Develop prototype for image recognition and AI based solution

We have a prototype project is already in ( the web, IOS and Android) for (xxxxxxxx) which is a platform using AI and image recognition technology to help finds things like objects, currently, we made a prototype that is based on 10 products if we uploaded the image or show an image to the platform ( using camera ) will give us the location by showing text “ aisle and bin “ from an API ( either on the web, IOS or Andriod ) So, what we are going to do on phase 2 is as the following: 1- We are going to integrate a smart digital shelve screen with a platform see below how to look like [login to view URL] 2- the request from the developer to achieve the scenario is as the following : User opens the app upload image or showing pictures, type text will give the location text same as we mentioned above “ aisle, bin “ but what we need to add : - - - adding a new feature to offer the exact location by using the smart shelve so the scenario will be that asked if I need help to find the location as soon as click on this option will give me a random icon anything banana, square whatever or special unique code number xxxx and tell me to go to that “ aisle and bin” which is already coming from the API and send this code or icon to specific smart shelve to display it there, so you will see this icon display on the location so at this time the platform will send a request to the digital display based on time slots that icon and make it flash or pop up for couple minutes to bring attention to the user then cleared out and so one. so can send many requests as possible should not conflict between requests, so each request and icon is unique. 3- a new option for the user to create a digital shopping list with this option the user should be able to create their own list adding items from the products that we added already this list can add with the same procedure when trying to find things either typing text and the suggesting the product or find it by image so if location needed then following step #2 regarding finding this item or the object. Users can finish the shopping list by scanning each item either using a scanner or QR scan and will give total and option to pay. We need an estimation plus time to update this across the platform -web - android -IOS