ALBERT

Self-learning vision system based on artificial intelligence

Catalog 2019

Available April 2017. Contact us if you wish to become an integrator.

Overview

Models

Downloads

How it works

Application Examples

Accessories

Quote

Media gallery

ALBERT is a complete and independent unit for visual inspection,based on the most advanced artificial intelligence techniques.

ALBERT learns the characteristics of a product directly from the production line and autonomously assesses its quality. ALBERT is very simple to use and does not require complicated programming procedures by experienced users, so it is quickly ready to control new products with different characteristics.

No traditional machine vision system is able to analyze complex objects or products with high variability as simply as a human operator would: ALBERT, on the other hand, interprets the concept of "quality" just like the fastest and most trained of your quality control operators. ALBERT is able to adapt to the production requirements of the moment since its “severity” level can be increased or decreased at the touch of a button, thus loosening or tightening the product acceptance criteria.

Each time, ALBERT chooses autonomously which will be the features to monitor that best describe the quality of your products. At any time and with a simple click, ALBERT can learn how to sort a new product or adapt to changing production conditions.

KEY ADVANTAGES

Simple
Learns and assesses the quality of your products directly from the production line without complicated settings.
Intelligent
Independently decides whether a product can be accepted and can control it in a more strict or tolerant way according to different production requirements.
Self-learning
Quickly learns the characteristics of a new product under inspection.
Suitable to identify complex defects
Is capable to understand the quality of products even with complex features and high variability.
IP65 Rated

APPLICATIONS

Ideal to quality check a variety of baked goods such as croissants and cookies but also frozen products, chocolates and various other foods, even when presented in a disorderly manner and with different orientations.

Albert - Self-learning vision system based on artificial intelligence

Inspection of croissants

Model		ALBERT-01
Description		Self-Learning Vision system based on artificial intelligence
Application		In-line inspection
		Working Distance (mm)
Field of View	mm x mm	300 260 x 190 400 350 x 260 500 440 x 325 600 530 x 390
Minimum Working Distance	mm	100
Optics		8 mm f1.4-f16 with manual focus adjustment
Lighting system		LED diffuse strobe illuminator, 5700 K white
Line speed 1	m/s	≈ 1
Number of parts per second 2		20
LED indicators		Yes (STATUS and SEVERITY LEVEL)
N ° of storable images 3		≈ 800K
Ports
Input
Synchronization input		1, opto-isolated (on top of the unit)
Output
for tower light		2 lights, 1 siren (on top)
for ejector (s)		6, opto-isolated (on top)
Synchronization output		1, opto-isolated (on top)
Communication
Ethernet		2 (on top)
Wireless		Wi-Fi (802.11n)
USB 3.0		4 (front of the unti)
HDMI		1 (front)
DVI		1 (front)
Power Requirements
Voltage	V, DC	24 ± 5%
Maximum power consumption	W	150
Cable		CBPWALB01 length 5 m IP68 (included)
Mechanical specifications
Width W 4	mm	330
Length L	mm	311
Height H	mm	171
Weight	Kg	10
Material		AISI304 stainless steel, anodized aluminum, scratch resistant polycarbonate (Lexan Margard®)
Mounting		4X M6 holes (optional mounting accessories available)
Environment
Operating temperature	°C	10-40
Storage temperature	°C	0-50
Humidity		20-85% (with no condensation)
IP class		65
Installation		indoor use only

ALBERT’s Field of View at 400 mm working distance.

Notes

Approximate value. Higher speeds are possible. Please contact us to check compatibility with your production line.
Estimated value. The number of inspected parts per second may vary depending on their size and the speed of the line.
Estimated value based on 250 Kbytes images stored in 200 GB SSD memory.
Wireless antenna included.

Technical documents

Data table

XLS

ALBERT-01 Manual v1.0

PDF

Albert datasheet

PDF

DP24793-A.pdf

PDF

ALBERT-01.step

STEP

Software tools

ALBERT-01 software package

ZIP

Press & marketing content

ALBERT brochure EN

PDF

Smart vision and optical solutions for the food and beverage industry

PDF

Installation

ALBERT is extremely easy to install: just attach it to any mechanical fixture by means of the four threaded holes on top of the unit, making sure to respect the correct working distance from the conveyor belt.

Once connected to a 24V power supply, simply press the "ON" button and wait less than a minute for ALBERT to be ready for use.

The basic settings are extremely simple and fast: the process of adjusting the focus and identifying the product to be inspected is assisted by convenient software tools.

The interaction with ALBERT is possible both through the physical interfaces on the product or by connecting the unit to a tablet or industrial PC.

There are 4 M* holes M6 to mount ALBERT on the production line at the correct working distance. ALBERT can be configured using the buttons on the product or by tablet / PC.

Learning

The learning process is easily performed by presenting some products on your production line and activating ALBERT in "LEARN" mode during normal operation.

Unlike traditional vision systems, ALBERT autonomously learns the characteristics of your production in a few minutes: it is normally sufficient to present a few tens or a few hundreds of products during production to allow ALBERT to learn their characteristics without complicated settings.

ALBERT is able to tolerate up to 20% defective products during the learning phase, without affecting its ability to sort products correctly.ALBERT will be ready to check your production once the status bar is full.

Moreover, whenever the goods on your production line change or anytime you want to adjust your quality control process to new production parameters, you can just press the "LEARN" button and ALBERT will adjust itself accordingly.

Even during the learning phase, ALBERT continues to monitor production, quickly adapting to the new inspection criteria without having to stop the line: no other vision system is so flexible and easy to configure.

ALBERT in ''LEARN'' mode

Sorting

Once the learning process is complete, ALBERT is ready for the sorting phase or "CHECK" phase: the products deemed inconsistent with the desired level of quality are reported via an integrated light bar and can be rejected from the line by interfacing ALBERT with the most common ejection systems thanks to the preinstalled opto-isolated outputs.

ALBERT is able to store images of defective products also keeping track of the reasons for rejection: this data can then be analyzed to improve the production process.

You can also adjust the "severity" level of the control parameters without having to stop the line: a dedicated slider bar allows the user to loosen or tighten the sorting criteria, quickly and easily adjusting ALBERT to new quality parameters.

ALBERT set to low (left) or high (right) severity level.

Interface

ALBERT communicates its status through a LED bar that turns red when defective products are detected.

ALBERT is also preset to be interfaced with an industrial tower light already installed on your production line and reports defective products through appropriate output signals that can trigger up to six eject stations.

If you wish to view the images that ALBERT is acquiring, you can do so wirelessly through an industrial tablet PC without losing IP65 rating or by connecting ALBERT to a monitor after removing the front protection panel. Connecting ALBERT to a monitor / tablet PC is also required to adjust the basic settings and to monitor rejection statistics on an external screen.

ALBERT interfaces.

ALBERT is designed to also control products characterized by a high degree of variability and impossible to parameterize through a traditional vision system, specifically in the food industry, but not exclusively. The most typical areas of use are the inspection of baked goods, frozen products, sweets, fish or meat. ALBERT is also ideal for products that are presented in a disorderly manner or with different orientations (provided there’s some spacing between them), or whose packaging cannot be represented by a predetermined pattern. In all of these cases, ALBERT makes it possible to control the products avoiding excess scrap or continued assistance by operators experienced in programming.

ALBERT is suitable for use on food lines thanks to the IP65 protection, the adoption of materials compatible with the food industry and the engineering solutions adopted.

Bakery products with variations in color, shape or other attributes

ALBERT is the ideal inspection solution for production lines of bakery products, such as biscuits, where traditional vision systems fail because product acceptance is not determined by a single parameter but is rather a combination of multiple subjective variables.

Works with a variety of conveyor belts including mesh or white

Inspection of cookies

ALBERT learns to know your production

Due to continuous and genuine changes in products such as chocolate or shortcrust pastry, no traditional on-line vision system is able to quickly learn and properly inspect this type of products like ALBERT does.

In fact, ALBERT can learn the natural change in color of the ingredients of a new batch in less than 5 minutes without the need to adjust complicated parameters each time.

Zoomed in image

Inspection of shortcrust pastry

Frozen products with variations in color, shape or ingredients

The acceptance criterion for frozen products is often a complex combination of many parameters. Unlike traditional vision systems, ALBERT is flexible and quickly learns the characteristics of products such as frozen pizzas, semi-finished meat or fish products, allowing you to loosen or tighten the sorting parameters by simply adjusting a dedicated slider bar on the main interface.

Frozen food

RT-KWP5170	Stainless steel PC Quad Core processor with fanless cooling system full IP65,1 x GbE, 2 x USB 2.0, 1 x RS-232
CBETH001	Ethernet cable for Panel PC, ODU plug S11KFC-PD8MCD0-xxxx + RJ45 connector with F/F RJ45 inline coupler, length 5m, IP65
RT-JD-0700GB-2	USB Keyboard/Mouse desk set
RT-UT10	UniqTablet Elcom UT10, Intel Atom, Touch screen, IP54
CMHOALB01	Support plate
RT-DRP-240-24	DIN rail power supply 240V ac - 24V dc 240 W
RT-70261132	Power cord with schuko plug - open end cable,3 m 10A 250V, single-phase
RT-DRP-480-24	DIN rail power supply 240V ac - 24V dc 480 W
RT-DRT-240-24	DIN rail power supply 400V ac three phase - 24V dc 240 W
RT-DRT-480-24	DIN rail power supply 400V ac three phase - 24V dc 480 W
RT-WTB9-3P246	Background suppression sick photoelectric sensor 20 - 350 mm detection range,PNP output, block style
RT-69942075	LED signal tower with buzzer, 2 light elements, clear, green/red (LED colour), 24 V ac/dc
LTBP288036-W	High power strobed LED backlight, 288 x 36 mm lighting area, white
LTBP048216-W	High power strobed LED backlight, 48 x 216 mm lighting area, white
LTDV6CH	Strobe controller 6 channels
LT3ZPZ400-00-X-W-24V	LED bar light, 400X19.7 mm illumination area, white, 24V
LTICGR1000-D1-PS-EU	Analogue lighting controller unit, 1 channel, 24V, 2A, constant mode, EU power cord, power adaptor 24V plug
RT-EXT-24V-F-3M	Illumination cable, side A flying leads, side B SM 3 way female connector, 24V - 3m
CBETH002	Ethernet cable, ODU plug S11KFC-PD8MCD0-xxxx + RJ45 connector, length 5m, IP65
CBGPO001	Output signals cable, ODU plug S11K0C-P14MCC0-xxxx with flying leads, length 5m, IP65
CBPH001	Photoelectric sensor cable, ODU plug S11K0C-P04MJG0-xxxx + M12 connector, length 5m, IP65
CBPH002	Photoelectric sensor cable, ODU plug S11K0C-P04MJG0-xxxx with flying leads, length 5m, IP65
CBTL001	Tower light cable, ODU plug S11KAC-P04MJG0-xxxx + M12 connector, length 5m, IP65
CBTL002	Tower light cable, ODU plug S11KAC-P04MJG0-xxxx with flying leads, length 5m, IP65
CBPWALB01	Power cable , ODU plug S11K0C-P03LPN0-xxxx with flying leads, length 5 m, IP65
RT-10060911	Set of 2 8” x 10 “ white balance/exposure cards - 18% grey and 90% white for color calibration

Request a quote

Simply fill in the form below to receive a quote or to request further info about this product: a sales engineer will get back to you within 24 hours.

* Please note: all fields are required.

First name

Last name

Job description

Company

Address

City

ZIP code

Phone

Country

State (US)

Your message

Copy the Opto Security Number below
(4 characters)

* After submit we'll ask you to check your consense to use Your data

E-mail

Please, insert a valid E-mail and accept conditions.

: With this you authorize the treatment of your personal data and give us a consense to contact you in order to work on your commercial requests and to be able to send you our best offers to cover your needs. For more details, please, read here(MANDATORY)
: With this you authorize the treatment of your personal data and give us a consense to contact you and send you email newsletters about products, sales, events, company announcements and to be able to send you our best offers to cover your needs. Newsletter are sent (roughly) twice a month. We will NOT spam or abuse your email in any way. For more details, please, read here(OPTIONAL)
: With this you authorize the treatment of your personal data and give us a consense to contact you and send you our product catalog by traditional mail (every two years). For more details, please, read here(OPTIONAL)