Tracking a rodent?

I have developed a reputation locally regarding mice in my house. When I first moved to Pennsylvania, I caught about 20 mice in 2 weeks inside my bedroom. Needless to say, I started to get aggressive about trapping them. It started with regular mouse traps, sticky traps, the kind that have little flaps that trap them side, wind up multiple-mouse catchers, and now an electronic one.

But, I'm not satisfied. It seems you must find their regular paths to place the traps for them to find it. So, I was thinking about video surveillance.

USB port Cameras are very cheap, and I've read they are sensitive to infrared if you remove the infrared filter. So, is it viable to use one to detect if and where mice are running around in my bedroom?

Will the cheap cameras detect the infrared wavelength of a mouse's body heat? (I have doubts) But, if it does, I imagine it would work well as a tracker. Perhaps I could also use one of the parabolic or spherical mirrors to get a 360 of the room?

The idea would be to allow the computer to track and record any mice over the course of a night. Then the next day view where the most used paths were. No actual 3d tracking or anything complex like that. Perhaps just color an image where any motion was detected.

With that knowledge, I can place the traps where the mice will find them. What do you all think?

Joe D

Reply to
Smiley
Loading thread data ...

Very cool idea, let me know how it works out. I have this same issue in my garage.

Reply to
Shawn Brown

Sounds like an interesting project. Could you not just use the camera as is and look for motion in areas. Then after a certain amount of time indicate the area with lots of motion. Keep us informed.

Alan

Reply to
A.P.

I think this is your best bet. The cameras will not see the body heat of a mouse, but infrared illumination will allow them to "see" in the darl. Cameras that advertise low brightness, and cameras that are B&W, generally do not have an infrared cut off filter.

The filter is a piece of what appears to be light blue glass. It may be tough to remove.

set up the camera in such a way that the illumination levels are consistent, and look for regions of change from image to image. filter out smaller "particles" and larger "particles" , leaving only the ones you expet to be the size of a mouse in the image.

google morphology, blob detection, center of mass, image processing.

Reply to
Blueeyedpop

Heating seeking rodent destroyer. Add a high speed missile projector and you could have something. Won't have much house left though.

Reply to
Ralph

what about something like cmucam ?

formatting link
Just need to add an ir lense and pan and tilt servo's.

formatting link
guys may have a board(with camera) that is better suited.

allectronics had a small b&w camera with ir leds around the lense for = illumination

formatting link
formatting link
've got one , but haven't got around to using it yet.

Alex

Reply to
Alex Gibson

This project may have some ideas for you:

formatting link
See the "Examples" and take a look at the video - pretty impressive.

Cheers,

-Brian

Reply to
Brian Dean

Hello Joe, There was an off the shelf program I played with from download.com. It's called "Active WebCam" and it will detect motion with a webcam. The demo is free and the full version is $19. Matt Meerian

Reply to
matt

some comments.

#1 - popcorn [salted, buttered, and popped of course] removes the need to find their regular paths. I once caught a mouse in a large deep plastic bag hanging from a doorknob - the mouse went in to get the smelly food in the bottom, and couldn't climb back up the sides of the bag.

#2 - some of the little cams have a group of 6 or so IR LEDs which provide illumination in the dark.

Reply to
dan michaels

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.