Overview
Our Pharma Division offers innovative and flexible custom designed inspection machines for IV bags.
This machine can perform two different types of visual inspection: in manual mode (with an operator) or in automatic mode (by cameras supported by Artificial Intelligence).
Artificial Intelligence is the core of our machine: it allows to obtain an extremely accurate inspection of particle and cosmetic defects, and thus to overcome many past issues and/or false rejects (such as air bubbles, silk-screening on containers, aesthetic defects of the containers that are considered non-relevant).
The extremely compact design, combined with the handling flexibility provided by a robot, makes ROBOX a unique modular machine.
Main features
Automatic version:
⦁ Robotic automatic bag loading system
⦁ First station for automatic inspection (2 cameras)
⦁ Rear light illuminator
⦁ Second station for automatic inspection (2 cameras)
More features
⦁ Radiant light illuminator
⦁ Patented inspection software supported by Deep Learning based on Artificial Intelligence
⦁ Robotic automatic rejection
⦁ Robotic automatic bag unloading function
Automatic version:
⦁ Robotic automatic bag loading system
⦁ Station for human operator visual inspection
⦁ Rear light illuminator
⦁ Operators can safely touch the bags during inspection
More features
⦁ Option for robotic automatic rejection
⦁ Robotic automatic bag unloading function
Machine Output
15 to 300 BpM
Compatible containers
Case History
Deep Learning
and Artificial Intelligence
A novel inspection machine has been designed and developed with the use of a specifically designed cellular neural network (CNN) coupled with an off-the-shelf neural network trainable solution. The novel machine, thanks to the computational versatility of the CNN, is capable of reaching high standards of assessment drastically decreasing the risk of operator-based errors during production.
The dedicated CNN developed by PBL is composed of two parts:
⦁ a first network that has the scope to perform a defined series of image processing steps in order to enhance the features of the objects that have to be detected.
⦁ a second network that is responsible for the feature extraction of the different defect classes that the network is trained to detect
With this dedicated configuration, PBL was able to developed a customized Artificial Intelligence solution for all the different types of inspection machines of its customers.
In addition, it is important to highlight that the neural networks developed by PBL can be trained off-the-shelf, thus ensuring a stable software solution that does not evolve in time, unless the user decides to modify the software.
Artificial Intelligence Technology
Thanks to our proprietary AI-technology we are able to offer the unmatched ability to recognize, in real time, the presence of particle defects and leaks in both rigids and flexible containers.
Contact us
