Enhanced computer vision with Raspberry Pi HQ camera

Last year I wrote about reading my district heating meter by means of computer vision. Already within that post I shared my observations about the fact, that the screen on the meter is a bit complex to read. This is due the “disconnected segments” and the “surrounded decimals”. This combined with my initial setup using Camera Module V2 ended up being too unstable.

After fixing the reading manually multiple times, I decided that this not the right way to continue. I have now replaced the camera with Raspberry Pi High Quality Camera combined with a 16mm lens.

Testing out the new camera

The new camera has better image quality, manual focus and shorter minimal focus distance. All of these are very welcome for my use case and the difference is really noticeable.

Old view on the left and new one on the right

With these improvements I have had more reliable results. And actually I was also able to simplify the image processing quite a bit. So all in all an excellent upgrade!

Preprocessed image. Digits are now nice and clear!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.