how to recognize multiple objects at the same time and the number of times it occurs

Question

I am extremely new to this technology so i don't know where to start. i need this camera to look at a sheet of paper with shapes on it (rectangle, square, circle, square) and tell me the number of each shape next to the actual shape itself. each shape is 3 cm wide or in diameter and it will be on a small sheet (half the size of computer paper)

For example, if two black circles, one black triangle, four black lines, and two black squares are displayed on the sheet, the output would be:

2 (RED CIRCLE)

1 (RED RECTANGLE)

4 (RED LINES)

2 (RED SQUARE)

sorry i had to put the name of the shape instead of the actual shape but i need the actual shape to be displayed and the output has to be all RED (the shape and the number)

if the red feature is not possible AT ALL then please help me on how to just display the number of each shape and the shape itself

can somebody give me step by step instructions on how to do it. BTW i don't have the camera with me and i don't have the software so if possible try to explain in laymans terms. i've made raspberry pi drones so i have some programming knowledge but i have never dipped my toe in object recognition 

Thank you anybody for even looking at this post and bigger thanks to anybody that wants to help. I am truly grateful for any help   

Answer 1

How about this one: https://www.pyimagesearch.com/2016/02/08/opencv-shape-detection/

or you can search the web for "opencv shape detection" and find other hits.

To port this to JeVois, you would basically eliminate all code related to video capture and to display, and you instead just focus on the core algorithm (since jevois handles capture and video output for you).

This tutorial will show you the steps (though it is a bit old now and predates JeVois Inventor; editing the code is much easier now in the inventor):

http://jevois.org/tutorials/ProgrammerPythonDice.html