Skip to content

A module which detects products on the rack and finds misplaced, understock and overstock based on the planogram

Notifications You must be signed in to change notification settings

kaushalg47/Product-management-AI

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Product-management-AI

A smart object detection for grocery stores and super markets which detectes if the products and under-stock, out of stock or misplaced.

Detection

Detection was done by fine tuning a pre-trained YOLOv8 model on a 110k SKU dataset, which gave it capabilities to achieve 98% accuracy on detecting any type of product

Classification

Classification was optimized by extracting the flattened feature vector using Inceptionv3 CNN model and comparing it with sorted feature vectors dictionary of the inventory.

Rack identification

the average of the maximum and minimum limit of the product will create a dictionary of the limits and the products will be added to this dictionary based on the center coordinates. which also saves the process of seperately trainig of rack identification

processed Image:

Processed Image

Author

Kaushal G / @kaushalg47

About

A module which detects products on the rack and finds misplaced, understock and overstock based on the planogram

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages