Replacing reduced quality images with original quality images in Word to Filtered HTML conversion

Question

I am creating a Word to HTML converter using Microsoft.Office.Interop.Word by saving each input Word document as Filtered HTML. But I am seeing a loss of image quality in the generated Filtered HTML as compared to the original embedded images. Following 4c860944-c627-4fed-afd4-a7473ab1dee6 , I am trying to automate the process of replacing the lower resolution images in the files folder for filtered web with a copy of the higher resolution original. I am saving the Word file as .zip to extract the original images for the replacement. However, I am seeing the following issues in this process :

1) Filtered HTML converts all my .tiff images to .png so that it can be viewed on browsers. Images obtained from zip extraction retains .tiff files in the same format. So browser compatibility is getting affected with this approach.

2) Filtered HTML does not save any image in the header and footer. But zip extraction does. So this is causing order of images to be affected when an image is present in the header or footer in Word.

3) Shapes, WordArt and other media apart from images are being saved as images in the filtered HTML folder. This is causing a mismatch in the numbering and ordering of images between the zip extracted images and the filtered HTML images.

What is the best way to obtain the original resolution images in the same order and name as the filtered HTML folder? Is there any other approach to automate this? Any help on this is appreciated.

Thanks