The four main components of a machine vision system are the lens and lighting, the image sensor or camera, the processor and the method for communicating the results, whether by physical input/output (I/O) connection or another communication method.
Machine vision can use color pixel scanning and often uses a much larger array of pixels. Software tools apply to the captured images to determine the dimensions, position of the edges, movement, and relative position of the elements relative to each other.
The machine vision lenses capture the image and present it to the sensor in the form of light. To optimize the machine vision system, the camera must paire with suitable lenses. Although there are many types of lenses, machine vision applications typically use fixed focal length lenses.
Three factors are important when choosing:
- field of vision,
- working distance,
- camera sensor size.
There are many different ways to apply lighting to an image. The direction in which light arrives, its brightness, and its color, or wavelength, compared to the color of the target, are very important factors to consider when designing a machine vision environment. While lighting is an important part of getting a good picture, there are two other factors that affect how much light exposure a picture gets.
The machine lens includes a setting called an aperture that opens or closes to allow light to enter the lens. Combined with the exposure time, this determines the amount of light incident on the pixel array before the lighting applies at all. The shutter speed or exposure time determines how long the image projects on the pixel array. In machine vision, the shutter controlles electronically, usually in the milliseconds.
After capturing the image, software tools are applied. Some apply before analysis (pre-processing), while others determine the properties of the object under study. In the preprocessing phase, effects can be applied to the image to sharpen the edges, increase the contrast, or fill the spaces. The purpose of these tasks is to increase the capabilities of other software tools.
The goal of machine vision
Here are some common tools you can use to get goal information:
- Pixel Count: Specifies the number of light or dark pixels in the object.
- Edge Detection: Find the edges of an object.
- Measurement / metrology: Measuring the dimensions of an object (for example, in pixels, inches, or millimeters).
- Pattern recognition or template matching: Finding, matching, or counting specific patterns. This may involve locating an object that may be rotated, partially obscured by another object, or have different ones
- Optical Character Recognition (OCR): Automated reading of texts, such as serial numbers.
- Bar Code Reading, Data Matrix and “2D Bar Code”: Acquisition of data contained in various bar coding standards.
- Blocks: Checks the image for the presence of discrete spots of interconnected pixels as a landmark of the image.
- Color analysis: Identification of parts, products and items by color, quality assessment and isolation of elements by color.
As indicated by DZOptics, the purpose of obtaining control data is often to use it for comparison against target values to determine a “pass / fail” or “continue / discontinue” result. For example, when verifying a code or barcode, the retrieved value can compare to the stored target value. In the case of measurement, the measured value compares with the correct values and tolerances.
When verifying an alphanumeric code, the OCR text value compares to the correct or target value. For surface defect inspection, the measured defect size compares with the maximum size allowed by the quality standards.
Communication in machine vision
After obtaining information using the processor and software tools, this information can pass to the control system via many standard industry communication protocols. Major computer vision systems often support EtherNet / IP, Profinet and Modbus TCP protocols. RS232 and RS485 serial protocols are also common. Digital I / O often works into the systems for startup and simple reporting of results.
Computer vision communication standards are also available. Understanding the physical principles and capabilities of computer vision systems can be helpful in determining whether an application is suitable for camera-based systems.
In general, whatever the human eye can see, the camera can see (sometimes more, sometimes less), but decoding and transmitting this information can be complicated. Working with a supplier with experience with these systems, lighting and methods can save you a lot of time and money in the long run.