I have a raspberry pi pico and OV7670 cam. I want to connect them and get a camera output as a video, So I can use image recognition to identify objects (tennis balls) to collect them using a setup with motor drive+motor setup. Please help with your ideas and reference links. Thanks in advance.
1 Answer
You can use the RPI pico C sdk to write firmware code for reading image data pins of off ov7670. This explains how to setup the PCIO C SDK https://datasheets.raspberrypi.com/pico/getting-started-with-pico.pdf
Look at the ov7670 schematic here to figure out the ov7670 hookup/wiring to PICO http://web.mit.edu/6.111/www/f2016/tools/OV7670_2006.pdf
You'd realize that PICO's GPIO sampling rate is slower than ov7670 data rate. You might be able to do something to compensate for the data rate mismatch, maybe only read a few selective lines every frame. Might not work for video, but maybe stills, or video in grayscale on a lower resolution might work.
Let me know if you need more help!
-
I find out that pico can't read with ov7670 efficiently. So I barrow pi to do this work. Its a raspberry Pi 3 B model. The thing is I can't install opencv in the board. When I tried to download all package its struck on middle and when I try to download zip. its very slow and download for 13 to 15mb then Its retiring to download. How should resolve this. can you help me. Feb 27 at 4:09
-
YOCTO can be used for making custom Linux builds. Yocto uses meta layers making build builds for different targets / systems and adding different packages. There is is a meta layer for Raspberry Pi, you can look at it here github.com/agherzan/meta-raspberrypi Learn a little about YOCTO, and then use meta-rapberrypi to add opengl etc and other packages of your liking to your build.– ManzurFeb 28 at 1:26