IOT Based Object Counting Using Image Processing

Project Title

IOT Based Object Counting Using Image Processing

Project Area of Specialization Internet of ThingsProject Summary

Object counting is frequently encountered problem in production system operating at high speed, it is very difficult to perform measurement without using machine vision. In this project, a machine vision application is proposed, which can perform object-independent contactless product counting and defective products are sorted out .It is a low cost,very fast and high precision measurement system which will be frequently used in production lines. This system increases the instantaneous production of production facilities and also smart enough to identify the shapes of products placed in a crate. It deliver complete, exact and identical products to end users.

Project Objectives

1. To deliver complete, exact and identical products to end users.
2. Contactless Counting.
3. Error control i.e locate empty spaces in a pack.

Project Implementation Method

This Project is is software based as well as hardware baised. A conveyor is basically used in which products are passed over and faulty products are sorted out on this band. An image of the product through a camera integrated in this system and faulty products are determined.

Benefits of the Project

In  industry, inventory management is very tricky as products are pretty hard to track in real time because something is always add, remove or move. Artificial intelligence (AI)  systems can perform automatic counting and localization that will aid us in inventory accuracy. In this way right quantity of products at the best possible price, ensuring not a single panney is wasted on inaccurate or extraneous orders

Technical Details of Final Deliverable

First of all we will mount cameras so that we can able to get pictures from all directions and dimensions. Then we have to process that particular image. During image processing our 1st step is to  applied gussian filter to get noise free image then convert this image into gray image.After that we will go for otsu thresholding which will provide the image with just 2 pixels I.e foreground and background and then sobel  edge detection and Hough transformation will perform

Final Deliverable of the Project HW/SW integrated systemCore Industry ManufacturingOther IndustriesCore Technology Internet of Things (IoT)Other TechnologiesSustainable Development Goals Industry, Innovation and Infrastructure, Responsible Consumption and ProductionRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	19500
camera	Equipment	3	1500	4500
conveyor belt	Equipment	1	10000	10000
crate	Equipment	5	1000	5000