LAUSR

In order to realize the three-dimensional reconstruction of the cerebellum structure of alligator sinensis, it is necessary to observe the structure by using a microscope. In the process of slice making and micro-image shooting, individual operation differences lead to good or bad quality of micro-images. To solve this problem, the feature linking model (FLM) is used to enhance the micro-image of alligator cerebellum and improve the quality of micro-image. The transverse section and longitudinal section were prepared by using the tissue of cerebellum of alligator sinensis as a raw material, and the micro-images were obtained by the light microscopy. The different enhancement methods were used to process the transverse section micro-images, and the enhancement was evaluated subjectively by direct observation. It was found that histogram equalization (HEQ) and fuzzy set method (FSM) did not enhance the darker parts well, while enhanced the brighter parts excessively. And the enhancement with pulse coupled neural network (PCNN), spiking cortical model (SCM), and FLM are better. On the basis of subjective evaluation, three objective evaluation indexes (contrast, spatial frequency, and gradient) were used to compare and analyze the image quality improvement of PCNN, SCM, and FLM, and we found that FLM had the best enhancement on transverse section micro-images. Then, FLM was used to process longitudinal section micro-images, and the objective evaluation indexes before and after processing were compared. We found FLM also had a good enhancement on longitudinal section micro-images. The results show that FLM was effective in enhancing micro-images of the cerebellum of alligator sinensis, and FLM improved the quality of the longitudinal section micro-images more obviously. The proposed method can improve the quality of the microscopic image of the cerebellum of alligator sinensis so that higher quality images can be obtained, which is beneficial to improving the utilization rate of the material.