Understanding how machine vision works will help you identify whether machine vision solves specific manufacturing or processing problems. People are often unclear about what machine / computer vision can and cannot do for a production line or process. Understanding how it works can help people decide if it will solve the problems in a given application. So what exactly is computer vision and how does it actually work?
Machine vision means using one or more cameras to inspect and analyze objects in an automated manner, usually in an industrial or manufacturing environment. The obtained data can use for process control or production activities. A typical use may be an assembly line, where a camera works after an operation performed on a part, which takes and processes the image.
The camera can be programmed to check the position of a particular object, its color, size or shape, or whether an object is present or not. It can also search and decode standard 2D matrix barcodes or even read printed characters.
After checking the product, a signal is usually generated that determines what to do with the product next. The part can be discarded in a container, directed to a branch conveyor or passed on to other assembly operations, while the inspection results are monitored in the system. In any case, machine vision systems can provide much more information about an object than simple absence / presence sensors.
Machine vision is typically used, for example, to:
- Quality Assurance,
- robot / machine guidance,
- testing and calibration,
- real-time process management,
- Data Collection,
- machine monitoring,
- sorting / counting.
Many manufacturers use automated computer vision instead of inspection staff because it is more suitable for repetitive inspections. It is faster, more objective and works around the clock. Computer vision systems can inspect hundreds or thousands of parts per minute. It provides more consistent and reliable inspection results than humans.
Manufacturers can save money and increase their profitability by reducing defects, increasing revenue, facilitating compliance and tracking parts with computer vision.
Analogies to machine vision
The discrete photocell is one of the simplest sensors in the field of industrial automation. The reason we call it “discrete” or digital is that it has only two states: on or off.
The principle of a diffuse photocell is to emit a beam of light and detect if light has bounced off an object. If the subject is not present, no light can reflect into the photocell receiver. An electrical signal, usually 24 V, connect to the receiver. If an object is present, the signal turns on and fonction in the control system to perform an action. When the object deleted, the signal turns off again.
The diffusion photocell can also be analog. Instead of just two states, ie off and on, it can return a value indicating how much light returned to its receiver. In the case of the photocell shown, it can return 256 values, from 0 (meaning no light, dark or black) to 255 (meaning a lot of light or white). The photocell on the left returns 76 or dark gray. This is about 30% of the maximum value of 255. If a lighter object gets in front of the sensor, it returns a higher value. If the result is 217, which is about 85% of the full range 255, it means a much lighter shade of gray.
Imagine thousands of tiny analog photocells arranged in a square or rectangular array pointed at an object. This would create a black and white image of the object based on the reflectivity of the location the sensor is aiming at.
Individual scan points in these images referr to as “pixels”. Of course, thousands of tiny photoelectric sensors work to create the image. Instead, the ((scan lens focuses the image on the semiconductor matrix of the light detectors. Arrays of light-sensitive semiconductor devices such as CCD or CMOS work in this matrix. The individual sensors in this matrix are pixels.
