显著特征融合下的虚拟仿真实验室布局最优规划方法

The Optimal Planning Method for Virtual Simulation Laboratory Layout under the Fusion of Salient Features

为了科学、合理地规划虚拟仿真实验室布局,在有限的空间达到利用率最优,提出显著特征融合下的虚拟仿真实验室布局最优规划方法。将虚拟仿真实验室布局的三维模型转换成二值投影图像,分块处理实验室布局图像获取图像块,运用主成分分析(PCA)方法将图像块映射至高维空间后,提取其全局显著特征图,依据中心图像块和相邻的图像块的颜色差异提取图像局部显著特征图;通过多尺度多向变换分解方法融合全局和局部显著特征图,将显著特征融合结果作为支持向量机(SVM)的输入,完成虚拟仿真实验室布局识别,并利用遗传算法对布局好的实验室做最优规划。实验验证:该方法的显著特征融合结果细节信息丰富、全局效果清晰,可以科学地规划虚拟仿真实验室布局,避免穿墙、重叠、通行受阻或使用受限的情况。

In order to scientifically and reasonably plan the layout of virtual simulation laboratory and achieve the optimal utilization rate in the limited space,an optimal planning method for virtual simulation laboratory layout under the fusion of salient features was proposed.The three-dimensional model of the virtual simulation laboratory layout was transformed into a binary projection image,and the laboratory layout image was processed in blocks to obtain image blocks.After the image blocks were mapped to the high-dimensional space by using the principal component analysis method,the global salient feature map was extracted,and the local salient feature map of the image was extracted according to the color difference between the central image block and the adjacent image blocks;The global and local salient feature maps were fused by multi-scale and multi-directional transformation decomposition method,and the salient feature fusion results were used as the input of support vector machine to complete the layout recognition of virtual simulation laboratory,and the genetic algorithm was used to make the optimal planning for the well arranged laboratory.Experimental verification:the salient feature fusion result of this method has rich details and clear global effect.It can scientifically plan the layout of the virtual simulation laboratory,and avoid the situation of passing through the wall,overlapping,blocked access or limited use.