics of mushroom image information; algorithm for extracting mushroom contour; discrete Fourier transform to mushroom contour. A new method to calculate mushroom centroid coordinates based on the boundary information was put forward. Fourier descriptors can be used to conduct various operations to mushroom contour, such as translation, rotation, and scale. Besides, Fourier inverse transform allows reconstruction of mushroom contour. Finally, by analyzing mushroom image, the algorithm for locating mushroom and Fourier description to describe mushroom contour are greatly effective.

       

    /

    返回文章
    返回