【摘要】 将立体视觉应用于森林资源调查是一种全新的手段 ,能高效地对样地进行调查 ,并且为高分辨率遥感图像提供一种获取林分调查因子的技术 .该文从林业环境的实际情况出发 ,探索立体视觉技术应用于森林资源调查的可行技术路线 .针对树干边缘的特点 ,提出了Canny算子检测的树干边缘和外极线约束下、交互式的区域相关立体匹配策略 .采用在中国林业科学研究院院内树林中FUJIFILMFinePix 6 90 0Zoom数码相机拍摄的立体图像对 ,实现了立体视觉技术测量树木点位和胸径 ,并估计了在实验条件下的量测精度 .更多还原

【Abstract】 Application of stereo vision technology on forest resource inventory will be a novel means, by which the plot inventory can be efficiently carried out. In addition, it will provide a new technique of gathering stand inventory factors for high resolution remote sensing image. Based on the forest environmental facts, the practical technical routine of application of stereo vision technology on forest resource inventory is explored. To improve calibration precision, nonlinear error in measure model is adopted in this paper. Aimming at the features of tree edge, the interactive area correlation stereo matching, which is constrained by epipolar line and tree edge detected by Canny operator, is presented. With the stereo image pairs of forest in Chinese Academy of Forestry taken with FUJIFILM Fine Pix 6900 Zoom digital camera, the position and the DBH of trees were measured while the measurement precision at experimental conditions is evaluated.更多还原