Hi,
I’ve change the lens of the pixy to suit my application better and realized that now, the xloc of the object reaches its max when the object is around 3/4 of the video. After that it goes again from 1. which means that the object can be in two separate places and the algorithm reads the same location.
Is there anywhere in the code which defines the usable part of the sensor (perhaps I need to change that to better define my new FOV and focal length?
Many thanks,
Khris
Hello Khris,
Does the new lens cause the active video to be smaller than the video window in PixyMon?
Pixy’s firmware doesn’t make any assumptions about the lens optics. Pixy’s firmware only knows the width and height of the CMOS image sensor: all of the object block information including width, height, x and y locations are in pixel units with respect to the image sensor. If a different lens “crops” the image or results in a narrower field-of-view (zoom), Pixy will still return pixel units.
Can you send a screenshot of PixyMon?
Edward
Hi Edward,
Since I posted i tried changing back to the original lens and that made no difference. With the new lens (!!Image 1) I don’t seem to be getting any crop.
Image 2, 3 and 4 shows the output from reading position via UART at different positions. As you can see it seems to overflow x_loc at around 2/3 of the screen.
Hi Edward,
I’ve connected via i2c and it works fine on i2c. Might be a problem with the UART interface? I am running the latest version of the pixy firmware and piximon.
Thanks,
Khris
Hello Khris,
I see you are using the IR-lock version of Pixy. You will need to get support from IR-lock:
Edward
Hi Edward,
I am not using irlock’s firmware, I simply called my signature IR_lock.