基于小波域隐马模型的树木类图像分割算法

Multiresolution segmentation approach for tree image based on hidden Markov tree model

为了克服自然状态植物图像提取困难,尤其是树图像在自然生长状态下背景非常复杂(天空、建筑物、植物等)的问题。提出一种基于小波域隐马模型的树木类图像分割算法。该算法利用形态小波的优良特性,结合隐藏马尔可夫树模型,实现了基于图像纹理的树木类图像分割,并得到了较为理想的实验效果。

The extraction from plant image was difficult. Varying image characteristics caused by differing levels of foliation and fluctuating lighting conditions. This paper proposed a new segmentation approach, based on morphological wavelets and the hidden Markov tree model （HMM）. The wavelet transform is suited to images containing singularities, the HMM provides a good tool for distinguishing between different textures.