Visual Inspection


Visual Inspection, or Visual Testing (VT), is the oldest and most basic method of inspection. It is the process of looking over a piece of equipment using the naked eye to look for flaws. It requires no equipment except the naked eye of a trained inspector.

Manufacturing operations strive to deliver the highest quality during every stage of the production or assembly process. Over half of these quality checks involve visual confirmation to ensure the parts are in the correct locations, have the right shape or color or texture, and are free from any blemishes such as scratches, pinholes, foreign particles, etc. Automating these types of visual quality checks is very difficult because of the volume of inspections, product variety, and the possibility that defects may occur anywhere on the product and could be of any size.

When an inspection process is automated it is intended that this system simulates the most possible intelligence and human  experience  at  a  high  processing  speed.  To  achieve this  goal  the  various  solutions  are  built  around  cameras, computers and lighting systems, which all together make up the  so-called  machine  vision  system  or  automatic  visual inspection   system.

These   systems   use various  techniques of image processing to extract information with techniques supported by automatic decision  of  artificial  intelligence  area,  such  as  artificial neural networks.

These automatic visual inspection systems are based on image processing techniques.

The term image processing generally refers to processing of a two-dimensional picture by a digital computer. In a broader context, it implies digital processing of any two-dimensional data. A digital image is an array of real numbers represented by a finite number of bits. The principle advantage of Digital Image Processing methods is its versatility, repeatability and the preservation of original data precision. The various Image Processing techniques are:

  • Image preprocessing
  • Image enhancement
  • Image segmentation
  • Feature extraction
  • Image classification


Our goal in visual inspection is to:

  • Develop innovative systems, with a better usage of computing resources
  • Develop simpler systems, where ease of use is balanced with Quality of Service (QoS) and Quality of Experience (QoE)
  • Develop richer systems, as users are always asking for more computing power in less space


Our recipe
  • no limits in complexity when designing systems
  • timing and performances as primary targets
  • “Zero warning” strategies

CanEye and CoverEye are two our systems for specific visual inspection use case.

  • Check EVERY product in a high-speed metallic can assembly line for EACH of these defects:
  • Shape damages, deformations
  • Texture impurity
  • Missing reference points
  • Uncomplete glueing of external border

  • Image resolution: from XGA and more (up to 5 MPix)
  • Speed: more than 600 pcs. / min.
  • Bursts: 900/1000/1200 pcs. min. -> 20 fps
  • Hard realtime behavior with 40 ms. timeslots


Our solution
  • In-house written algorithm for analysis in realtime
  • x86 Core based architecture, tested also on FPGA and ARM



coverEye from dpcontrol on Vimeo.