Research On Automatic Detection And Manipulation Of Nanomaterials Inside An SEM

Carbon nanotubes(CNTs)are regarded as one of the most promising nanomaterials to replace silicon in order to break the bottleneck of Moore's law.The research of nano-devices based on carbon nanotubes is the research hotspots of the world.Accurate three-dimensional automated operation of carbon nanotubes is extremely important for the efficient assembly of nano-devices.Nano-manipulated non-contact methods,such as dielectrophoresis,allow for two-dimensional manipulation of nanomaterials,but it is difficult to handle the flexibility of 3D operations.However,the precise identification and localization of carbon nanotubes alone is a key problem in the fabrication of nanodevices.Based on the analysis of nano-manipulation technology at home and abroad,a nano-robot system based on scanning electron microscopy(SEM)is built.The single-rooted carbon nanotube automatic recognition and picking algorithm based on SEM image information feedback is researched,optimization and verified by experiments.The specific research contents are as follows:Firstly,an automatic binarization method based on contours of objects and a tip recognition algorithm of carbon nanotube based on gradient inversion are proposed for the three-dimensional automatic operation of nanomaterials.Two-dimensional contour information of carbon nanotubes and probes is extracted and To complete the carbon nanotube and probe the automatic identification and positioning;Secondly,the carbon nanotubes were separated from the probe when the carbon nanotubes overlapped with the probe,and then a contact detection algorithm based on the similarity of the curve was proposed to detect the contact between the carbon nanotubes and the probe.The contact types of carbon nanotubes and probes were studied and analyzed,and different solutions were proposed for different contact types.Finally,the feasibility and robustness of the algorithm are verified by experiments.The algorithms are optimized and accelerated so that the algorithm can realize the real-time processing of SEM images under the existing hardware conditions.The system and algorithm can be applied to the three-dimensional automatic operation of nanomaterials,which can improve the production efficiency of nano devices and shorten the production cycle of nanodevices.