Automating with
Machine Vision Systems

Any industrial facility with a repetitive process can benefit from introducing vision. Machine vision has heavy presence in the automotive, plastics, food and packaging, medical device, and electronics industries, among others.

Vision systems are extremely customizable, and as such, can be used in any number of creative ways. Major machine vision applications are:

Appearance Inspection
Ensure a product is free of scratches, burrs, dents, short shot, flash, pinholes, or discoloration, even on complicated or low-contrast parts.

Presence/Absence (Counting)
Confirm that all components of an assembly or package are present and of the proper type, and count features like holes or slots on a single part.

Positioning
Verify proper position, orientation, and seating of many components, guide an industrial robot arm to correctly pick and place targets, or confirm the marking position of a KEYENCE laser marker.

Measurement
Measure in the X, Y, and Z planes to accuracies as small as ±10 μm.

Code Reading
Combine KEYENCE machine vision and inkjet printers to easily print and confirm date and lot codes with the help of an OCR (optical character recognition) built-in library, then compare to the read data of a bar code by simply checking a box.

KEYENCE's machine vision lineup consists of area camera, line scan, and 3D imaging technologies:

Area cameras are the most commonly used, and will solve the majority of vision applications.

Line scan cameras build 2D images one line at a time, sort of like a scanner, and are useful when images of large products, cylinders, or continuous webs are needed. Either the camera or part must be in motion during capture.

3D cameras capture industry-defining images of small parts with variable surface conditions. Simultaneous 2D and 3D capture allows for stable measurement in three planes.

Machine vision systems are made up of cameras, lenses, lighting, and image processing units. Each component is selected according to the application:

Camera
KEYENCE CMOS image sensors convert light into digital image data, to be sent to the controller in as little as 2 ms.

Lens
Lenses are used to focus the light onto the camera's image sensor. KEYENCE's high-resolution, low-distortion lens options ensure stable inspection across the whole image.

Light
Light selection is critical to any machine vision configuration; what the camera can't see, a machine vision system cannot inspect. The shape, size, and color of lighting, as well as installation distance and angle, can all be optimized to emphasize the features being inspected, and prevent any influences by the surrounding environment.

Image Processing Unit
Image processing units, or controllers, use pre-defined algorithms to process the image data and extract the important information.

Line scan cameras are used to capture images of moving parts one line at a time. Individual lines are stacked one atop the other in an image processing unit to ultimately build a 2D image. This style of image capture makes line scan particularly well-suited to inspections of continuous webs, cylinders, and large parts. Read on to find out why, or click here for more information:

Continuous Webs
Line scan cameras capture lines at either a specified rate, or according to an encoder's output. "Continuous" capture can be maintained without interruption, so that from one image to the next, 100% of a continuous web has been captured and inspected.

Cylinders
With a traditional area camera setup, a minimum of three images must be taken to see the full 360° surface around a cylinder. Additionally, due to the curved surface, part features may be distorted or difficult to evenly light. With a line scan camera, the entire cylinder is captured in one "unwrapped" image, free of any disfigurement.

Large Parts
Since images are built in pieces, high-resolution images can be captured and transferred more quickly than an area camera of similar size. For example, a KEYENCE line scan camera can capture images up to 67 MP at speeds of 6.1 μs/line. Additionally, lighting is much easier. Only one line across a large part needs to be illuminated at once, rather than the entire target.

3D cameras are used to inspect features which can't be seen with a traditional camera, or require manual inspection. What differentiates them from other technologies is their use of height data to create contrast, instead of color, so that precise inspection can be completed even on low-contrast parts. KEYENCE's 3D cameras simultaneously capture 2D image data, without the need for external lighting, so that measurement can be performed in all three planes; X, Y, and Z. Go to our product pages to learn more.

Pattern Projection lighting – 2D & 3D inspection

XT Series – 2D & 3D Measurement

RB Series – 3D Vison-Guided Robotics