Plastic Bottle Caps Full Automated Optical Visual Inspection System
Inspection Standard
| Model | Number of camera | Inspect | Inspection content | Precision | Speed |
| KVIS | 1 set | Concave area | Black spot,stains (color different) | ≥0.2mm | 800-1000 Pcs/mins |
|
| 1 set | Top area | Black spot,stains (color different) | ≥0.2mm |
|
| 4 set | Side area | Black spot,stains (color different) | ≥0.2mm |
|
| Printing missing | ≥1.0mm |
|
| Injection incomplete | ≥0.5mm |
|
| Sealing not good | ≥1.0mm |
|
4 set |
Ring area | Black spot, stains (color different) | ≥0.2mm |
|
| Incomplete | ≥0.5mm |
|
Inspection principle
For beverage, pharmaceutical, food and other manufacturers, the
appearance defect of bottle caps is a common problem in the
manufacturing process of products. Due to the increasingly high
requirements for product packaging refinement, quality and
continuous mass production, traditional manual inspection is easy
to cause leakage. Inspection and false inspection lead to the
outflow of defective products and low efficiency. The market poses
new challenges to the original quality inspection of bottle caps.
The emergence of the AI-based fully automatic bottle cap visual
inspection system has solved this difficulty well and has become a
new choice for more and more industrial enterprises. The bottle cap
inspection system can not only ensure the quality of packaging, but
also detect the material of the bottle cap, avoiding the entry of
materials that do not meet food safety standards into the
production process, and ensuring the quality and safety of
beverages and foods.
Inspection details
Based on computer vision and pattern recognition theory, adopts
oscillating feeding method, equipped with AI deep learning system,
uses CCD industrial camera to obtain online images of bottle caps
in real time, and conducts multi-angle and multi-directional
inspections on both sides, inside and side of bottle caps Shooting,
determine whether it meets the requirements through image
positioning and analysis, select bottle caps with defects in
appearance, and realize online rejection of defective products to
improve industrial production technology.
Hardware platform
◆ The main control platform adopts standard industrial computer,
with vision acquisition board and signal control board.
◆ The industrial camera adopts high-precision mega-pixel industrial
cameras such as CCD and CCD.
◆ Camera lens: Industrial lenses such as fixed focus and adjustable
focus are used.
◆ Light source: optional white light, red light, blue light,
infrared; surface light source, strip light, coaxial light, ring
light.
◆ Algorithm design: use OpenCV and other design modules.
Test content
The detection position can be selected from the top surface, bottom
surface, and inner/outer wall circumference of the bottle cap.
◆ Cap positioning: presence or absence, slanted cap, screw cap,
double cap, cap height, round cap, reverse cap
◆ Cap mixing: color, printing (misplaced pattern, wrong printing),
code, embossing
◆ Cap direction: align with product packaging or label
◆ Appearance of bottle cap: Dirty, lack of material, black spot,
flash, wrinkle, extrusion
◆ Mixed bottle caps: According to the rapid modeling and variety
management of different products, count the detection results and
automatically eliminate waste products.
Company profile
KEYETECH has always been committed to the application of artificial
intelligence in the field of vision technology, replacing human
eyes and brain decision with machine vision and AI reasoning
calculations, and integrating the quality detection and sorting to
industrial products in traditional manufacturing to make the
production more digitization , Intelligence, visualization, and
persistence. Also greatly improves the efficiency and accuracy of
industrial detection, and enhances the intelligent level of the
manufacturing industry.
