从上图中可以看出，本文提出的改进Retinex算法对三维场景对象的提取

从上图中可以看出，本文提出的改进Retinex算法对三维场景对象的提取误差更小。最后，采用本文提出的方法对三维场景对象进行可见性判定，统计其判定误差，结果如下：

結果 (英文) 1:

As can be seen from the diagram, improvement of the Retinex algorithm presented the scene object extraction error smaller. Finally, to scene objects using this method for visibility determination, statistical error in its determination, the result is as follows:

結果 (英文) 2:

As can be seen from the figure above, the proposed improvements Retinex algorithm for extracting three-dimensional scene object error is smaller. Finally, the proposed method for three-dimensional scene object visibility determination, statistical error of its determination, the results are as follows:

結果 (英文) 3:

Can be seen from the diagram, the improved Retinex algorithm is proposed in this paper to extract the 3D scene objects smaller error. Finally, using the method proposed in this paper for 3D scene object visibility judgment, the judgment error statistics, the results are as follows: