An example chip photo(black & white image)

I assume the elongated white blobs are wires?

If so, I would calculate the area of each white blob, and threshold it in order to remove the small white stuff.

Then do a distance transform from black. Any blob that contains a 'too high' pixelvalue in the distance transform is not a thin elongated structure. You might even detect the synapse-like endings this way...

Thanks for your response.

You are right (blobs long and thin with regular width are wires)

I think the area calculation may be one of the intelligent filtering of the irregular shapes. But, I don't understand what you mean by distance transform from black and 'too high' pixel value. Would you be more specific on it?

Actually wire width is sometimes known, but not always. We may have wires with different widths.

Thanks in advance.

The area threshold was to remove small things.

You want to remove the thick things too. Area won't help you because long wires also have a large area...

With a distance transform you calculate for each pixel how far it is from a (in your case) black pixel.

I included an example, which is in fact square Manhattan distance.

For the synapse-like endings of the wires, you allow a certain distance. Any value higher than that in the distance-transform, belongs to a thick blob in the original image that you want to remove.

BTW it will help if you tag your question with a couple of topics, e.g. 'image processing' and 'image segmentation'.

Thank you so much for the response and suggestions

You are welcome!

Somehow the image I included disappeared, so here is again a distance transform of your image:

I think the distance measure would be one of the ways to remove think blobs from the layout picture. Thanks for your idea.