I am on a quest to efficiently segment zoning map images using color information. For this task, I am looking for a cutting-edge model that can accomplish this using either deep learning or traditional techniques.
I welcome any insights, code snippets, or recommended resources that can set me on the right path for this zoning map segmentation project. Your valuable assistance is highly appreciated!
Please feel free to adapt and expand upon this question to align with your specific project requirements and the guidance you're seeking.
To provide some context, I initially ventured into k-means clustering, but I grappled with noise issues. Although post-processing techniques like morphological closing and opening helped alleviate noise problems, they unfortunately led to a loss of crucial zone information. I also explored alternative methodologies, including connected component analysis, contour-based estimation, and median filtering, in an attempt to address noise-related challenges.