欢迎访问林业科学,今天是

林业科学 ›› 2004, Vol. 40 ›› Issue (4): 145-147.doi: 10.11707/j.1001-7488.20040425

• 论文及研究报告 • 上一篇    下一篇

基于分数布朗随机场与分形参数的原木漏节图像处理

戚大伟   

  1. 东北林业大学,哈尔滨150040
  • 收稿日期:2002-10-24 修回日期:1900-01-01 出版日期:2004-07-25 发布日期:2004-07-25

Log with Knot-Hole Image Processing Based on the Discrete Brownian Random Field and Fractal Parameter

Qi Dawei   

  1. Northeast Forestry University Harbin150040
  • Received:2002-10-24 Revised:1900-01-01 Online:2004-07-25 Published:2004-07-25

摘要:

本文提供了一种基于分形理论中分数布朗随机场模型和分形参数H值的X -射线原木漏节图像处理方法。分数布朗随机场模型是描述自然景物的有效方法,在图像区域的小范围内,灰度表面具有统计意义上的自相似性,但在不同图像区域的交界处,这种分形的规律性将会被破坏,在此求出的分形参数H值将会发生奇异,据此可以判断出该处为图像的边缘或交界处。从试验的结果可以看出这种方法对X -射线原木漏节图像非常有效,同时对计算机进行自动模式识别有重要意义

关键词: 原木, 漏节, 分数布朗随机场, 分形, 图像处理

Abstract:

A reliable method of X-ray image of log with knot-hole processing based on the fractal discrete Brownian random field model and fractal parameter H was presented in the paper. It is known that the fractal Brownian random field is valid to describe the image of nature scene. In only tiny region of the picture, the surface of gray levels is self_similar in statistics. But for the edges, the point located in the boundaries between adjacent regions, this regular properly will be lost. The fractal parameters of these points will be out of range. In this case we can finger out the edges of knot-hole by calculating the fractal parameter H. The experimental results showed the method was very successful for X-ray image of log with knot_hole. It was significant to computer pattern recognition automatically.

Key words: Log, Knot_hole, Discrete Brownian random field, Fractal, Image processing