Question about 2nd monitor area

[atraver1]

Hello greg,

i was just wondering if it was possible to create a 2nd monitoring area and track both?

[gregtinkers]

It is certainly possible, though it may tax the power of the pi. If you wanted to, you could move the image processing into a method and each areas coordinates to the method where the original image would be cropped and processed. Give it a try!

[atraver1]

I'm sorry but that is sort of confusing to me. Is the image processing the draw_rectangle function? Or is it the whole getting the image reference and image comparing contours. I don't want to ask you to do it for me but could you show me an example of that. I am just lost in this whole process. I am still a little new to programming.

…

On Fri, Mar 3, 2017 at 12:44 PM, Greg Barbu ***@***.***> wrote: It is certainly possible, though it may tax the power of the pi. If you wanted to, you could move the image processing into a method and each areas coordinates to the method where the original image would be cropped and processed. Give it a try! — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#8 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AY9HGL-wYZAXmK9Ma88RxL5DuRLU3IoEks5riFGNgaJpZM4MST3N> .

[gregtinkers]

The program monitors the area defined by the draw rectangle function. Basically, it saves a portion of the frame without a vehicle and then compares it over an over to that same portion in the current frame. When the comparison shows a difference, it begins the speed calculation using a contour to define the change (a box drawn around the area that is different.) The part of the frame that it is using is defined by the rectangle that you draw. To monitor two rectangles you would need to first draw both on the frame, remember what each area looks like without a car, then repeat the process of comparing those original images to the corresponding area on the current frame.

This post on pyimagesearch may help you understand the principle of detection.

Since you are new to programming, I'd suggest writing a program with a very basic motion detection so that you understand the principle. Carspeed uses the basic principle, but restricts the processing to the area in the rectangle. When it detects movement, it uses the change in location of the contour between frames to compute the speed.