Arduino capture still image with camera

Hello, I am extremely new to Arduino so please forgive me for my ignorance on the topic. I have been searching for information on using a camera with an Arduino. My goal is to capture a still image and interface with an image recognition API such as Google's Vision AI. Based on the image recognized, move a couple of servos to particular potions.

I have not found a way to capture an image using Arduino. Is this not possible? I have found other hardware such as the ESP32, but my understanding is that this is doesn't use Arduino hardware.

Can I capture an image with Arduino? If so, could you please kind enough to point me in the right direction of a tutorial or more information? Thanks.

I have been trying to get this one working How to Use OV7670 Camera Module with Arduino​ Uno it is reproduced on a number of maker web-sites.
My eyes are not up to the job of making the 'adaptor board' that reduces the Arduino 5v logic down to 3.3V.
I ordered an Arduino Due board on friday to see if I can bypass the voltage issue or does it just move it?
mick in glen innes 2370

ESP32 is different from an Uno (for example), but it can still use the Arduino IDE. So you can develop Arduino applications and upload to an ESP32. There is an ESP32 board, the ESP32-CAM that already comes with a camera attached.

It is possible to capture very small images, but few genuine Arduinos have enough memory to actually work with images. As mentioned above, there are some more powerful microprocessors, like the ESP32, that can be programmed using the Arduino IDE, but even they are quite limited for applications like machine vision.

If you are serious about machine learning and image processing, the Raspberry Pi is about the smallest micro where such applications are practical. A laptop or desktop running windows, mac OS or linux is much better.

Sure you can, try an Arducam OV2640.

Its 3.3V\5V and saves images to SD, code will fit on a UNO if you insist.

A DUE would be a better choice, far more memory, faster etc.

However capturing an image is one thing, how were you planning on getting the picture into the Google's Vision AI ?

Thank you for all of the suggestions. It sounds like the Raspberry Pi may be the better board to work with on this project for capturing an image. I will look into the suggestions though.

As far as getting the image to Vision AI (though I may use a different service), my plan was to use a WiFi shield and connect the Arduino wirelessly to make HTTP requests.

This may be of interest?

Interesting project, but it looks like it took a heroic effort to get the vehicle classifier to even run on the ESP32.

It does not appear to be practical for general applications.

This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.