Enhancement of Micro-Images Using Feature Linking Model for Cerebellum of Alligator sinensis

Bei Li,Zaiqun Liu,Feng Hu,Chengyan Ke

doi:10.1109/access.2019.2925772

Abstract

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.

Highlights

With the rapid development of computer technology, biological micro-imaging technology and equipment have been constantly updated [1], and the establishment of threedimensional (3D) visualization models based on the anatomical structure of virtual human, animal and organ have become a hot topic of research [2]
The central nervous system of new-hatching Alligator sinensis is about 4.5cm in length, which can be divided into telencephalon, diencephalon, mesencephalon, cerebellum, medulla oblongata and spinal cord from front to back
Microsections of the cerebellum of Alligator sinensis were made from new-hatching Alligator sinensis samples, and microscopic images of the cerebellum were obtained by the automatic scanning microscope

Summary

Introduction

With the rapid development of computer technology, biological micro-imaging technology and equipment have been constantly updated [1], and the establishment of threedimensional (3D) visualization models based on the anatomical structure of virtual human, animal and organ have become a hot topic of research [2]. Studying the structure of the cerebellum of Alligator sinensis and constructing the three-dimensional visualization model can grasp the three-dimensional structure more clearly, which is of great significance to the research and teaching of Alligator sinensis. The 3D visualization model has certain reference value for people to study the rise and fall of ancient reptiles, paleogeology, and biological evolution. In the process of brain slicing of the alligator [5], there is a gap in the quality of the slices due to the individual differences of the operators. At the same time, when using a microscope to observe and shoot micro images, the quality of micro-images will be affected by artificial

Methods

Results

Conclusion