基于图像的林木蓄积量计算方法研究与仿真

摘要

研究森林木材蓄积量准确计算问题.在大面积森林区域中树木种类比较多,不同种类的树木交织生长,差异化的林木在形态学特征和可识别传感信息区别较大,很难运用传感器对交叉分布的林木特征准确区分采集.传统方法多是基于小区域传感器网络对森林木材存储状况进行检测的,一旦区域较大,林区植被结构复杂,传感网络很难进行细化区分,计算准确性不高.提出了一种基于图像树木轮廓特征分类的森林木材蓄积量计算方法.提取图像中的森林不同树木区域的轮廓特征,建立轮廓特征分类模型,按照树木不同的种类对木材蓄积量进行计算,从而实现森林木材蓄积量计算.实验结果表明,这种算法能够有效提高森林木材蓄积量计算的准确率.%Research accurate recognition of forest timber volume. Tree species are complex in Torest and have great differences. This paper put forward a forest timber volume calculation algorithm based on contour features classification of trees images. First, the contour features of different trees were extraction from images. Then according to different types of trees, the wood volume was calculated. The experiment results show that the algorithm can improve the accuracy of identification.