Kvantron Smart vision systems
The automation of product control is:
- Measurement of geometric parameters of objects (size, shape, thickness and other parameters)
- Rejects and unit accounting
- Sorting and automatic entry of product data into the database
- Labeling and more.
Vision systems are an excellent tool for improving product quality and record-keeping. This is achieved through feedback to actuators, increased productivity and data collection for analytics.
When it comes to vision systems, the most often presented is a 2D system, which is based on an industrial camera and special software.
2D-systems can be divided into 3 types:
With frontal illumination (reflected light mode). An image of the object itself is formed on the photosensor. The algorithms are based on video analytics methods.
Back-illuminated (shadow measurement method, transmitted light mode). An image of the shadow cast by the object is formed on the photosensor. In general, this method is significantly more accurate than the front-illuminated method, but is not always applicable due to limitations in the placement of the backlight. The algorithms are based on video analytics methods.
Based on neural networks with deep learning. They are used in tasks with complexly classifiable objects, when it is too difficult or impossible to use video analytics methods. For example, recognition of people, machines, or variations of complex objects (if the “pattern matching” method of video analytics is not applicable).
3D methods are even more varied than 2D methods. With 3D scanners, a measurement consisting of three coordinates (X, Y, Z) is formed, which is then processed in the controller.
A 3D vision system consists of a laser scanner, a controller (PC) and software. Laser vision systems are widely used to control geometric parameters of products, such as thickness, width, length, complex shapes. Feature of 3D-systems is to obtain three-dimensional point (X, Y, Z) directly from the scanner. If for 2D systems it is obligatory to have contrasting objects, characteristic elements and surface features in the frame (for example, edge, step, hole, pattern, change in reflectivity), then 3D sensors work with any surface. So why aren’t laser sensors used everywhere for metrology and geometric control of products? They are not much more expensive than solutions based on 2D vision systems. It is all about the specifics of work.
In 2D systems the task is to precisely select a contour and determine its characteristics. The task of 3D laser systems is to precisely identify the energy center of reflected probing (laser) radiation that forms Gaussian (normal) intensity distribution on photosensor. The problem is that any irregularity of the measured object (shape, steps, corners, edges, holes, change of reflectivity, etc.) distorts the intensity distribution on the photosensor, which leads to the wrong shift of the energy center. This generates errors in measurements.
In general, 2D systems are used for high-contrast image tasks, and for 3D systems for uniform image or height/thickness object measurement tasks.
Vision systems based on 3D are divided into several types:
The laser probe has a probing illumination in the form of a dot (spot). The most accurate method of measurement, but the speed is not more than 30 thousand measurements per second.
The laser profiler has a probing illumination in the form of a line (strip). It has medium accuracy and average speed.
3D scanner based on structured illumination. The method allows you to digitize an entire area that is in its field of view in a single measurement. The method is the most productive, but the most inaccurate.
Many other methods, such as time-of-flight (TOF), interference, low-coherence interferometry (white light), adaptive focal length control, and many others.