Working on an image classification project (microscopic images) , have some doubts

Currently, I am working on an image classification project. The data set contains very high resolution images taken via an electron microscope. Hence, I have few and limited instances.

I have done EDA and made up a deep CNN to go about it. The results are not very satisfying. Even tweaking the model did not work. I got similar results in cross-validation as well. I also performed data augmentation, but I do not possess enough knowledge of it, can anyone provide any guidance?

Also, if I do ensembling in such a case would it be beneficial (given fewer instances)?

Also, I am planning to use a pre-trained model for performing image classification. But, after going through blogs, Kaggle topic discussions, and some research papers I am confused if any of them will work on a smaller number of instances?

Also, as you might have guessed by the type of my doubts, I am new and naive in this field. I have worked with numerical data but images are new for me. So any guidance will be appreciated.

Thanks in advance :)