A fusion framework for automatic neuron reconstruction

Neuron reconstruction is an important step to study the relationship between brain neuron morphology and function. Existing neuron reconstruction algorithms have achieved relatively good results under certain conditions. In the context of whole brain research, the conditions are more demanding, so the previous methods cannot guarantee high-quality results in any brain image data. This paper proposes a fusion framework for automatic reconstruction, which is dedicated to improving the applicability of traditional reconstruction algorithms and obtaining high-quality neuron morphological data. Experiments show that this framework is more flexible than traditional reconstruction methods and has broad application prospects in the current research background.