论文部分内容阅读
为了完成基于多自由度的机械臂人脸素描,实现精确的空间点对点映射控制,提出素描绘画算法。该算法由改良型的神经网络算法和人物图像矢量化算法组成。该算法通过合理分解目标空间,严格选择采样点,在保证精度的情况下,具有实用性强,计算量小的特点,可适用于大量嵌入式平台。这保证了机械臂人脸素描的快速性和通用性。
In order to complete the human face sketch based on multi-degree-of-freedom manipulator and realize the accurate point-to-point mapping control of space, a sketch drawing algorithm is proposed. The algorithm consists of an improved neural network algorithm and human image vectorization algorithm. By reasonably decomposing the target space and selecting the sampling points strictly, the algorithm has the characteristics of high practicability and small computation amount, which is suitable for a large number of embedded platforms. This guarantees the rapidity and versatility of robotic face sketching.