Are items randomly spaced on the conveyor or is the position registered via fixed product locator?
Similar to your post #49.
Is there a space between items or is it possible that some items butt up to the next in random order?
Max.

 

Are items randomly spaced on the conveyor or is the position registered via fixed product locator?
Similar to your post #49.
Is there a space between items or is it possible that some items butt up to the next in random order?
Max.

items are randomly spaced on the conveyor and there is a space between items as shown in the diagram

 

items are randomly spaced on the conveyor and there is a space between items as shown in the diagram

So a simple photo cell detector could be used as a counter from detection position to rejection position.
Max.

 

So a simple photo cell detector could be used as a counter from detection position to rejection position.
Max.

Yes it can be possible but what happens if the speed of the conveyor belt will vary with time. we can't get exact position of object at variable speed but encoder can give the position of the object

 

Doesn't matter if a simple item counter is used.
I am talking for the rejection process.
Max.

 

Doesn't matter if a simple item counter is used.
I am talking for the rejection process.
Max.

I am totally agree with you but Why most of places encoder used, why they don't use sensor in place of encoder in rejection system ?

 

Not sure what are ;most' places as I have not found that the norm in material handling/rejection systems that I have come across and repaired.
Max.

 

From what I am seeing in the OP's system, it is not necessary to count the items, just the position of the defective item, this system will not be concerned if there is a 'missing' item.
Just the location slot or position on the conveyor.
More mechanical details of the conveyor and items is necessary for more informed decisions.
Max.

If the mechanical arrangement is not able to keep the items with a constant spacing then that spacing may change between the scan position and the eject position. Not all packages travel nicely. AND, at 500 items per minute they will really be flying by. But they do need to be presented to the camera in a manner that provides a uniform view of the bar code, so there will need to be attention given to that aspect. After viewing the exact position will be less critical, except for the reject position.

 

If the mechanical arrangement is not able to keep the items with a constant spacing then that spacing may change between the scan position and the eject position. .

Spacing may change but not the identified number status of the item, as soon as the reject item # reaches the ejector it is activated.
This is similar to the beer/soft drink canning where the level is detected by low level radio active sensor and the low container is ejected down the line which is moving quite fast.
There are systems that operate up to 2,000 cans/minute.
Going back to my sensor on the conveyor itself, it would be a simple solution as the conveyor position required to move the item to the reject position could be measured exactly by rotating proximity flag etc.
Which can also be termed an encoder, I just prefer, Rotary Sensor.
Max.

 

Back to updates

I have seen some camera's on internet used for barcode reading

1. congnex https://www.cognex.com/products/machine-vision/2d-machine-vision-systems/in-sight-8000-series
2. Flir vision camera
3. Bamur Vision Camera

any suggestion, which would be good

 

can any moderator merge my closed thread post here in this thread

https://forum.allaboutcircuits.com/threads/help-need-to-develop-program-flow-for-project.157372/

 

Product details

• The system should be used for barcode, batch code verification on the labels
• To check and verify OCR, Barcode, Pharmacode, 2D code on Label. These OCR contains the detail of the product like Batch no., M.R.P., Manufacturing date, expiry date etc.
• Track and Trace System to get the data of the product for Pharma product.

Here is something

It is not at all clear that you should use a camera at all. If you are reading bar codes, a holographic bar code scanner may well be a better choice. Cameras are good for arbitrary features and orientation sensing but barcode scanners are purpose built to pick up barcodes quickly and accurately.

 

It is not at all clear that you should use a camera at all. If you are reading bar codes, a holographic bar code scanner may well be a better choice. Cameras are good for arbitrary features and orientation sensing but barcode scanners are purpose built to pick up barcodes quickly and accurately.

I want to use a camera just for the experient purpose because my main to develop vision system to sort product ie. Sorting bottles, sorting cups and so on

vision camera can be sort things

 

I am looking to use camera because in future I will try to develop vision system to sort product ie. Sorting bottles, sorting cups and so on

vision camera can be sort things

Yes, but if you have a product possibility you are hobbling it immediately by choosing a more expensive and less effective component on the basis of future development. Make it modular and add machine vision in as an option if someone needs it. Bar codes don’t justify a machine vision system, why would the customer pay for it?

 

Yes, but if you have a product possibility you are hobbling it immediately by choosing a more expensive and less effective component on the basis of future development. Make it modular and add machine vision in as an option if someone needs it. Bar codes don’t justify a machine vision system, why would the customer pay for it?

I haven’t started yet to implement. Still working on planning

I have changed mind I am not focusing on barcode. I do not want to use barcode scanner to scan barcode I am more concern about other product that's why want to use vision camera for sorting products

 

I haven’t started yet to implement. Still working on planning

I do not want to use barcode scanner to scan barcode I am more concern about other product

Well, given that these sorts of systems have a long and complicated development history, I would start reading books on them to learn from the experiences of engineers who have already done this. Specifying a camera, basically at random on the advice of Internet "experts" (myself included) doesn't sound like a good business plan to me.

This isn't new development, this is reinventing the wheel. You are much better off reading the wisdom of wheel makers before deciding you know what you need. What might be a fruitful direction is to look at F/OSS options for both software (Python has a good vision library, for example) and hardware (which wouldn't be a commercial MV camera, rather something much less expensive) and perhaps innovate in the area of cost and value.

In any case, good luck with your project, I wish you and your future customers great success.

 

In any case, good luck with your project, I wish you and your future customers great success.

Let’s see what happen I am trying my best. I am reading articles and gathering information

 

This might do it for you.

DIY Artificial Intelligence
https://www.microcontrollertips.com/diy-artificial-intelligence-faq/?spMailingID=1956&puid=420371

Intel’s AI Academy has a vast repository of learning materials, including advice for free courses, a library of recorded training, technical case studies, and research papers. There are even practice input videos for testing image recognition. For example, we can train a model and then do practice runs with ready-made sample videos of bolts on a conveyor belt to use as inference input rather than purchasing a conveyor belt and video camera.